Who should Green voters back with their constituency vote in next year’s Holyrood election? I am not aware of Green candidates standing in any of the constituencies this time, although that’s a decision for individual branches to make. Assuming I’m right about those branch decisions, the dedicated Green voter will have to look elsewhere if they cast a constituency vote.

Is there a clear policy answer?

Surely this is the best place to start? Where a Green voter faces a choice of Labour, SNP, Lib Dem or Tory in their constituency, which of those four has been the best on the that motivate our voters? To oversimplify, I’ll look at social justice, civil liberties, the economy and of course the environment.

The Tories and the Lib Dems obviously score more poorly on social justice, the economy and the environment given the Westminster coalition, but if civil liberties are your priority, they might make sense (with a couple of serious reservations). In fact, if civil liberties come top for you, you’d be voting for anyone but Labour, although most of their failures on this front have been at a Westminster level, not Holyrood, but should we disassociate the two when Labour campaigns in Westminster elections on knife crime etc?

On the economy the choice is harder. All four other parties backed the bankers and fell for the idea that boom and bust was over, during an unstable boom. None of them question the economic system so comprehensively excoriated by Neal Ascherson in the Sunday Herald a while back (perhaps the best thing I’ve read all year). None of the three parties who held Westminster office during 2010 have shown any inclination to make the tax system more progressive, given the impact of the increase in allowances. Similarly, neither the SNP’s Local Income Tax proposals nor their Council Tax freeze can plausibly be identifed as progressive. It also seems unlikely that any of the other parties will propose a revenue-raising alternative to passing on the cuts, an area where another progressive proposal from elsewhere might well have tempted Green tactical voters to commit a first vote.

On the environment, none of them have a great case to make, (though I have to bow to James’ more detailed knowledge on this score), but there are points of difference. The SNP score for being anti-nuclear, but lose for being pro-coal. Labour score for being against some coal at least, but lose for being pro-nuclear. On climate change targets, the Westminster Coalition parties and the SNP lose points for voting for very weak targets last year, where Labour get a grudging half point for abstaining. Not one of them gets a single point on transport: all four other parties back every one of the unpopular motorway schemes currently under consideration, and all four back airport expansion.

In short, there’s no clear guide on policy for the Green voter looking at the constituencies, and it would depend on what each voter’s policy priorities were.

And looking at the tactical votes?

Now we’re talking!

One of the widely touted advantages of PR, of course, is an end to tactical voting. In general that’s true – with STV, you should just vote your first preference first, then rationally go down to the penultimate candidate or party. With a pure party list system it would only make sense not to put your real first preference party first if you thought they couldn’t win, and if there was a decent enough compromise party worth backing instead.

However, Scotland’s system, as the fellow anoraks who read this blog know, isn’t pure PR – it’s AMS with 73 seats elected by the tired old First Past The Post system and then “topped up” by the second vote in the region.

Of course, before you start counting the list votes for each region, you divide each party’s vote by the number of constituencies they won, plus one. That neatly avoids having to divide by zero, of course, which is infinity, or zero, depending on your mathematical know-how.

This means that, in Glasgow for instance, Labour list votes have always been irrelevant. In 1999 and 2003 they won every constituency, and even in 2007 only a Nicola-shaped pocket of neon yellow punctuates the sea of pale red.

This makes it a hard region for Greens, Conservatives and Lib Dems to win seats in off the list, but Nicola’s win made it that little bit easier for the rest of us. If she’d lost in Govan, Patrick would have lost on the list, simple as that. So tactically, Green list voters (and indeed Lib Dem or Tory list voters) want to see her win again, and if Labour were somehow to lose another constituency that would help Glasgow’s three smaller parties too: that would be where the pure tactical interest lies.

To take a slightly different example, a hypothetical Green voter who lives in Edinburgh Pentlands would have had some tough choices over the year. From 1999 onwards it was clearly the Tories’ number one target in the Lothians, and one thing has been constant about their results in the capital’s region: they get two MSPs elected. In 1999 they lost out in Pentlands and took the compensatory list seat. In 2003 and 2007 they won Pentlands and freed up a slot in the lists.

If David McLetchie hadn’t won Pentlands in 2003, Colin Fox wouldn’t have claimed that last list slot behind Mark Ballard. Curiously, therefore, the purely tactical constituency vote in Pentlands for a Green, or even for a Socialist, would be for Tory MSP David McLetchie. There’s virtually no risk Parliament would have any more Tories in it, so why not?

Who might tactically vote Green?

The flip side of this question is to ask when it might suit supporters of other parties to lend a second vote to the Greens, and the obvious examples are regions where a party’s list votes simply don’t get MSPs elected. And there are loads of them.

For this purpose we can ignore the Tories, not just because Green is the second vote of typically about one Tory in fifty, but also because they have won list seats in every region at every election. Having opposed PR. Good work.

The SNP are also in a different category – they have won seats in every region at every election too, but we do attract a fair number of SNP second preferences. In fact, the case has previously been made that, given those pesky d’Hondt divisors, voters who prefer the Green position on the constitution get more bang for the buck voting Green on the lists.

Those entire regions where list votes simply don’t elect anyone from larger parties are most interesting, though. The table shows where list votes were simply discarded, election by election.

1999 2003 2007
Central Labour Labour Labour
Glasgow Labour Labour Labour
Highlands and Islands Lib Dem Lib Dem Lib Dem
Lothians Labour Labour & Lib Dem Lib Dem
Mid Scotland and Fife Labour Labour Lib Dem
North-east Labour & Lib Dem Lib Dem X
South Labour & Lib Dem Labour & Lib Dem Labour
West Labour Labour Labour

There has therefore been larger-party wasted list votes in every region and at every election except the North-east last time. In some cases it may be hard to predict where that will apply, but in others it’s a virtual certainty.

If you were a Labour voter in Central, Glasgow or West, wouldn’t you rather express a preference that might elect someone on the list? For sure, there’s a substantial statement made by all those discarded Labour list votes – “we are really loyal to Labour” – and you never know for sure how the constituency vote will go, but voting for another party could make a real difference. Are those voters really neutral about whether Glasgow has more SNP, Lib Dem, Tory or Green MSPs? Or do we need to explain the voting system better? The same applies to Lib Dems in the Highlands and Islands, whose list votes have never helped anyone get elected.

There are Labour activists who work on this basis, who split their vote to get Greens in on the lists instead of SNP or Tory MSPs. There has been (confidential) suggestions from an SNP activist that they might think about it in a region or two. But will it spread? And will the public follow suit?