Steven Hill responds to Lee Drutman: Fusion yes, but more study of modern-day uses needed

Drutman’s article is a thoughtful yet ultimately unconvincing argument that Fusion Voting is better than Ranked Choice Voting

[Editor’s note: this article is Steven Hill’s response to political scientist Lee Drutman’s recent article on DemocracySOS, "Yes, Fusion does offer a new horizon for US Politics" (which was Lee’s response to Hill’s previous critique of fusion voting). This kind of thoughtful debate about which political reforms are most salient for the current moment, as well as the strategy and tactics for enacting reform, is a core part of our mission at DemocracySOS. We hope you find this back-and-forth discussion informative and interesting.]

The one thing that most political reformers agree on is that the current political system in the United States is broken. And for lots of reasons. The antiquated “winner take all” electoral system, privately financed campaigns, lack of free media time for candidates and parties, no direct vote for president, anti-majoritarian Senate, no national elections commission to establish minimum election administration standards, no universal voter registration, and nonsensical voter laws that are manipulated to keep certain classes of voters from participating -- the list of our democracy deficits is long.

We can be cheered by the fact that most established democracies in the world have already settled these problems, but they have taken a very different tack than the antiquated, still 18^th-century based dis-United States of America.

So then which direction(s) should we in the US pursue when it comes to political reform? Which pathways show the most promise? If we don’t figure out how to unite and collaborate, and even support each other within our differences, our separate efforts will not amount to much. The mountains we have to climb are high and steep.

On the other hand – it’s a big country. A one-size-fits-all approach to reform probably won’t be successful. What works in California may not work in Mississippi or New York or Wisconsin or Georgia. The US has so many elections, at local, state and federal levels, so we may need different reforms for those different levels (a point Lee alludes to in his article).

It is in this spirit of collaboration amidst disagreement that I offer my thoughts to Lee Drutman, as well as my other friends and allies who are supporters of fusion. Here’s my basic take: a lot of the positives that Lee claims for fusion are not supportable by enough examples or data to form hard conclusions. And a number of his conclusions regarding ranked choice voting are contradicted by a fairly extensive track record of RCV elections in the US and around the world going back many decades.

Lee offers a very thoughtful and spirited analysis, and he and I have basic agreement that the antiquated, 18th century “winner take all” system results in a number of troubling toxicities, and that the ultimate solution to the failures of “winner take all” is to enact proportional representation. We also both agree that fusion is important and has resulted in some notable achievements, especially in New York, where it has seen the most active use in recent years.

But where we part ways is in our mutual understandings of how important fusion is in the pantheon of reforms, and in the bigger picture focused on reaching a genuine system of multiparty democracy founded on the bedrock of proportional representation. Lee portrays RCV as more or less a dead end toward the goal of PR and multi-party democracy, and portrays – but fails to prove – fusion as the true pathway forward. Indeed, Lee make some pretty bold claims considering how few places actually use fusion – it’s banned in 43 states and only used with regularity in one state, New York – and the paucity of available examples that would allow studying the impacts.

Drutman’s three arguments for fusion

Lee advances three main arguments to support his view that fusion is a vibrant political reform and in fact is far more important than the most successful political reform of our times, ranked choice voting. Lee’s first argument is that RCV is allegedly “candidate-centered” while fusion focuses on parties, and fusion will result in more political parties than RCV, and since parties play a crucial role of “shaping and organizing politics,” anything that appears to undermine them, such as “candidate-centered” RCV, is problematic.

His second argument, related to his first, is his contention that, because fusion voting is more of a party-based reform, and RCV is allegedly a “candidate-centered” reform, fusion has a better chance than RCV of being the catalyst for the US transitioning to proportional representation. Lee writes, “the core question among reformers appears to be whether the pathway to proportional representation goes through parties or candidates. Looking at history, it’s clear to me that the pathway can only go through parties.” Lee also ends his essay by saying that RCV “is not a pathway to more parties or proportional representation.”

And thirdly, Lee finds fusion voting compelling because he believes that, even if it doesn’t result in proportional representation, it offers a unique opportunity to rebuild a vanishing political center, and he appears to believe that fusion will result in more centrist parties, and even further he says it offers a better avenue than RCV to neutralize the rise of the MAGA authoritarian right.

There is little present-day or historical evidence, and Lee does not present a strong factual basis, for any of his claims. I will now address his three arguments, starting with the last one first.

Fusion is better than RCV for promoting centrism?

Lee writes, “I find fusion voting compelling because it offers a unique opportunity to rebuild a vanishing political center;” he also writes, “it can create an instant home for anti-MAGA Republicans to support Democratic candidates without supporting the Democratic Party” and “fusion is more practical and tactical, it counters the threat of the far right.”

This is interesting in theory, but we haven’t seen that happen actually in practice, either today or historically. Lee does not ground his argument in credible examples, it’s pretty much all conjecture to the point of sounding like wishful thinking. In fact, Lee acknowledges that “in recent history, fusion parties have primarily existed on the edges of the political spectrum, not the center.” 

Fusion has been around for about 150 years, and mostly has helped parties on the left/progressive side to influence the center-left party – such as William Jennings Bryan and the Populist Party fusing with the Democrats in the late 19th century – or parties on the right to influence the center-right party, like the Conservative Party in New York, whose votes for Republican candidates have been crucial to some victories for governor and U.S. Senator.  Historically, there have been few “centrist” minor parties that were able to facilitate voters from one of the major parties actually supporting candidates from the other major party.

Lee does cite “the role that fusion parties played during Abolition and then during the great farmer-worker upsurge of the post Civil War era,” but those were efforts on the center-left and not examples of centrist minor parties spurring crossover appeal among candidates from the two major parties. Also, at this point those examples are over a century in the rear view mirror. Intellectually interesting, but not very persuasive.

Richard Winger from Ballot Access News, who is a walking encyclopedia of the history of minor parties, emailed me with the observation that, “centrist parties are new in U.S. history. There were none, as far as I know, until the John B. Anderson candidacy of 1980, and that wasn't a formal party, although technically it was a party in a few states.  Also after the 1980 election Anderson tried to create a new centrist party but it didn't ever get going.” That included in states that had fusion at the time. The Independent Party of Oregon has tried to act as a centrist party, at times endorsing GOP and Democratic candidates in different races utilizing Oregon’s aggregated variety of fusion, but it has run few of its own candidates and it mostly has focused on a handful of issues such as campaign finance reform. It is more of an influence group than a political party.

As his lone example for his meager hope, Lee cites the brand new Moderate Party of New Jersey – which doesn’t really yet exist. It’s more of a party in formation, since it does not have formal ballot status in New Jersey and in any case does not plan to run any of its own candidates. Instead, it hopes to nudge both Democratic and Republican Parties toward the center, rather than replace or compete with them. And it hopes to use fusion to do that, except one problem – fusion is illegal in New Jersey, as it is in 42 other states.

So the Moderate Party hopes to change that by challenging New Jersey’s ban in state court, but there’s an additional problem – the US Supreme Court declared in a previous 1997 decision, Timmons v. Twin Cities Area New Party, that the banning of fusion is legal because states have a legitimate interest in fostering “a stable two-party system.” Chief Justice William Rehnquist deployed a ridiculous argument that justified discrimination against the rights of political parties, yet it presents a serious obstacle to the spread of fusion to other states. Considering all of these headwinds, even the Moderate Party’s founders acknowledge that its chances of success are low.

Given the lack of historical or even present-day examples, it is puzzling to me that Lee would place so much faith in this as a legitimate pathway toward centrism or neutralizing the authoritarian MAGA right. In fact, if that is the goal, RCV has a far better track record than fusion. Look at the GOP gubernatorial primary in Virginia, which used RCV and resulted in the moderate Republican Glenn Youngkin defeating the “Trump-in-heels”  candidate and going on to win the governor’s race in a Democratic-leaning state. In Utah, a heavily GOP state, two dozen towns and small cities have used RCV for local elections, resulting in a range of different types of Republicans and Democrats getting elected, and earning praise from Utah leaders spanning the political spectrum.

Perhaps the best known example was the RCV election of Democrat Mary Peltola in Alaska to its lone congressional seat, keeping out right winger Sarah Palin. Alaska has a national reputation as a very Republican state, but in actual fact 60 percent of registered voters are independents, having chosen "Non-Partisan," "Undeclared" or Alaskan Independence Party as their political affiliation. Certainly those independents are conservative small-government supporters, yet they ranked their ballots for the moderate Democrat Peltola over a former reality TV celeb like Palin who resigned halfway through her term as governor to gallivant in the Lower 48. RCV also helped moderate Republican Senator Lisa Murkowski overcome a MAGA far-right challenge; Murkowski likely would have lost the GOP nomination under the old closed primary system.

Additionally interesting, RCV also has been credited by a number of Alaskan leaders and the Anchorage media in contributing to a cross-partisan environment in which moderate Republicans and moderate Democrats have actually formed a coalition government together (see my previous DemocracySOS article about this).

And Lee was quoted in the Washington Post citing Australian scholars who point out that “ranked-choice voting has exerted a moderating force on Australian politics.”

So it’s pretty clear that if the criteria is to promote centrism and keep out the authoritarian MAGA right, RCV has a much better track record then fusion. Because at least RCV has a track record of promoting centrism, whereas fusion has none.

Does fusion promote parties & multi-party democracy better than RCV?

Now I will turn to Lee’s other two arguments, taking them together because they are related: that fusion promotes political parties more than the allegedly “candidate-centered” RCV, and for that reason it is a better vehicle for enacting proportional representation. On both counts, Lee has presented a false choice based on an apparent misreading of the track record of ranked choice voting and other electoral systems, both in the US and in other democracies.

Let’s examine first the claim that RCV is candidate-centered and that it allegedly focuses on candidates, while fusion focuses on parties.  

Empirically, that’s a false statement. Look at RCV elections in Ireland, Australia, Malta, and in the US states of Alaska and Maine (and perhaps soon Nevada) -- all of those places use partisan elections with vibrant political parties, even as voters pick candidates instead of a political party. Certainly these places do not use continental European-style party list systems, in which voters select a party instead of individual candidates, but that hardly means the elections in these places are not focused on political parties. Any suggestion to the contrary badly misreads the politics of these countries and US states.

Even a cursory glance at the headlines that surrounded the controversial Alaskan elections in which Democrat Mary Peltola beat Republican Sarah Palin twice within a few months reveals that these elections were intensely focused on parties, even as voters ranked individual candidates. The Irish and Australians would be quite surprised to hear that their system is “candidate-centered” to the detriment of political parties, given the robustness of their multi-party competition, the stickiness of partisan voters’ preferences (in which families have voted for the same long-standing parties for decades, though that has declined somewhat in recent years) and the complexity of multiparty coalition negotiations that have been necessary to form governments.

Even more puzzling, Lee favors an open list PR system, writing, “I also prefer an open-list system of proportional representation to a candidate-based system of ranking, the single-transferable vote.” But as Canadian political scientist Henry Milner has pointed out on DemocracySOS, depending on what kind of open list system you select, those also are candidate-centered, some more than others. In many open list systems, voters have the option of voting for either candidates or a political party. And in Finland’s “fully open” PR list, the voter can only vote for individual candidates, not political parties. Some of these open list systems behave very much like party-based proportional RCV elections, but without the advantage of transferable ballots, which helps voters to not waste their votes on small unelectable parties or longshot candidates.

In fact, in one recent op-ed that Lee co-authored he discussed his support for an open list system, writing “Voters just pick and rank candidates from the party lists.” So somewhat inexplicably Lee has not explained why his preferred “open list system,” in which voters are voting for candidates, is acceptable to his political sensibility but ranked choice voting, which also allows voters to rank candidates, somehow falls short.

Lee is a well-informed political scientist, and these points are perfectly obvious. So I am puzzled why he does not account for them in his analysis. Here’s my conjecture on what’s going on: his perception about RCV in the US fails to take into account that the RCV movement gained a toehold, and then a foothold, in the political landscape by reformers winning the first RCV victories in nonpartisan elections in cities. I ran the first successful RCV campaigns in San Francisco, Oakland and other cities, and at the time these early victories were the first electoral system reforms decided at the ballot box in many decades. The reasons why we targeted nonpartisan city elections was purely a pragmatic decision based on: 1) limited financial resources for reform, 2) Americans’ general lack of familiarity at that time with alternative voting methods, and 3) narrow reform opportunities based on very local conditions.

Those early successes kicked off the modern-day movement for electoral system reform, and for many years winning victories for RCV in nonpartisan cities was the bread and butter of the movement. It wasn’t until 2016 – 14 years after the first nonpartisan victory in San Francisco – that the growing movement for RCV had gained enough momentum and resources to win at the ballot box in Maine, and this time to be used in partisan elections.

Given this history, to some observers RCV in the US appears to be exclusively nonpartisan and “candidate-centered.” But that has never been the intention, and it does not need to be the future of RCV in the US. Indeed with the passage of RCV in Maine and Alaska, and with Nevada set to vote on it and potential campaigns coming in November 2024 in Arizona and Oregon, that points the way to more party-based elections in RCV’s future.

In some of these states, RCV is combined with a top-four open primary, which Lee and others believe is anti-party. I understand that viewpoint, but I think in truth the bigger picture is more complicated. I live in California which has a top two open primary, yet it’s pretty hard to argue that California does not have strong political parties, including not only Democrats and Republicans but four minor ballot-qualified parties. Certainly minor parties are damaged by a top-two primary because they rarely finish in the top two and make it to the November election. But regardless of the primary situation, most legislative districts are one-party, “winner take all” fiefdoms, and in those districts the Democratic or Republican parties are strong enough to consistently prevail.

Does fusion create more parties? No evidence.

Lee also says that “The value of fusion is that it creates more parties” and it creates a pathway to “move beyond the winner-take-all system” and to move “beyond the two-party system.” But there’s really no credible evidence in recent years to support that view.

Granted, fusion is practiced in only seven states (and Lee’s preferred “disaggregated fusion” is practiced in only two states). But even in New York, which is considered the apotheosis of this reform, there is no evidence that fusion has created an impressive number of more parties or moved beyond the two-party system. In fact, there are fewer minor parties in New York today than there were in 2002. And even though California does not have fusion, it has more ballot-qualified parties (six) than New York (four).

I do agree with Lee in his portrayal of the Working Families Party as cleverly exploiting fusion to exert influence in NY politics, and it is the best example of fusion showing promise. Lee writes, “Over the last 25 years the year-in, year-out work of interviewing candidates, running issue campaigns, training volunteers and turning out its supporters has made the WFP a serious actor in state politics. Politicians court the party’s support, and the WFP is known in the state for extracting serious policy concessions.” Lee lists a number of issues in which the WFP has been influential, including minimum wage hikes, paid family leave, public financing of elections, an end to the Rockefeller Drug Laws, higher income taxes on the wealthy, rent regulation, and climate regulation.

I celebrate this as a valuable feature of fusion in New York, even as I recognize that the success story has not been replicated anywhere else in recent years. And that dynamic is not all that different from other lobbying and influence groups, such as environmental groups, or a tenants union, or a voting rights group. Those groups also are seriously regarded in state politics and courted by politicians.

And Lee’s citing of the need of WFP to often align with top Democratic candidates, like former Governor Andrew Cuomo, as a tactic to maintain its ballot status and avoid splitting the liberal-left vote that could help elect Republicans, amounts to an admission that there are severe institutional constraints that fusion voting does little to counteract. Does fusion truly create a real multiparty democracy? Or mostly maintain the winner-take-all status quo, albeit with occasional sparks from the political margins? I think the evidence is quite clear that fusion results in the latter, not the former.

Fusion leaves winner-take-all intact

Due to these institutional constraints, it seems hard to imagine that fusion, even if it spread to a lot more states, would ever result in multiparty democracy. How could it? It leaves the “winner take all” electoral system 99.9 percent intact. I suspect Lee would agree with my observation, since he writes that, “within the system of single-winner elections,” the WFP as an influence group “is as close as one gets” to proportional representation.

Which is to say, you don’t get very close at all. It is simply not credible to argue that the presence of two minor parties in New York have “altered” that state’s “winner take all” system into a true multi-party democracy anymore than California’s four minor parties have transformed the Golden State into a real multi-party democracy.

Lee also makes several other observations that I have a nitpicky quarrel with, here are two of them that stand out.

He does acknowledge that RCV has done well in primary and local nonpartisan elections, but he seems to think that’s true only in “small electorates.” He doesn’t really define what “small” is but is the RCV electorate of New York City, which is larger than 38 states, or the congressional districts in Maine and Alaska, with around 700,000 residents, or a mayor’s race in San Francisco with 250,000 voters, considered small to him?

Also he says, “Most reform campaigns fail,” apparently unaware that RCV is on an unprecedented roll. It has won its last 21 ballot measures, usually by enormous landslide margins, and is now used in over 60 cities and a number of states, including New York, Maine, Oregon, California, Illinois and others for local, state or federal elections. It is the most successful political reform of our times. Get fusion into half that many cities or states, generate some real-world data for analysis, and then this discussion will be grounded in something more meaningful and informed. Without more examples and data, I can see no factual basis or credible evidence for Lee’s viewpoint that fusion voting will lead to PR, or multiparty democracy, or more parties, or more centrism, or that it is better than ranked choice voting in those ways.

But it doesn’t have to be fusion over RCV or vice versa. As I stated at the outset of my article, it’s a big country and there is room for different approaches to political reform. Or for innovative hybrids. I believe fusion and RCV could coexist and even work well together. In fact, I would argue that fusion is stronger with RCV since WFP and other minor parties would be more liberated to run their own candidates without the charge of “spoiler.” Yet they would preserve the option of playing that cozy insider game by endorsing (or threatening to withhold an endorsement from) Democratic candidates. Lee writes that he is working on a longer paper on the importance of party-centered reform, and I hope his more extensive work provides more explication and evidence of his basic arguments.

Lee, yours has been a powerful voice making the case for proportional representation, and you have added a new and inspiring dimension to the longtime advocacy of people like myself and Rob Richie from FairVote, as well as many others. Collaboration among fellow political reformers can work well as long as the differing advocates remain respectful and to some degree supportive. I do wish you, the Working Families Party and other fusion advocates the best of luck with your important work.

Steven Hill @StevenHill1776