Winner of the New Statesman SPERI Prize in Political Economy 2016

## Sunday, 19 April 2015

### From seats to governments: UK general election arithmetic

This is written for UK readers, who will know who all the parties are. However you do not need to know the details to get the main points.

How do you translate seats into governments? So many parties, so many variables. Here is a suggestion, based on the idea that the number of seats that the big two are likely to get is much more uncertain than for the other parties. Then split those other parties into three groups: the ‘left bloc’ (mainly SNP, but also Plaid, Greens, Galloway and SDLP), the LibDems, and a ‘right bloc’ (NI Unionists plus UKIP). If Sinn Fein win 5 seats which they do not take up, there are 645 seats to play for, and a total of 323 seats gets you a majority.

Assume the left block gets 58 (50 SNP, 3 Plaid, 3 SDLP, 1 Green, 1 Galloway), the LibDems get 24, and the right block get 13 (e.g. 3 UKIP). That is a total of 95, leaving 550 to divide between Labour and the Conservatives. We can now split the result into a two dimensional set of possibilities depending on how the Labour/Conservative battle goes:

1) Con 323+ (Lab 227-) seats - Simple Conservative government

2) Con between 310 and 322 seats (Lab 240-228) - Coalition continues.

If you add the right bloc to 310 you get 323. In this case the Conservatives could shun the LibDems, but I suspect they will prefer to continue to work with the LibDems than having to rely on the right bloc, and the LibDems would find the attractions of continuing in government too strong to say no.

3) Con between 286 and 309 seats (Lab 264-241) - LibDems decide.

In this situation the LibDems have indicated they would ‘talk to the Conservatives first’, but in principle they could go either way: see below.

4) Lab between 265 and 298 (Con 285-252) - Labour SNP understanding.

265 plus the left bloc of 58 gets to 323. Would the LibDems get a look in? There is a possibility near the Lab 265 mark, where Labour might want to avoid the chance of one of the smaller left block parties causing difficulties. But it is not clear why Labour would want a coalition with the LibDems in this case, rather than something similar to their SNP understanding.

5) Labour between 299 and 322 (Con 251-228) - Labour choice

Labour would still have the option of an SNP arrangement, but it could alternatively form a coalition, or make an arrangement, with the LibDems. Some suspect Labour would find the politics of not being dependent on the SNP attractive, but they would probably want a new LibDem leader in exchange.

6) Lab 323+ seats - Simple Labour government.

At the moment the polls and models are suggesting we are in Labour-SNP territory, but only just, and things could easily slip into the ‘LibDems decide’ zone, with other outcomes still quite possible. If the LibDems do have to decide between Cameron and Miliband, things get very interesting.

There seems little doubt that Clegg and some other senior LibDems would favour continuing with the current coalition: the fact that Clegg has indicated he will talk to the largest party first indicates that, because in itself talking to the largest party makes little sense when the left block is so much larger than the right block. However most of his party members would more naturally favour the opposite arrangement. Furthermore, LibDem policies are generally closer to Labour than the Conservatives.

There are plenty of superficial sound bites that will be brought into play by the LibDem leadership to favour choosing the Conservatives: besides largest party talk, there is also the line of not wanting to be associated with a party that wants to break up the union (also largely nonsense). But what LibDem members really need to think about is the survival of their party if they continue with the current coalition when they have perfectly feasible alternatives. Imagine, for example, if the referendum went against continuing EU membership. Even if it did not, imagine the next two years when uncertainty about EU membership held back investment, and growth faltered under renewed austerity. Imagine the NHS finally collapsing from lack of funding. The Conservative party would recover from all those things, but the LibDems would always be seen as the party that chose to let this happen. I’m afraid I do not know enough about LibDem internal decision making to know how far the party would in practice be able to overrule its leadership in the days after the election. (If you do, please comment or email.) If they cannot, or chose not to, this zone merges with ‘coalition continues’.

Of course the number of seats obtained by the LibDems, and other minor parties, are uncertain, although less so than the Con/Lab split. However realistic changes to these numbers just move the location of these various zones, rather than change the zones themselves. The zone margins are also probably not precise: for example it is conceivable that one of the major two would prefer a LibDem coalition to having a majority of only 1.

One final point about the role of the SNP. If the SNP seat count was much smaller (which seems very unlikely to happen), we would have a symmetrical position. In essence it is a three party model, with the LibDem position being ‘forced’ on the boundaries because the largest party has the option of instead going with the minor parties. A large SNP block creates an additional zone, where Labour can only govern with SNP support. Now suppose you are indifferent between the policies of Labour and the SNP, but are inclined to vote SNP because you think that would give Scotland more influence. Is there a downside? There would seem to be two.

The first concern would be that some English voters might not vote Labour because they did not want Scotland to have this additional influence. You could see Conservative claims of Labour-SNP chaos as code for this. The second is that it may deprive Labour of being the largest party. However that does not matter in any constitutional sense here, because the way I have set up the problem each minor party (in terms of seats) will only form a coalition with one of the major two. (The DUP might be an exception to this rule.) It only matters to the extent that being the largest party in terms of seats might influence the LibDems choice! Interesting times.

1. All this by the LibDems to avoid a multiplier of 1.5, a simple fiscal stimulus, and instead to push a UK-is-Greece fallacy?

As Krugman has just said of pausing Greek austerity, which can be echoed for a leftist alliance in the UK, "so is it going to happen? Well, it’s the right thing to do — which tells you nothing."

2. Excellent post. I've been trying to explain this to some nationalist friends - voting SNP will make a Tory government more likely.

I'd add another point as well. As you say, some English voters will be unhappy if the SNP have some kind of power. I suspect they'd be less unhappy if Labour are at least the largest party.

1. ""@Anonymous
Excellent post. I've been trying to explain this to some nationalist friends - voting SNP will make a Tory government more likely.""

What?

The tories will be lucky to get one seat in Scotland. "SNP understanding" is as good as Labour seats. It's the seats that the tories win and Labour don't that is the issue. IF you want the tories out.

323 Lab/SNP is the same as 323 Lab

Scotland will probably be 100% anti-tory. England is where Labour need to take seats away from the tories.

Sorry but you're wrong.

Voting SNP will NOT make a Tory government more likely under any circumstance if the "understanding" is legit. The anti-tories need 323 seats for Labour to win. If anything, the SNP wave taking seats will make it more likely that tory-dem can't happen. SNP will be taking lib-dem seats.

3. I think the "Lib Dems Decide" category is a bit misleading, as I imagine it's unlikely that the Lib Dems would support a Conservative government that made any significant concessions to UKIP for their support. Other than that, quite an interesting way of viewing the election. Certainly goes some way to undermining the inevitable "Miliband government is illegitimate" line.

4. The chances of an outright Tory majority could be higher than people think. William Hill have the tories at 15/2 for an overall majority and 12/5 for 301-325 seats. These look like value bets to me. There isn't the anti tory vibe required to keep them out in my view. It's a first past the post system thats supposed to produce result. It's looking bleak at this stage for progressive forces. If you can take the next three weeks off work grab a red rosette get to your nearest marginal and get stuck into the ground war. Writing or reading blogs is not going to help at this point.

5. I haven't followed this as closely as I usually do but wasn't there also a distinct possibility of one party (Labour? SNP?) saying they would not enter a coalition, but saying instead there would be the possibility of a Minority Government? What then?

Secondly, the LibDems have come second to both Tories and Labour all over the place for years. You may be right that UKIP will not get so many seats, but they will surely still steal away a lot of votes from both Conservatives and Labour. The LibDem faithful have always been a pretty loyal bunch, and when LibDems get in somewhere they do tend to keep it for a long time. Last election there were a lot of people saying right up to polling day that they would vote LibDem, but in the end backed away perhaps because they couldn't imagine the LibDems as a party of Government. Well, now the LibDems *are* seen as a party of Government, so perhaps will gain more of those lost votes from last time, keep a lot of their faithful, and end up winning a lot more seats, far more in line with their share of votes cast since with the UKIP effect the number of votes needed per constituency to be elected is likely to be smaller than in 2010.

Basically, UKIP lowers the bar, allowing the LibDems to jump over it more often, giving them a seat share far more in line with their vote share.

1. I fear that any optimism over the Lib Dem's performance is sadly misplaced. Firstly, their USP has always been as a "reasonable party of opposition", which was lost when they joined the government. Indeed, in the coalition, they have reaped none of the benefits of government and all of the downside. Secondly, they have always been the beneficiaries of tactical voting, and are much less likely to receive that as a result of (1). Indeed, they had picked up a lot of Labour voters who were disillusioned with Blair, who are not best pleased to find themselves "supporting" a government of the right.

2. We shall see. There was no discernible UKIP effect before, and I suspect Cameron and Co are more worried about this than are the other parties, although Labour coud lose out too. Are you suggesting that the LibDems are not least likely to lose votes to UKIP?

I'm not looking at other policies or views, just this one factor which is perhaps the main change from 2010 for the Nation as a whole. As the party that came second in more Labour seats than the Tories, and second in more Tory seats than did Labour, who other than the LibDems will benefit so much from UKIP chipping away at the two big parties' core votes? The Cameron meme is that Labour will benefit most, but Labour loses votes to UKIP too which adds to the general feeling that Cameron is really out of touch with anything except his own propaganda machine.

6. Great analysis and explanation. The only insight I can offer to the Lib Dem decision making process is from their party conference - very democratic involving votes by party members, but not rapid by any means.

I'd imagine a leadership change would take place 'outside' of this usual process - would it be popular with their wider membership?

7. ""@Anonymous
Excellent post. I've been trying to explain this to some nationalist friends - voting SNP will make a Tory government more likely.""

What?

The tories will be lucky to get one seat in Scotland. "SNP understanding" is as good as Labour seats. It's the seats that the tories win and Labour don't that is the issue. IF you want the tories out.

323 Lab/SNP is the same as 323 Lab

Scotland will probably be 100% anti-tory. England is where Labour need to take seats away from the tories.

Sorry but you're wrong. Voting SNP will NOT make a Tory government more likely under any circumstance if the "understanding" is legit. The anti-tories need 323 seats for Labour to win. If anything, the SNP wave taking seats will make it more likely. SNP will be taking lib-dem seats.

1. If Clegg uses the criteria of party with most seats to decide who to form a coalition with, SNP rather than Labour could matter. Clegg's logic makes little sense, but that will not stop him doing it.

Unfortunately because of spam with embedded links (which then flag up warnings about the whole site on some browsers), I have to personally moderate all comments. As a result, your comment may not appear for some time. In addition, I cannot publish comments with links to websites because it takes too much time to check whether these sites are legitimate.