Candidate cloning

(Executive summary)

John Dendahl, chairman of the New Mexico Republican Party, quietly offered "more than $100,000" to the Green Party if they would run candidates in NM's first and second Congressional Districts. The Greens were relatively strong in New Mexico and had already been spoilers in local races. The difference was that the Republicans were now willing to pay cash for services that had previously been free.
— W.Poundstone: Gaming The Vote, Hill & Wang 2008, p.110. (This was in July 2002. Later, the Greens reported that the offer was $250,000. Dendahl told CNN that he got the money and instructions from "somebody in Washington" but refused to say who.)
In the June 2006 special election for the Congressman from California's 15th district, Republican Brian Bilbray was a foaming-at-the-mouth anti-immigration hawk who supported building a fence clear from the Pacific Ocean to the Gulf of Mexico... in one of his ads he was shown driving a bulldozer, keeping "Tijuana sewage" off California's beaches. William Griffith, a Republican running independent using $2000 of his own money, claimed to be even more anti-immigrant than Bilbray. How that was possible was unclear.
Then something odd happened. Phone calls started urging voters to vote for Griffith. Radio ads too. Griffith didn't know who was behind them. Both later turned out to have been funded by Democratic candidate Francine Busby. — W.Poundstone: Gaming The Vote, p.120, paraphrased.
Despite having $21 million in campaign cash, Pennsylvania's Republican uber-hawk Senator Rick Santorum had a tough race for re-election in 2006 against Democrat Bob Casey. A May 2006 SurveyUSA poll had found him the least popular US senator. (Santorum claimed in June 2006 to have seen secret documents showing the USA had finally located Saddam Hussein's "weapons of mass destruction." But neither Santorum, nor anybody else, ever revealed those locations.)
But the candidate having a really tough time was the Green Party's Carl Romanelli. He had almost no money and needed 67070 signatures to get on ballot. There were only 20000 Green party members in Pennsylvania. Santorum and Romanelli had diametrically-opposite stances on Iraq, abortion, and gun control.
Suddenly, Romanelli's funding increased by a factor of 2000. All of it came from (what the press identified as) GOP/conservative donors (some crudely disguised) and was funneled to a signature-collecting company.
— W.Poundstone: Gaming The Vote, p.124, paraphrased. (Poundstone estimates that Santorum, by donating money to Romanelli, effectively got about 15 times more votes-versus-Casey per dollar than if he'd spent it directly on his own campaign.)

Nicolaus Tideman emphasized the importance of making election methods "immune to candidate cloning" in his work.

In plurality voting, introducing an extra candidate who is enough like another, can paradoxically cause both to lose (because the vote gets split among the clones) – even if the voters really like the point of view they (both) represent! The very popularity of a view can cause its defeat!

Is that "democracy" or "a joke"?

This effect has even caused parties to intentionally aid opponents of their point of view ("helping a spoiler")! And it also has (more often) caused parties to intentionally hurt allies of their point of view!

Borda voting suffers from the opposite problem – "teaming" – a party could (if elections were conducted by Borda voting) win by simply sponsoring lots and lots of clones of their candidate to enter the race, which would artificially "magnify" his advantage. What a joke!

Range voting is immune to cloning – so such boorish election manipulation attempts just won't work under range voting. (Don't think this really matters? Wrong – check out the enormous distortion that occurred in this French Election.) So we can concentrate on who is the best candidate instead of who uses the most Machiavellian tactics.

Further reading

T. Nicolaus Tideman: Independence of Clones as a Criterion for Voting Rules, Social Choice & Welfare 4 (1987) 185-206.

Sam Howe Verhovek: Green Party Candidate finds he's a Republican Pawn, New York Times 8 Aug. 2001 page A10.

B.T. Zavist & T.N. Tideman: Complete independence of clones in the ranked pairs rule, Social Choice and Welfare 6 (1989) 167-173.


Return to main page