[ARIN-consult] Consultation on Implementing Single Transferrable Voting for ARIN Elections

[William Herrin]

On Thu, Jan 6, 2022 at 8:38 AM ARIN <info at arin.net> wrote:
> One recommendation that has arisen from this governance review is to replace the current first-past-the-post (FPTP) voting system with single transferrable voting (STV), also known as ranked-choice voting. STV would provide the ability for the community to realize the following primary benefits:
>
> • Greater voting choice. Voters rank candidates in order of preference;
> • A more proportional result. Successful candidates more broadly represent voter sentiment versus a single issue. There is better representation of member minority views;
> • No wasted votes. Fewer votes are cast for losing candidates or needlessly cast for run-away winners;
> • Designed for multiple winners; and
> • Reduced opportunities for tactical voting.
>
> STV has some disadvantages:
>
> • It is possible for a single candidate to win without crossing the winning threshold;
> • The process is more complicated to understand and implement than FPTP; and
> • In some cases, ballots that don't rank all candidates may be discarded.

Though unlikely, it is mathematically possible for the folks who
receive the most first-round votes to all lose the election while the
second-to-last candidates win. It's also possible for the candidate
ranked LAST by a majority of voters to win election.

Suppose you have five candidates for two positions. The votes are:
30% 1, 3, 2, 5, 4
20% 3, 1, 2, 5, 4
19% 4, 1, 2, 5, 3
10.5% 2, 3, 1, 5, 4
10.5% 2, 4, 1, 5, 3
5% 5, 4, 1, 2, 3
5% 5, 3, 1, 2, 4

In the first pass, one receives 30%, two receives 21%, three receives
20%, four receives 19% and five receives 10%. Candidate five is
eliminated.
5% of candidate 5's second choice was candidate 4. The other half was
candidate 3.

In the second pass, one has 30%, two has 21%, three has 25% and four
has 24%. Candidate two is eliminated.
Candidate 2's second choices were again split half and half between
candidates 3 and 4.

In the third pass, one has 30%, three has 35.5% and four has 34.5%.
Now the peculiarity of STV versus ordinary ranked choice voting comes
into play. Candidate 3 has exceeded the election threshold so his
"excess" votes are distributed to the other candidates. Let's say the
excess is 1%. The next choice for everybody now in the candidate 3
bucket is candidate 1, so all of that excess goes to candidate 1.

In the final pass, one has 31%, three has 34.5% and four has 34.5%.
Candidates 3 (originally 20%) and 4 (originally 19%) are elected,
between them having received less than half the first-round vote. Note
that elected candidate 4 was ranked LAST by a majority 65.5% of the
voters.

FPTP uses the same numbers above except voters cast a vote for both of
their first two picks, that is:

30% 1, 3
20% 3, 1
19% 4, 1
10.5% 2, 3
10.5% 2, 4
5% 5, 4
5% 5, 3

This results in:
69% 1 (elected)
65.5% 3 (elected)
34.5% 4
21% 2
10% 5

A simple plurality vote would use only the first choices:
30% 1
20% 3
19% 4
10.5% 2
10.5% 2
5% 5
5% 5

For a total of 30% 1, 21% 2, 20% 3, 19% 4 and 10% 5 electing
candidates 1 and 2 who between them achieve a majority of the votes.

Regards,
Bill Herrin


--
William Herrin
bill at herrin.us
https://bill.herrin.us/