Reply #48: wrong - Proportional Representation CAN Work in the US

[liberalpragmatist]

First, let me make a point about PR in general. No system guarantees representation to disadvantaged groups and minorities than proportional representation. Most legal experts and political scientists have recommended PR for Iraq (see: http://www.fairvote.org/articles/washingtonpost070604.htm).

Also, your point about coalition governments and PR being bad is a gross generality. It is true that when a *pure* form of PR is used, instability and extremism can result - witness Israel, Italy, and, of course, Weimar Germany. However, the various systems collectively known as "proportional representation" have been highly tweaked and today, most estabilished democracies in the world use it. Most nascent democracies, like South Africa, explicitly chose it because they felt it would be the only way to reflect the diversity of the country. New Zealand in '93 adopted PR with good results. Germany and the Netherlands, in particular, have had EXCELLENT results with PR over the last 50 years. Germany, in particular, uses a mixed-member system that preserves the single-member districts, and supplements them with party seats. What stays in place are two large, stable parties, the CDU and the SPD and a few minor parties - the Free Democrats, the Greens, and the East German Party of Democratic Socialism. Coalition govts in these countries are generally stable and usually last a full-term. So coalition politics *can* work, and in systems that set a high enough threshold of the vote to enter Parliament (in Germany it's 5%), stability is the general rule. That doesn't mean there's never instability - but that's true of our system too. In the Netherlands, for instance, a couple years ago, there were two elections within months. But in those countries those are the exception, not the rule.

Indeed, there is a strong movement to adopt a form of PR in Great Britain and in Canada, which, along with India, France, and the US, are the only established democracies that don't use PR.

THAT said, you are right about a few things. All the examples I have cited are parliamentary systems, where coalition governments are necessary and even desirable. Purer forms of PR would work less well in a presidential system, and we have our own history and traditions that need to be respected in any kind of change. Because we are a presidential system, it is desirable to maintain a system that is at its core, 2-party (although having some minor parties in Congress would not be a bad thing). Multiparty presidential systems have a poor track record, and having amorphous, changing coalitions running for president or dominating Congress could lead to more gridlock with our system of checks-and-balances.

Even so, there are reforms that could move us towards a system that rids us of the bad effects of single-member, winner-takes-all elections (incl. extremism, lack of minority or female representation, lack of other voices, lack of ANY third-parties) while recognizing the essential uniqueness of the American system. In a 1998 Boston Review collection of articles on PR, the responses seemed to reach a certain consensus: our current system could be improved and most settled for the first system that I am going to describe:

1. Choice Voting: Though usually grouped as a form of proportional representation, choice voting actually isn't one. What it calls for are multimember districts, usually of between 3 and 5 members, although conceivably it could go higher. A threshold is established by taking 1 over the number of seats + 1. So, in a 3-member district, the threshold is 1/4.

Voters rank their choices and vote by individual, not by party. All first choice votes are counted first. Any candidate that wins above the threshold, in this case, 1/4 is automatically elected. All surplus votes for the candidate - anything above 1/4 - are automatically recast for their second choice. If still no candidate has passed the threshold, then the lowest scoring candidate is eliminated. This process is continued until all the seats, in this case, 3, have been filled.

Such a system could work well in the US if we were to adopt 3 to 5 member seats wherever applicable. Similar forms of voting were used on many town councils in the early 20th century and Illinois used a similar system until the 1970s. There is a strong movement to return to it in that state.

The main downside to this are the complexity of the vote count and counting the ballots. It should be noted that there's nothing very complex for the voter - they simply have to rank their choices like in Instant Runoff Voting, and that's it. However, vote-counting mechanisms would have to be well-developed. Australia offers an example - the 20-million strong country uses it for senate elections.

For more info on choice voting, see: http://www.mtholyoke.edu/acad/polit/damy/BeginnningReading/PRsystems.htm#stv

An alternate form of this would use a semi-proportional voting system called cumulative voting. This was actually the system Illinois used. Every person has as many votes as there are seats. Voters can distribute their votes however they want. For instance, in a 3-member district, they could cast all 3 votes for one candidate, 2 for 1, or one for all three. Such a system tended to reinforce a two-party system but also promoted diversity of views, and allowed both ethnic and political minorities (i.e. downstate Democrats and Chicago Republicans) to get representatives.

See: http://www.mtholyoke.edu/acad/polit/damy/BeginnningReading/semiproportional.htm#cumulative

2. Alternate Vote Plus - The other proposal is one that was devised by an extensive official commission in Great Britain in a 1997 study on scrapping the first-past-the-post system of elections.

Called Alternate Vote Plus, the system calls for IRV in single-member districts, and for having roughly 20% of parliament elected by party, as at-large representatives elected in super-districts - groupings of countywide or citywide parliamentary districts. See: http://www.electoral-reform.org.uk/votingsystems/systems4.htm#AV+

In the US, such a system could work like this: expand the house to about 585 (a 1/3 increase). Distribute the votes by state as is done currently. Ban gerrymandering, and have 3/4 of each states seats elected in single-member districts by IRV. However, also have citizens cast a vote for a party. Once all the districts within a state are tabulated, the party vote is tabulated, and the at-large seats (the extra 1/4) would be distributed to make the overall share more closely approximate the party vote. Theres a complex mathematical formula to doing this and it wouldn't be exact, but it would allow for some proportionality, while still making majorities dependent on winning lots of individual seats. A threshold could be set natioally state to keep out narrow, extremist or explicitly regional parties. And, if allowed by the Supreme Court, small states, for example the Great Plains and Mountain states, could form multistate at-large districts for at-large seats, while preserving their own individual seats for each state.

I believe either system would be superior to our current one. Even cumulative vote, or limited vote (yet another semiproportional system, related to cumulative voting) would be superior. None would be too foreign for America. All would put a premium on electing inviduals, not parties, and all would reflect the diversity of views better, without wrecking the two-party system or creating greater instability in the long term. Of course, any change would probably produce some short-term instability as the political system adjusted.