Thursday, May 13, 2010

Rooting For Clegg and the Lib Dems, Part II

Even though the flawed plurality voting system resulted in his Liberal Democrats actually losing seats in the recent UK elections, Nick Clegg today sits as Deputy Prime Minister, Liberal Democrats now occupy a handful of cabinet seats in a coalition government led by Tory (conservative) David Cameron, AND the citizens will have an opportunity to vote in a referendum on an alternative voting system. The Tory/Lib Dem policy agreements can be found here.

The lesson for American third parties? Actually, there are a few lessons:

1. Televised Debate Participation: Third party participation in nationally televised debates greatly changes the campaign dynamic. Had Nick Clegg not been allowed to participate in the debates, it's doubtful that he would have had the credibility necessary to be part of coalition government--especially given the fact that his party actually lost seats in the election.

That third party participation in debates changes the campaign dynamic is not news. Indeed, Ross Perot's showing in 1992 and 1996 led to the Republicrats creating debate participation criteria that effectively rule out anyone but them from participation. It will be difficult for American third parties to make progress at the national level without such participation.

2. Coalition Politics: The Liberal Democrats were formed in the late 1980s when the Liberal Party and Social Democratic Party merged. The Social Democrats included former Labour Party politicians. The merged party had enough credibility to win seats in the Parliament.

I know it's difficult to imagine here in the US, but I can envision a scenario in which a variety of third parties coalesce , recruit former Democrats and Republicans, and win some seats in the Congress. There actually is some precedent for that here in Wisconsin, where the Progressive Party controlled state government for a brief period in the 1930s and won a few US House seats.

3. Voting Reform: Clegg's maneuvers resulted in the conservatives agreeing to have a national referendum on voting reform. It's not clear yet what kind of reform proposal will be voted on, but it will probably be along the lines of a system that will ensure that a seat cannot be won with less than fifty-percent of the vote.

In essence, Clegg has succeeded in putting the ball in the court of the UK citizens. If they want to see fairer, more representative elections, they will have to support the referendum. Expect Labour and the Tories to fight like hell to defeat whatever proposal comes forward.

In the US, we're seeing a growing number of establishment politicians running as Independents. Most want to follow the Lieberman Model in Connecticut: take advantage of one's name recognition to score a narrow victory in the rotted plurality voting system.

All the Democratic Party outrage at Lieberman doesn't seem to translate into any action on their part to change the voting system. Consequently (and as is typically the case in American politics), it's going to be up to grassroots activists to do the heavy lifting.

1 comment:

Steve Barney said...

Whatever voting system is put to a referendum, the system should be transparant enough to allow the public to easily see how many ballots of each voter-type were cast. By "voter-type" I mean a voter who cast a ballot with a given ranking of the candidates, such as A>B>C, or B>C>A.

I am not aware of any country with a preference ballot system, such as Australia's instant runoff system, that is that transparent. If the UK does that, analysts and mathematicians may be able to tell us some very interesting and suprising things about the results of the evident will of the people, as expressed by those ballots.

In some (probably rare) cases, they may be able to show that the same candidates would have won under any reasonable (non-absurd) ranked preference voting rules, including plurality; for example, in a 3 candidate contest, if the winner gets more than 2/3 of the vote (1st ranking), then we can be sure that the winner was elected by the will of the people.

Notice that 2/3 is a supermajority. In general terms, the required supermajority formula is (N-1)/N, where N=#candidates [*]. If there are more than 2 candidates, a simple majority is not necessarily enough! To be sure that the preferred candidate won, we would need more information than that to be able to determine the strength of a simple majority winner's mandate; that is, we would need to know the number of each voter-type. Given that information, in many (probably most) cases, analysts may be able to show (mathematically prove) that they would have gotten different results under some different rules.

All of the above assumes that the voters voted sincerely, but that's another whole can of worms.

*: This, and much more, was mathematically proven by Donald Saari, with whom I am the co-author of a voting math paper titled "Consequences of Reversing Preferences," in the journal _Mathematical Intelligencer_, Fall 2003. For beginners, I recommend "A Fourth Grade Experience", by Donald Saari [math.uci.edu/~dsaari/fourthgrade.pdf].