Strategic Thoughts

bannerspacerAbout Me | Mail Me | Linksbannerspacer2

 

Please See NoSTV.org

November 26, 2008

STV Makes Your Vote Worth Less

Some folks claim that BC-STV (The Single Transferable Vote) will "make your vote count". Actually BC-STV can make your vote worth less and make your MLA less accountable.

Inequality is inherent in BC-STV. The Northeast (Peace River) would get two MLAs while the Capital Region would get seven; some voters would see their vote dead-ended, not electing anyone and not transferred, while others would see their vote help elect more than one MLA. What's fair about that?

Ballot showing vote for the existing electoral systemNo one should vote for STV unless they can understand how votes are counted. Supporters of BC-STV say that it doesn't matter if people don't understand how votes are counted. They argue that most people don't know how their car works but they can still drive it. That analogy is misleading. Voters are asked to compare BC-STV to the existing first-past-the-post system (FPTP) and decide between the two on election day, May 12th. In deciding whether to buy a GM or a Toyota vehicle, prospective purchasers need to know a lot more than just how to drive each car, such as whether the company will be in business next year and whether the warranty will be any good. In deciding between voting systems, British Columbians need to know a lot more than just how to vote. They also need to know how votes are counted in order to be able to adequately compare FPTP and BC-STV. B.C. voters are not test driving BC-STV; we could be living with it for decades

BC-STV would result in 20 multiple-MLA constituencies, each of which would elect from 2 to 7 members of the legislature for a total of 85 MLAs. The constituencies would not be equal. Large multiple-MLA constituencies would be formed by merging the single MLA constituencies from the 2009 election. The Capital Region would have 7 MLAs; the Northeast would have 2 MLAs; most constituencies would have 4 or 5 MLAs. The area covered by the Northwest, which would have 3 MLAs, is more than 5 times larger than the Republic of Ireland.

In the existing first-past-the-post voting system, whoever gets the most votes wins. Voting is simple; every voter gets one vote for one MLA. Up until 1986 BC had some constituencies that elected two MLAs, but there is nothing similar between 2-MLA constituencies in 1986 and 2-MLA constituencies under BC-STV. In 1986 voters in 2-MLA constituencies could cast two equal votes and the two candidates who got the most votes won. With BC-STV a voter would rank candidates starting with a "1" for the candidate most preferred, and continuing with 2, 3, 4 and so on. The numbers are instructions on how to count each voter's single vote.

The first word in STV is "single". That is exactly what it is, a single vote, even though a constituency will elect from two to seven MLAs. With BC-STV the minimum number of votes required to win is called the Droop quota: one plus the number of valid votes divided by one plus the number of MLAs to be elected. In percentage terms the quota is equal to 12.5% of the total votes cast in a 7-MLA constituency, rising to 33.3% of the total votes cast in a 2-MLA constituency. Those percentages are important because any votes in excess of the quota get redistributed to other candidates based on the instructions each voter gave by way of candidate rankings.

The vote count under BC-STV starts by counting how many first preferences each candidate receives. If any candidate receives more than the quota, the candidate is declared elected and the excess number of votes above the quota is allocated to second preferences. Since voters are unlikely to have the same second preference, the second preference of every voter who ranked the elected candidate a "1" is examined and a fraction of each vote is reallocated to second preferences (the fraction being the ratio of the excess over the quota divided by the total first preferences).

When there are no longer any excess votes from winning candidates to transfer, the candidate with the lowest number of first preferences is eliminated and the votes for that candidate are transferred at their full value to their second preferences (full value because no portion of their vote counted towards their first preference). It can take a dozen rounds of adjusting votes before the count is finished. The last candidate to be elected usually has fewer votes than the Droop quota, because there remains one position to fill and no further votes to transfer. The remaining candidate with the most votes in the final round is declared elected. That means that in a 7-MLA constituency, the 7th candidate to be declared elected wins with less than 12.5% of the total vote.

The Citizens' Assembly attempted to explain the complicated counting procedure by providing a cartoon animation showing the vote count in a make-believe election with 3 representatives to be elected and 5 candidates running. Click on their animation to see how fractions of second and third preferences are transferred. In the simplified version those who voted for the fictitious candidate Eric Elderberry never have their second preferences considered, while those whose first choice was for Amanda Apple and whose second choice was for Bill Banana also get to have a fraction of their vote count towards their third preference. It is unfair to have some votes stop with a first preference while others get fractionally distributed over several candidates.

For a real-life example, look at the actual count in the May 24, 2007 Republic of Ireland election for the district of Dublin North which had 4 representatives to elect and 13 candidates. The website ElectionsIreland.org allows you to see how votes were counted in each of Ireland's multiple-member constituencies by just clicking on the name of the constituency and then on the top tab for "count details". Browsing through those actual vote counts is a good way to see how complicated vote counting is under STV, and that there are always some voters whose vote doesn't count as much as the vote of others because it is cast for someone who survives until the last round of the count, doesn't get elected and doesn't have second preferences transferred. In Dublin North, the vote count took 10 rounds of redistributing votes but the 5,256 people who voted for Brendan Ryan (who lost) did not have their second preferences transferred. You can pick any other real life example of STV and you'll see that there are always some voters whose vote doesn't get transferred and whose first preference doesn't win.

Anyone who has ever had more than one boss at the same time for the same job knows that accountability can go out the window. With BC's existing system of single MLA constituencies, accountability is clear. If you don't like what your MLA did, or what your MLA's party did, vote for someone else. With 5 MLAs representing one enormous constituency each could say a problem is the other guy's responsibility or fault.

From an MLA's point of view, large multiple member regions would make it impossible to service all the school boards, municipal councils and community organizations that would be in regions two to seven times larger than our existing constituencies. Supporters say that candidates would carve out their own constituencies within the large regions, which, if true, is another way of saying they would ignore large numbers of voters in the region since they would know that they could get elected, not with the most support, but with a minimum support of 12.5% to 33.3%.

Our existing first-past-the-post (FPTP) system is not perfect, but it is better than BC-STV.

 
 

About Me | Mail Me | Navigation | Top
2008 David D. Schreck. All Rights Reserved.