The Bow Group argues that AV will bring an end to conviction voting

Richard Mabey is the author of Death of the Conviction Voter - Fairness and Tactics under AV, which is published today by the Bow Group.

It is no closely guarded secret that most proponents of electoral reform wanted to see some form of proportional representation (PR) on the table in the upcoming referendum.  The argument they make is that forms of PR are the ‘fairest’ voting systems available.  Reluctant to make the same argument for AV, advocates of reform have advanced the curious proposition that AV is not the ‘fairest’ system, but simply that it is ‘fairer’ than the current First Past the Post system.

But is the type of AV on the table in the referendum really ‘fairer’?  My paper, Death of the Conviction Voter - Fairness and Tactics under AV, published today by the Bow Group, tackles these questions by looking at the only places in the world that use the system.  We have on the table a surprisingly rare form of AV, which is only used at a local level in the Australian states of Queensland and New South Wales.  Whereas the flaws of AV in general have been well documented, the additional challenges posed by this particular form of AV to the ‘fairness’ argument have not yet fully come to the fore. 

The referendum ballot paper will ask us whether we want to change our national voting system to “the Alternative Vote”.  It gives no guidance as to which type of AV we would get.  The version of AV proposed for the referendum is better described as Optional Preference Voting (OPV), the salient feature of which is the ability of a voter to mark on the ballot paper as many or as few candidates as he or she wishes.   It has been remarked that this form of AV produces an inequality where some voters get one vote and some get three.  But what effect does it have on results?

The paper uses data from the 2007 and 2011 state elections in New South Wales and Queensland to examine outcomes under OPV and the likely influences on voter choice.  The data suggests the following surprising conclusions:

1. What has been proposed by proponents of AV as the certainty of the winning candidate receiving at least ‘50% plus one’ of voter preferences is not true for the form of AV on the table.
2. This form of AV offers a far wider range of tactics to voters and political parties than those available under both FPTP and other forms of AV.
3. Up to 65% of voters might cast only one preference in a future general election.
4. Voters who mark their preferences with conviction will be disadvantaged in the electoral process.
5. AV, in the form proposed, would not increase fairness in voter outcomes.

An institutionalisation of tactical voting under OPV will leave us with divisive and unfair elections.  Noting that other forms of AV, such as those facilitating the national voting systems of Australia, Fiji and Papua New Guinea, do not entail these flaws to anywhere near the same extent, my paper argues that this ‘optional’ factor delivers yet another blow to the fast-fading AV logic.  Even if you think the current voting system should be changed, is it really worth changing our constitution for a system that will lead to less fair outcomes?  I urge you to vote ‘No’ on 5th May.