Approval Voting - could it work?

I recently conducted an e-mail interview with one of the founders of Citizens for Approval Voting and Americans for Approval Voting.  I actually had to do some real research before I conducted the interview, but I think you'll find the results interesting and enlightening.

Quickly tell us a little bit about yourself and your organization.





My name is Rob LeGrand. I ran for Texas State Representative (District 50) in 2002 as a candidate of the Libertarian Party and am currently a Ph.D. student in computer science at Washington University in St. Louis, Missouri.





I founded Citizens for Approval Voting and Americans for Approval Voting with Robert "Rock" Howard and Dr. Steven Brams in late 2002.  CAV is an educational and outreach organization dedicated to introducing voters everywhere to Approval Voting; AAV is a political action organization dedicated to the passage of legislation that mandates Approval Voting for single-winner public elections.





Can you explain Approval Voting to my readers in one short paragraph?





Approval Voting is a single-winner election system. The current system, plurality voting, presents a ballot listing the candidates for a given race and allows only one vote for one candidate. Approval Voting uses the same ballot format but allows a voter to approve any number of candidates, effectively allowing a vote for or against each candidate. Approval Voting is a simple reform that gets rid of the spoiler and lesser-of-two-evils problems of plurality and would result in a level playing field for all parties and candidates.





Has anyone shown any interest in using the Approval Voting system  for any municipal or local elections?





Yes, my hometown of San Angelo, Texas, showed interest in using it within single-member districts to elect the city council, but the plan was not adopted. Using city-wide cumulative voting, a kind of proportional representation system, was also considered and ultimately rejected. I believe Rock Howard has made efforts to promote Approval Voting for elections in Austin, Texas.





Who first came up with the idea for Approval Voting?





Approval Voting was invented and investigated independently by several people in the 1970s, including Robert Weber and Steven Brams. It was first named by Weber and most thoroughly researched by Brams.  Brams wrote a book called Approval Voting with Peter Fishburn in 1983.





Let's talk about Presidential elections.  How would Approval Voting mesh with the Electoral College?





In any number of ways. Ideally, each state would choose its electors using Approval Voting (whether by popular vote or legislative vote), then the electors would convene and use Approval Voting to elect the President and Vice President. There is nothing in the Constitution of the U.S.A. (see Article II and Amendment XII) that would preclude such use (but "majority" would mean majority of voters, not majority of votes). Note that electors' use of Approval Voting would effectively create a candidate- withdrawal option. For example, electors pledged to vote for Ralph Nader could choose to approve the Democratic candidate as well if Nader's chance of winning were minimal, and they could do so without breaking their pledge to vote for Nader.





I am a Centrist. I understand there was an experiment with Approval Voting during the 2002 French elections. Explain how Approval Voting actually ended up benefiting the Centrist candidates.





A 2003 paper by Laslier and Vander Straeten describes an experiment in which some French voters submitted approval ballots in addition to their official plurality ballots. The results convincingly demonstrate that the vote totals of relatively centrist candidates like Bayrou were increased much more than those of relatively extremist candidates like Le Pen. This kind of result can be expected when Approval Voting is deployed for any single-winner election with many candidates, as voters will be free to approve all candidates they find preferable in addition to a lesser-of-two-evils candidate they find necessary.





Wouldn't certain laws have to be changed to allow Approval Voting in general elections? I mean, isn't there a law that says we can only cast one vote for each office?





I know of no such law, although it's possible that some states and localities have something similar codified. The greater challenge will be to overcome many Americans' intuitive one-man-one-vote objection, which can be answered by pointing out that each voter has exactly the same power as others under Approval Voting-- approving more than one candidate does not necessarily have a larger impact on the result than approving one.





What is the difference between Approval Voting and Cardinal Ratings?





Cardinal Ratings, also known as Range Voting, allows a voter to rate each candidate with a number in some specified range such as 0 to 10. The winning candidate is the one with the highest total rating. Approval Voting is equivalent to Cardinal Ratings where the only allowed ratings are 0 and 1. It has been shown that Cardinal Ratings is equivalent to Approval Voting when voters are rationally strategic--that is, it is always to a voter's advantage to vote each candidate at one of the extremes, so an electorate of rational voters would act as if they were voting in an Approval election. The primary advantage of Approval Voting over Cardinal Ratings is the simpler ballot and relative ease of voting equipment conversion. Warren Smith has recently founded a Center for Range Voting; see .





What is the difference between IRV (Instant Runoff Voting) and Approval Voting?





Both Approval Voting and IRV generalize plurality, but in different ways. While Approval allows voting for more than one candidate,  IRV allows voting a ranked ballot. The counting takes place in rounds. In each round, the first-place candidate on each ballot gets a vote; the candidate with the lowest first-place-vote total is eliminated from all ballots. The rounds continue until one candidate remains.  Essentially, IRV votes in your place in repeated plurality elections, voting for the highest un-eliminated candidate on your ballot. Unfortunately, this can result in compromise candidates being eliminated in an early round; see an example at . IRV often presents opportunities for strategically insincere voting, usually by voting a compromise candidate over a favorite.





In contrast, under Approval Voting, it is never to a voter's advantage not to approve a favorite candidate.





This question may be a little tough, but . . . what are some of  the arguments against Approval Voting and how do you answer those  arguments?





gives three objections to Approval, none of which survives close inspection:





"Approval voting does not solve the spoiler problem."

"Approval voting forces voters to cast equally weighted votes for candidates they approve of."





This is true, but the decision regarding how many candidates to approve is left to the voter. On the other hand, IRV forces strict ranking and, like almost all other ranked-ballot systems, sometimes rewards insincerity near the top of the ballot, effectively "forcing" a smart voter to vote preferences opposite to sincere preferences (say, voting A over B when you prefer B to A). See for a realistic example of IRV's failure to encourage sincerity. Approval Voting never rewards such insincerity. 

"Approval voting would challenge our notions of majority rule."





A notion of majority rule that prefers the victory of a candidate with 51% of the vote over a candidate with 75% approval *should* be challenged. In practice, however, a candidate that is the favorite of 51% of the electorate will almost certainly win anyway under Approval Voting when voters follow approval polls leading up to the election--that 51% would likely approve only that favorite. 





This question is a bit hypothetical, but if Approval Voting had been used in the 2000 Presidential election, do you think the outcome would have been the same?





Almost certainly not. First, using Approval Voting in the general election would largely remove a political party's incentive to conduct primary elections, since running two similar candidates won't split the vote as it would in plurality elections. So candidates with across-the-board appeal such as John McCain could have run for President alongside Bush and Gore. In fact, it's unlikely that the two front running candidates would have been Bush and Gore; McCain would likely have attracted enough approval to emerge as a favorite.





Also, the results from battleground states might have been quite different. In Florida, for example, Ralph Nader's candidacy obviously cost Gore the state. If Nader supporters had been allowed to approve their favorite of Bush and Gore in addition to Nader, Gore would almost certainly have won. Nader's true support would have been measured more accurately as well, since there were likely many Gore voters who actually preferred Nader but voted strategically.





In doing research for this interview, I came across almost universal praise for Approval Voting. My question then is why aren’t we seeing more movement in government to make Approval voting the standard?





The simple (and somewhat cynical) answer is that the people in power got there under the current system. More specifically, since plurality voting encourages a two-party system, the current two powerful parties have no incentive to open up the system to currently minor parties. Our two-party system will continue to be stable until public elections use a voting system that gets rid of the vote-splitting effect, like Approval Voting, or one of the major parties stumbles so badly that it is replaced. The latter hasn't happened since the Whig Party was replaced by the Republican Party in 1856 (which was composed mostly of former Whigs, just as the original Whig Party was made up of former Federalists).





Another answer is that most people distrust that which they don't understand. Most politicians don't want to advocate using a new voting system for fear of being seen as a nutcase (or as a traitor to political friends). The answer, of course, is education of the public of the advantages of Approval Voting.





You claim that elections will be “cleaner”, that Approval Voting will cause negative campaigns to backfire. But won’t it actually lead to more negative campaigning? Won’t politicians start  running “vote for me” as well as “don’t vote for him” ads?





The point is that “vote for me” ads will be much more effective in general than “don’t vote for him” ads under Approval Voting. When there are only two candidates with a chance to win, as under plurality, attack ads are about as effective as self-promotion ads, since there's only one opponent to attack. But when competition is opened up to more candidates, attack ads are less effective and more likely to backfire. Positive campaigns will likely be much more effective than negative campaigns.





Is there anything you’d like to add before we close?





I'd just like to request that readers not take my word for it--or anyone else's. Here are some sites that will give you the information to make up your own mind:





http://www.approvalvoting.org/


http://www.approvalvoting.org/articles.html


http://www.electionmethods.org/


http://bcn.boulder.co.us/government/approvalvote/center.html


http://www.accuratedemocracy.com/


http://www.sciencemag.org/cgi/content/summary/292/5521/1449


http://www.sciencemag.org/cgi/content/full/294/5541/303c





Also see the article "The Fairest Vote of All" in the March 2004 issue of Scientific American.





Thank you again for agreeing to answer these questions, and know that if any of my readers have any follow up questions I'll be getting back in touch.





It's a pleasure. Feel free to include questions and answers from CAV's FAQ at as well.

So readers, what are your views on Approval voting? Do you think it's better than the current system? If it is indeed a better system, how can we make it happen? It's an interesting subject and I would appreciate your feedback.  Drop me an e-mail at Here or comment on this article on my blog, A Little Left of Centrist.