[EM] A Vermont Rep comments on Vermont's House election

[Donald E Davison]

  ------------ Forwarded Letter -----------
Date: Thu, 20 May 1999 13:32:09
To: donald at mich.com (Donald E Davison), Rob Richie <FairVote at compuserve.com>
From: Terry Bouricius, Vermont House Rep
Subject: Re: Election Upgrades and Swing Voters
Mime-Version: 1.0

Here are some comments from Rep. Terry Bouricius of Vermont about the two
seat district idea in practice.
My comments are in ALL CAPS interspersed.
-Terry B.
P.S. I am a Progressive Party member elected from a single seat district.

At 08:36 AM 5/20/99 -0400, Donald E Davison wrote:

>  ------------- Forwarded Letter One ------------
>From: Richard Foy
>
>It is noted that the Democrats won the majority of the seats in both
>the single seat and two seat districts. I suspect this is a result
>of a redistricting that was controlled by the Democrats.
THIS IS CORRECT.  ALTHOUGH THE REPUBLICANS HAD A NOMINAL MAJORITY, THE
DEMOCRATS CONTROLLED THE HOUSE AND THE REDISTRICTING IN 1991 THROUGH
SWEETHEART ARRANGEMENTS WITH A FEW REPUBLICANS.

>I also note that the largest third party did better in the two seat
>districts than their percentage of the voters while all the other
>smaller parties did worse. That is the vote tended to become more
>of a three party vote.
I AM ONE OF THOSE ELECTED THIRD PARTY LEGISLATORS.  THERE WERE NO THREE-WAY
RACES.  THE PROGRESSIVE PARTY CANDIDATES WON IN RACES AGAINST DEMOCRATS
WITH NO REPUBLICAN CANDIDATES.  WE ARE "OVER-REPRESENTED" BECAUSE WE ONLY
RAN RACES IN DISTRICTS IN WHICH WE WON.  SINCE WE DID NOT GET 100% OF THE
VOTE, WE EXCEEDED OUR PROPORTION.  THE ONE OR TWO-SEATEDNESS OF THE
DISTRICTS IS IRRELEVANT IN THIS CASE.

>            Six Upgrades to the Single Seat District Method
>
>     If we are starting with Single Seat Districts using Plurality(FPTP),
>then what follows is a series of Upgrades, anyone of which, if taken, will
>be an improvement over our starting position.
>
>Upgrade One: Pair up all the Single Seat Districts to form two member
>districts. Two votes per voter in each district, which is
>Plurality-at-Large. Top two candidtes are the winners
BELIEVE ME, THIS IS NO IMPROVEMENT AT ALL.  IT SIMPLY ALLOWS A VOTERS TO
UNDERCUT THEIR FAVORITE CANDIDATE BY EFFECTIVELY VOTING AGAINST HIM/HER
WITH THE EQUAL-WEIGHTED SECOND VOTE.  A CANDIDATE THAT IS THE MOST
PREFERRED BY A MAJORITY OF THE VOTERS CAN LOSE IN SUCH AN ELECTION (I HAVE
SEEN IT HAPPEN!).

>Upgrade Two: Same number and size of districts. The change in this upgrade
>will be to only have one vote per voter, which is Limited Voting, a
>proportional representation method. The top two candidates are still the
>winners.
THIS IS THEORETICALLY BETTER, BUT IN PRACTICE IT WILL TEND TO REDUCE VOTER
CHOICE, AS INCUMBENTS WILL SEEK TO AVOID HAVING RUNNING MATES.  MOST
DISTRICTS WILL SIMPLY HAVE A NO-CHOICE ELECTION WITH ONE D AND ONE R
RUNNING FOR THE TWO SEATS.

>Upgrade Three: Same number and size of districts. Same one vote per voter.
>The change in this upgrade will be to allow the voter to rank choices, and
>the method in this upgrade will be Choice Run-Off(Alternative Vote). The
>top two candidates are still the winners.
>
>Upgrade Four: Same number and size of districts. Same one vote per voter.
>The change in this upgrade will be to use Choice Voting(STV) to work the
>ballots down to two candidates. Those two are the two winner in the
>district. Same quota is used for all districts - total votes divided by
>total members.
>
>Upgrade Five: Same number of districts. The change in this upgrade is to
>give the voter two votes, one for Choice Voting in his district and one for
>any party that has at least one candidate running in any district.
>     Choice Voting is to work the district votes down to two candidates.
>The top candidate is the district elected member.
>     All the candidates of all the districts are to be placed on party
>lists in an order according to the highest number of votes they reached
>during their district elections. The elected district members are to be
>placed at the top of their party lists. Independent candidates are to have
>a list of their own.
>     The party votes are tallied and the balance of the members to be
>elected are decided by the Party List method.
>     This upgrade is a form of Mixed Member Proportional
>Representation(MMP) with some differences. Upgrade Five uses Choice Voting
>in the districts and Open Party Lists.
>
>Upgrade Six: Same number of districts as five. The change in this upgrade
>is to combine the two votes per voter by allowing the voter to rank
>candidates and/or parties with one vote. There are more details to this
>Upgrade Six, but in general it is something like the following:
>     There will be no separate primary. The primary is to be in the general
>election. One member will still be elected in each district per Choice
>Voting. Party List will still elect the balance of the members.
>     The value of this upgrade is that it avoids the distortions of cross
>party voting that occur in MMP and primaries.
>
<snip>
> Don wrote: >>Note: I did not include the data of candidates that were
>>> nominated by two parties.
>
>LBR wrote: >This might prove among the most interesting bits of information
>> but would require additional analysis as to the reasons for nominations
>> by more than one party.  If parties can unify behind a single candidate,
>> what might be learned?
>- - - - - - - - - - - - - - - - - - - - - -
>Don writes:
>
>Dear LBR,
>    Offhand, it seems to me that most of the action of nominating
>candidates by two parties is the result of a shortage of candidates. Here
>are two examples:
>
>District: Orleans-Franklin-1     Number of seats:  1       District Results
>Starr, Robert A.                    Democratic/Republican       1160  95.7%
>Total Write-ins                                                   52   4.3%
>
>District: Windham-2-2            Number of seats:  1       District Results
>Bristol, Gordon E.                  Democratic/Republican        929  96.6%
>Total Write-ins                                                   33   3.4%

DONALD IS CORRECT.  IN SOME DISTRICTS, INCUMBENTS ARE ASSURED RE-ELECTION
SO NOBODY BOTHERS TO RUN IN THE OPPOSING PARTY PRIMARY.  THUS THE INCUMBENT
WINS THAT PRIMARY AS WELL WITH WRITE-IN VOTES.

[Bart wrote:] >It's hard to believe that Plurality-At-Large could produce those
>results.  I wonder how?
>
>Maybe other factors are at work, such as limited resources.  If you have
>52% Democrats and 48% Republicans, you would expect P.A.L. to elect two
>Democrats, i.e. the results should come out D-D-R-R.  But if you
>consider competition for cash, volunteers, air time, media/public
>attention, etc., it may be that either party is better off putting most
>of its eggs in one basket.  This would give you D-R-D-R.
>
>If the Democrats in this example tried to promote both candidates
>equally, the lead candidate could lose position, i.e. R-D-D-R.  This
>might be worth a try for the Dems (if they know for sure they are in the
>lead), but the Republicans are better off concentrating on one
>candidate, in order to keep at least one seat.
THIS IS EXACTLY WHAT HAPPENS IN VERMONT.  IN TWO-MEMBER DISTRICTS EACH
PARTY TENDS TO HAVE A SERIOUS CANDIDATE AND A"SPONGE" CANDIDATE.  THE
SPONGE IS SIMPLY INTENDED TO SOAK UP THE STRAY VOTES OF PARTY LOYALISTS SO
THEY DON'T GO TO THE LEAST UNDESIRABLE OF THE OTHER PARTY BY DEFAULT.
INCUMBENTS TRY TO AVOID HAVING STRONG RUNNING MATES, SINCE THAT CAN
INCREASE THE RISK OF THEIR OWN DEFEAT.  ON THE OTHER HAND THEY DARE NOT RUN
WITHOUT A RUNNING MATE UNLESS THE OTHER PARTY INCUMBENT ALSO HAS NO RUNNING
MATE.  SUCH QUID PRO QUOS ARE ARRANGED.

>Dear Bart,
>
>     The answer to your question, "I wonder how?", is "the swing voters".
>
>     In a Plurality-at-Large election, if a party has fifty percent plus
>one vote, that party will have the power to elect all the members. But, in
>some of the Vermont Two Member districts the two elected members are of two
>different parties. An example:
>
>District: Windsor-2-2            Number of seats:  2       District Results
>Lehman, Ralph W.                    Democratic                  1176  27.4%
>Kainen, Michael R.                  Republican                  1170  27.3%
>Bohi, Lynn                          Democratic                  1139  26.6%
>Trottier, Joe                       Republican                   803  18.7%
>Total Write-ins                                                    2   0.0%
>
>     I conclude from this that no major party has fifty percent plus one.
>In these districts the major parties each have a minority of the votes. The
>balance is held by the swing voters, who are known as voters who will vote
>for the candidate of one major party in one election and then maybe vote
>for the candidate of another major party in the next election. Only in this
>case these swing voters seem to be swinging their votes in the same
>election. They are giving one of their votes to one candidate of one party
>and then giving their second vote to a candidate of another party. There
>are two possible reasons for this action.
>     Most people regard politics as a game of "Us against Them", but there
>is a growing number of people who believe that all groups should be invited
>to the table to decide government policy. Some of these swing voters may be
>exercising this belief by making a point of voting for two different
>parties with their two votes.
>     The other reason is that some swing voters are voting for the two best
>candidates regardless of party. At times the result will be two elected
>members from two different parties.
THIS IS THE EXPLANATION MOST VOTERS IN VERMONT WOULD USE - THEY PUBLICLY
DISDAIN STRAIGHT TICKET VOTING.

>     The major parties will run two candidates because the party expects to
>capture the two seats, but the swing voters treat this as a primary inside
>the general election. The swing voters are deciding which candidate is to
>be the top candidate for each major party.
>     While it was not the law in Vermont for a primary to be included in
>the general election, it has worked out to be so in these cases, and if I
>may add, I feel it works nicely. I have long favored having the primary
>inside the general election. That is something I ask everyone to consider.
>This example and the examples below show that having the primary in the
>general election is both workable and acceptable.
AS SOMEONE SUFFERRING UNDER VERMONT'S VOTING SYSTEM, LET ME SAY IT DOES NOT
WORK  NICELY.  NOT THAT THE OPEN PRIMARY WE HAVE WORKS WELL EITHER.   I
BELIEVE PARTIES SHOUD BE ABLE TO SELECT CANDIDATES FAITHFUL TO THEIR
PLATFORM (BY CONVENTION OR OTHER MEANS), AND THAT ROLE SHOULD NOT BE GIVEN
OVER TO SWING VOTERS WITH NO CLEAR POLITICAL IDEOLOGY.  THAT ENCOURAGES
ELECTIONS TO DEVOLVE TO NON-POLICY CHOICES WITH SLOGANS LIKE "VOTE FOR
SMITH, HE LISTENS!"  OPEN PRIMARIES, SUCH AS VERMONT'S, HAVE HELPED CONFUSE
THE PROCESS OF PARTY CANDIDATE SELECTION WITH THE PROCESS OF POPULAR
GENERAL ELECTION.

[end]