David Kirk’s main argument (Solidarity 3/191) is that AV will help left-wing “propaganda candidates”. But with Australia’s AV left and not-so-left minority party candidates have generally done worse than with Britain’s FPTP setup. This seems odd, but it is a fact. Knowing that fact, left and pseudo-left groups focused on electoral activity — Socialist Party, Respect, Green Left — oppose AV.
With AV people know that their vote will count towards the result only when it transfers to the bigger party they’ve chosen as second preference, so they often cut out the middleman and vote for the bigger party direct.
In Australia the main bias of AV is to polarise electoral politics into two blocs, organised around the two big parties (Labor and Liberals) to which the smaller parties in each bloc transfer.
Argue for the British Labour Party not to do preference swaps? You could, I suppose, but the chance of anyone listening is zero. Under AV a party eschews preference swaps only if it has no interest in winning (and usually not even then).
The evidence from Australia is that parties’ recommendations on preference swaps have surprisingly great effect.
AV is no more democratic than FPTP, maybe less so. The detailed balance will depend on how the Lib Dems choose to work the system, and whether British voters react to AV differently from Australians, neither of which we know.
AV would however be more stable than FPTP (there is no outcry to change it in Australia), so voting in AV would gazump any other electoral change for, probably, decades.
The fact that AV will ensure that the Lib Dems “win” the next election however we vote (even though longer-term it may hurt them), and the fact that the referendum will be at least partly a referendum on the government, indicate no to AV.