Is preferential voting really that difficult?

There’s been a shortage of European elections in the last couple of months, but the British Labour Party is trying to make up for it by having a fascinating election for its leadership. Members should receive their ballot papers this week, with voting closing on 10 September.

Labour was defeated but not disgraced at the May general election, winning 30.4% of the vote, a modest increase on 2010. If it can get its act together, and perhaps bring itself to negotiate some sort of potential coalition with the Scottish nationalists, it should certainly be competitive in 2020. But first it needs a new leader.

This is the first leadership election under the party’s new rules, under which votes are exercised only by individual members and supporters – unlike the old electoral college in which MPs and affiliated unions each also got a third of the vote. That’s a big win for party democracy, but as with any such system (compare “open primaries” in the United States) it leaves the party vulnerable to charges that people with no loyalty to Labour are signing up in order to influence the result.

The candidate that Labour’s enemies are assumed to be supporting is Jeremy Corbyn, the one left-winger in the field and currently a hot favorite in the opinion polls and (even more impressively) the betting markets – Ladbrokes has him at an extraordinary 4-1 on. His three moderate opponents, Andy Burnham, Yvette Cooper and Liz Kendall, have manifestly failed to excite much enthusiasm.

I’m not convinced that Corbyn is as much of a certainty as those numbers suggest, but even so, the contest says a lot about the current state of British and even world politics. I’ll try and get to some of those things in future posts. For now, I want to point out the way it also shows up the strange British failure to understand democratic voting systems.

The leadership election is a single ballot, with optional preferential voting, which the British call the “alternative vote”. In 2011, a referendum overwhelmingly rejected a proposal to introduce this for national elections, and indeed for a parliament of single-member constituencies it would have been only a marginal improvement on the existing first-past-the-post system. But for a one-off position like the Labour leadership, it’s the obvious way to go.

Yet its subtleties, familiar as they might seem to some of us, still elude most of the British political class. Today’s exhibit is a piece in Friday’s Guardian by Alberto Nardelli. (Hat tip to Sol Salbe for drawing it to my attention.)

My point is not that Nardelli is outstanding in his ignorance: on the contrary, much of what he says is an admirably clear and accurate explanation of the system and its implications. But he also shows how easy it is to slip up. His particular theme is tactical voting: where a voter may seek to gain an advantage by voting for someone other than their first choice. He says:

In many cases, the purpose of tactical voting is not about supporting your preferred option, but about minimising the chances of a worst-case scenario when your preferred outcome is all but impossible.

Well, yes, that’s true. Indeed, under first-past-the-post voting, which Nardelli is presumably used to, that’s the normal case. But with preferential voting, it’s often somewhat different: it’s about trying to get your best alternative rather than your second-best by doing something that looks counter-intuitive.

For example, consider a rural electorate in Australia where everyone knows that the Liberals will have around 40% of the vote and Labor and the Nationals will have something like 30% each. Liberals and Nationals will swap preferences, so for a Liberal supporter what they most want is for the Nationals to come third; if Labor comes third, there’s a risk that their preferences will elect the Nationals. So it makes perfect sense for such a person to give Labor their first preference; Labor isn’t a threat, but they want to try to knock out the Nationals.

So, contrary to what Nardelli says, the fact that Corbyn can be assumed to lead on first preferences doesn’t mean that his supporters have no opportunity for tactical voting. That depends on the preference behavior of the other contenders.

If we assume (as Nardelli does, and as seems reasonable from the polls) that Corbyn will definitely come first on primaries and Kendall will definitely come last, then the opportunity for tactical voting depends on how tightly you think the three different anti-Corbyn candidates will swap preferences. If you assume that they will do so equally well (or badly), then there is nothing to be gained by tactical voting, either by a Corbyn supporter or an anyone-but-Corbyn supporter.

It’s simply not true to say that “If Kendall’s supporters were to place her first and indicate Burnham or Cooper as their second preference, or express no second preference at all, they would risk pushing Corbyn over the line.” If everyone’s preferences are equally tight, a vote for Kendall with a preference to the other two moderates is no different to a primary for Burnham or Cooper.

If, on the other hand, you think that, say, preferences from Burnham to Cooper will flow much more reliably than vice versa, it would make sense for an anti-Corbyn voter to vote “1” for Cooper, even if she was not their first choice. But in that case, for the same reason, there would be a possible (albeit very risky) advantage to Corbyn supporters in voting for Burnham, since he is less of a threat due to getting less of a preference flow.

Nardelli continues:

Supporters of Cooper or Burnham who view the choice as between “left” and “right”, and are worried – improbably – that Kendall might emerge as the winner of an anyone-but-Corbyn drive, should consider giving their second preference to Corbyn. With enough second preferences, he would most probably win the leadership.

But this makes no sense either: even if they prefer Corbyn to Kendall, that’s no reason for Cooper or Burnham voters not to swap preferences. They can have just as much effect by then putting Corbyn third – although since ex hypothesi that represents their genuine preference, it no longer counts as tactical voting.

Again, my point is not to beat up on a specific pundit. For anyone who observed the 2011 referendum campaign – where the use of “Australian” as a bogey word trumped any sort of actual analysis – the lack of British engagement with democracy seems very much the rule rather than the exception. I’ve lost count of the number of times I’ve criticised the BBC for describing someone as “winning” an election when in fact they had only a bare plurality and a majority coalition could easily be formed against them.

No doubt the Corbyn surge has many causes, but one of them is surely this entrenched perspective of first-past-the-post.

One thought on “Is preferential voting really that difficult?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.