Why Alternative Parties Must Get Range Voting, or DIE

by Clay Shentrup and Warren D. Smith

Harsh Reality

It's quite a surreal and hostile world we minor parties live in. Confronted with the latest missteps by our elected officials, we shake our fists in the air, engage in passionate discourse to find solutions, and ultimately send our best and brightest into the political cauldron. But the result is that, when our candidates affect elections at all, it is almost invariably to harm the mainstream candidate most similar to ours. It seems we are gluttons for punishment, much like the 97,488 Floridians who voted for Nader in 2000, only to help secure the victory for Bush.

I recently read an article by a nationally syndicated columnist, in which he suggested that Libertarians (and ostensibly Greens et al.) close up shop as a political party altogether, and focus their efforts to form an interest group like the National Rifle Association or the League of Women Voters. In his words they are currently little more than a "high-school-level debating club". While his remarks drew the ire of Libertarians far and wide, any among us with an ear for reality knows he has a point. America's minor parties can talk all day about their great ideas, but our number one priority is to find a way to win a significant number of contentious elections. If we can't do that, then this columnist's point sticks. Hence the question: how do we break out of two-party duopoly? We must find an innovative way to do that, because nothing we've tried so far has worked.

We Need a Better Voting Method

"Insanity" is doing the same thing over and over again and expecting different results. – Albert Einstein

The Status Quo

If you were trying to design the worst way to vote, you might:

But wait, that's our voting system now! There are a lot of voting methods, and Plurality Voting (aka "first past the post") is one of the worst imaginable – not just for minor parties, but for everyone. It can do astoundingly stupid things, such as this:

#voters their vote
28 A>B>C>D
25 B>C>D>A
24 C>D>B>A
23 D>C>B>A

Here, A would lose to any opponent in a head-to-head election by a huge 72-to-28 margin, far larger than the hugest "landslide" in US presidential election history. But thanks to the incredibly poor Plurality Voting system, A is elected with 28% of the vote. Two of the results of this behavior are stifling two-party domination, and low voter satisfaction with the results.

The Perfect Foil

There are those who believe that hope lies in a system called Instant Runoff Voting. But IRV has been used for the better part of a century in countries such as Australia, Ireland, Malta, and Fiji, and has produced two-party domination, just like the US has experienced with Plurality Voting. Australianpolitics.com even says IRV "promotes a two-party system to the detriment of minor parties and independents." Ironically, that duopoly has been broken in at least 21 of the 27 countries that use traditional runoff elections. These have subtly – but evidently importantly – different political dynamics than instant runoff elections. But (suicidally?!) throngs of minor party and independent voters have worked to replace the runoff system with IRV in places like Oakland, CA, presumably because they did not understand that. Myths and misinformation have unfortunately been perpetuated by the popular press and (worse) by IRV-advocacy organizations such as the Center for Voting and Democracy. Two of the top myths are that IRV "eliminates spoilers" and that it only elects "majority winners".

Many IRV enthusiasts, such as Richard Anderson-Connolly, dismiss such problems and refuse to back superior voting methods, citing the apparent "momentum" of IRV. First of all, the momentum claim is false. Most US citizens don't even know what IRV is. Not a single US state has adopted it. Approximately two dozen U.S. cities (the largest being New York City, in 1936) have implemented it, only to have almost entirely reverted to Plurality. In 2006, four US municipalities approved IRV in ballot initiatives, so IRV adherents, unaware of IRV's history, have gotten the impression that IRV is some kind of new and great idea, that will also save minor parties from the certain doom that has been the fate of every minor party in America's history. Well, no. IRV in America is like an airplane prototype that keeps coming back down shortly after liftoff. And again, if we look at countries where it actually has taken off, it has produced two-party duopoly.

Yet some IRV enthusiasts, fully aware of this fact, still cling to IRV, in the desperate hope that if it could somehow eventually take root, it might one day allow us to transition to a proportional representation system called Single Transferable Vote (which would provide a small number of seats for minor parties). The idea of proportional representation is that multi-winner bodies, such as legislatures or city councils, should approximate the diversity of the electorate they represent; so if 10% of the voters are with Party-X, Congress should be about 10% Party-X, or at least significantly more than 0%. IRV is just STV applied to a single-winner election, like mayor, governor, or president; so the transition seems feasible. But America's history shows that IRV has indefinitely stalled, and there's no reason to believe that the duopolistic forces of our mainstream political parties would ever work to overcome enormous legal hurdles in order to implement a system like STV, that would only weaken their power. And even if they did, their biggest roadblock would be the Supreme Court. Finally, STV isn't even a particularly good method for proportional representation. So let any delusions about IRV die. It is, as Devin Ray Freeman has said, a "bullet in the foot" for minor parties. The same is true of other methods, such as Condorcet, Borda, and possibly even Approval Voting.

The Real Deal

The solution to this predicament is a system called Range Voting, advocated by the Center for Range Voting, an electoral reform advocacy group founded in 2005 by Princeton math Ph.D. Warren D. Smith and engineer Jan Kok. With this method, voters simply score every candidate on scale of, say 0-10 or 00-99. The winner is the candidate with the highest average. This system is so simple that many cannot help but feel skeptical that it could work for large-scale political elections. Won't it produce mediocre, greatest-common-denominator candidates? Won't it be especially susceptible to strategic voting, in which voters belie their sincere preferences in order to get a better result for themselves?  Won't it be too complicated?

To answer these kinds of questions objectively, Dr. Smith turned to both mathematical and real world experiments. In 2000 he wrote an open source computer program designed to calculate the utility efficiency (resulting voter satisfaction) of numerous voting systems. Hundreds of millions of simulated elections were performed, using 720 different combinations of five basic parameter ("knob") settings. For instance, electorates ranged from 100% honest to 100% strategic, and from 100% informed to 100% ignorant. Here are the results from two of those 720 combinations, averaged over millions of simulated elections, expressed as voter satisfaction indexes:

Utility measurements: Group A: 5 candidates, 20 voters, random utilities; Each entry averages the results from 4,000,000 simulated elections. Group B: 5 candidates, 50 voters, utilities based on 2 issues, each entry averages the results from 2,222,222 simulated elections.
Voting system VSI A VSI B
Magically elect optimum winner 100.00% 100.00%
Range (honest voters) 96.71% 94.66%
Borda (honest voters) 91.31% 89.97%
Approval (honest voters) 86.30% 83.53%
Condorcet-LR (honest voters) 85.19% 85.43%
Range & Approval (strategic exaggerating voters) 78.99% 77.01%
IRV (honest voters) 78.49% 76.32%
Plurality (honest voters) 67.63% 62.29%
Borda (strategic exaggerating voters) 53.26% 51.78%
Condorcet-LR (strategic exaggerating voters) 42.56% 41.31%
IRV (strategic exaggerating voters) 39.07% 39.21%
Plurality (strategic voters) 39.07% 39.21%
Elect random winner 0.00% 0.00%
From /vsi.html

As we can see, Range Voting dominates the other methods, especially when voters are strategic. So we can sell Range Voting to everyone, not just minor parties. Democrats and Republicans will derive hugely greater average satisfaction with Range Voting. And with Range Voting, spoilers are truly eliminated, not just decreased as with IRV. By entering an election, a minor party candidate either wins, or doesn't change the outcome. He never helps the opposing candidate win. This is good for both major and minor parties.

Most importantly for us, Range Voting experimentally produces substantially better results for minor parties. This is largely attributed to what has been called the "nursery effect", but regardless of why it happens, it is an empirical fact that it does. Consider the results of a pseudo-random poll of 122 US voters, conducted simultaneously with the 2004 US presidential elections. Real election results are in the middle, with experimental Range Voting averages on the right:

Candidate Plur RV (0-99)
Bush(Rep) 50.7 40
Kerry(Dem) 48.3 55
Nader 0.38 25
Badnarik 0.32 9
Cobb 0.10 5
Peroutka 0.12 6
Calero 0.003 4

An online Range Voting poll for the 2008 US presidential election, with over 1,500 participants as of January 13, 2007, had Libertarian Michael Badnarik beating several major party politicians: John Kerry, Dennis Kucinich, Joe Biden, Mark Warner, Condoleeza Rice, Bill Richardson, Nancy Pelosi, Mitt Romney, Evan Bayh, Tom Vilsack, Newt Gingrich, Tom Tancredo, Harry Reid, Chuck Hagel, Mike Huckabee, George W. Bush, Michael Bloomberg, George Allen, Sam Brownback, and Bill Frist – and this in spite of the fact that the poll is topped by liberals Barack Obama, Bill Clinton, Al Gore, and John Edwards. Greens also beat several major party candidates. By performing that well in a real election, a minor party candidate could get significant publicity, and be taken seriously by the press, and by voters. People would stop thinking of alternative parties as a waste of time, and start paying more attention to their message. With any luck, they might actually start to win elections! At that point they could work to enact the proportional representation version of Range Voting, called Reweighted Range Voting. Not only is it an easy transition from Range Voting, automatically defaulting to standard Range Voting when used for single-winner elections, but it is superior to STV in virtually every way.

Despite all this evidence, there are still some I've encountered who remain unconvinced. They feel that we've got to work to get fairer funding and participation in debates, and that if we keep up the good fight, eventually we'll make some headway, even with Plurality Voting. But as the entire history of democracy in our world vividly attests, the only factor that has ever correlated with a healthy multi-party system in single-winner elections, is having the right kind of voting method.

A Historical Perspective

History shows that US minor parties all seem to reach their all-time peak within about 1-16 years of founding, then diminish. Once they've fallen to about 5% of what they were at their peak, they die. Their founders perhaps suffer from the delusion that their idea, that new socio-politico-economo-religious-ecological idea that is special to their party, is so wonderful or different that, maybe, just this once, that party will be able to overcome Duverger's law and win, or at least grow into a viable party.

And so, they foolishly choose once again to concentrate on their wonderful idea and to ignore (or nearly so) the idea of improving the voting system. They don't understand that it just doesn't matter what their idea is. No matter what it is, if one of the major parties gives some slight lip service to that idea, say equivalent to just 2% of what the minor party wants, then even Plurality Voters who like that minor party best, are fully strategically justified mathematically/statistically in voting major-party, and in fact do so – in around a 10:1 ratio. Nothing can overcome that ratio. And then Duverger's law wins and they die like usual. Here are the stats.

Libertarian Party
year canddt vote/total-votes =%
1972 J.Hospers 3,674 / 77,744,027 0.005
1976 R.MacBride 172,553 / 81,531,584 0.212
1980 E.Clark 921,128 / 86,509,678 1.065
1984 D.Bergland 228,111 / 92,653,233 0.246
1988 R.E.Paul 431,750 / 91,594,686 0.471
1992 A.V.Marrou 290,087 / 104,423,923 0.278
1996 H.Browne 485,000 / 96,275,401 0.504
2000 Browne 386,000 / 105,417,258 0.366
2004 M.Badnarik 397,265 / 122,293,332 0.325
2008 ?    

The Libertarians still seem alive and kicking, but judging from history, presumably also are on a downward trajectory, and even at their peak in 1980 never came close to four of the parties in the graph above. Today's minor parties are dwarfs even compared to the minor parties of the 1900s. Can Libertarians afford to sit around twiddling their thumbs, focusing on things other than trying to change the voting system? Probably not. Add many more years to this Libertarian trajectory and extrapolate for yourself. If the Libertarians drown, it will be because they waited too long to make voting reform their top priority.

Green Party
year canddt vote/total-votes =%
1996 R.Nader 685,297 / 96,275,401 0.71
2000 Nader 2,883,105 / 105,417,258 2.74
2004 D.Cobb 119,859 / 122,293,332 0.10

In 2004, when the Greens (judging by their election result) appeared about to die, Cobb and Badnarik both noted in their debates that they advocated IRV. Badnarik and his Libertarian 2004 running mate Campagna now both have endorsed Range Voting.

Conclusion

Throughout US history, alternative parties have inevitably died within a relatively short time frame, with 100% consistency. At the most, they have simply stopped running candidates, and evolved into political interest groups. History tells us that if we do not do something revolutionary, we are destined to meet the same fate. So take all of those those high-minded ideals you've got, and set them aside for a moment. Your single most urgent task is to pull out all the stops, and pool your resources together with those of other voting reform advocates, in order to get Range Voting. And I mean today. Join the Center for Range Voting. Participate in their online discussion group. Donate money or time. Endorse them. Help improve their web site. And most importantly, spread the word! Tell friends, elected representatives, and members of your political party to support Range Voting. It's life or death!

Return to main page