Lee Rotherham: AV would not end tactical voting

By Lee Rotherham

Christopher Hall’s article on Thursday set out a useful and detailed insight into the Liberal Party’s historic views of AV. In particular, it exposed how that party has for over a century  seen it as a distraction from the real debate between the current system and PR – a tradition that is still honourably maintained by some.

Listening to an academic seminar yesterday, I was struck by another historic misconception, relating to the lessons learned not from Cowley Street’s past, but from elsewhere.

Both sides in the debate have argued about the relative complexity of AV. No have argued that it is a complicated system, which in turn threatens a need for counting machines, a suggestion incidentally endorsed in the Times letters pages by the man who used to run the London ballot. Yes have counterclaimed that it is “as easy as 1-2-3”, focusing on the process of the vote itself rather than the broader mechanics.

But the complexity issue goes much wider.

Perhaps Australia can provide some assistance here. E.J Nanson was the principal advisor and architect of the federal electoral system. Even he acknowledged that AV contained what he styled systemic and substantial voting rule inconsistencies. The fact that he would later go on to introduce the very system of which he had been sharply critical did not endear him to a number of subsequent commentators. But what was it that complicated matters?

Part of the problem lies in the machinery. If the gear stick is as easy as 1-2-3, the engine turns out to be a bit more tricky to follow. It’s what happens before and after the paper is marked that’s the real conundrum, and Australia shows us how in the form of the party lobbying underlying it. Look here, here and here and you can start to see the much deeper politics underpinning AV in practice, in the form of the how-to-vote card – the physical embodiment of tactical voting. It’s even murkier than that, because there have been infamous cases of campaigners being caught impersonating activists from other parties and handing out fake ones (see here for instance for a flavour).

So contrary to Yes predictions, AV does not end tactical voting (or clean up politics for that matter). It incorporates tactical voting directly into party politics as a standard election given. 

It may be more helpful to us if we put it into a British context. As we unfortunately discover in the chart below (which you should click to enlarge), AV does not end tactical voting in the UK either.

In single member constituencies, people will still want to reduce the chances of their bottom choices and improve those of their top, and the political parties will be only too glad to help. Tactical voting will still exist under AV. It just becomes more frustrating for the voter to work out.

Imagine a seat where, say, the third party vote will decide the winner. The voter who fears these swing votes will, between the two remaining candidates, predominantly go to his lower preference choice over his own top choice will be incentivised to keep this candidate in play. If this swing candidate is his own second preference, that encourages a tactical vote putting his second choice over his first. In turn, as the flowchart explores, that can have some very bizarre and counterintuitive consequences.

Of course, with AV itself only making a difference in one seat in twenty, the issue will only hugely vex voters in a minority of seats. But if this paradox exists, what other unconsidered consequences of introducing the AV system will we also discover to our cost should it come into play? “Complicated” I fear isn’t the half of it.