Tactics 1

This is going to be the first of three interconnected blogs.

The first of these might be seen as raking over old coals, but it concerns a matter I consider important in several respects. First of all to get right just how and why the SNP lost their majority in May. Secondly to argue for the creation of an independence movement, which I hope the Convention next Saturday (14th January) will take forward. And lastly, to make a few points about the development of an elite group of independence bloggers.

I am, as I often am, grateful to the Rev Stewart Campbell for this tweet he put out just before the bells at New Year.

This refers to the “furore” (I use the word loosely) prior to the Holyrood election of whether independence minded voters should use their constituency and list votes to vote for the SNP in line with #BothVotesSNP – or, while using their constituency vote to support the SNP, they should vote for someone else on the List. Various established independence bloggers, such as Rev Campbell on Wings, Peter Bell, James Kelly and GA Ponsonby were quite clear that # BothVotesSNP (indyref2, as well as their own blogs) was the route to follow, while various rag tags such as me, squeaked from the gutter that this was just wrong.

The specific problem with this is that while Campbell is correct that the Green List vote increased by as much as 50%, it was from 4.4% in 2011 to 6.6% last May. In that regard four extra seats is arguably not a bad return, but to expect lots of List seats on the basis of a 2.2% increase in their vote is at best silly, arguably misleading and perhaps serves only to divert our attention from something else. Moreover, the main reason for the loss of majority was the SNP’s overall loss of 12 seats on the List, so we need to examine this rather than just revert to what seems obvious, in many regards because it has been voiced so often.

Of course, the whole matter is largely academic (at least until the next Holyrood election), but it seems that some people cannot let it go. One example is “The ‘Tactical Voting’ groupies are in denial” (http://indyref2.scot/the-tactical-voting-groupies-are-in-denial) by GA Ponsonby. While it’s a bit dated (7th May), there are two specific reasons for revisiting it.

First because its argument is demonstrably false, but seems to be becoming the accepted wisdom, even though. Secondly because of another article on the same website, published on 26th November by Peter Bell, “A dissenting voice” (http://indyref2.scot/a-dissenting-voice)., which, it seems to me is symbiotic to Ponsonby’s argument of six months ago, but ignores its problems and in fact, if anything, compounds them. Most importantly it concerns the Independence Convention Conference on 14th January.

Ponsonby argues that the proposition was made that the SNP could secure a majority purely because of the constituency seats they won. Frankly, I don’t remember that argument being made at all, and I would very much like to see an instance of it. Certainly the polls suggested a very strong constituency showing, but with “local issues” always were possibly going to intervene, so winning the lot was never a certainty. Moreover, they would have had to win 65 of 71 constituency seats (or 92%) to have a very slim majority of one. In fact, they won 59, or 83%, which was an increase of 6 constituency seats.

However, I, and others, had argued for some time that, based on a series of opinion polls prior to the election indicating a high degree of success in the constituency section, that it was hard to see the SNP picking up too many list seats. Ponsonby’s response to this is

“We can never know how many SNP voters were influenced by the claim that their List vote was wasted, what we do know is that the fall in the nationalist List vote cost the party List seats and ultimately a majority.”

This quite simply is not true. Let’s see why.

The table below sets out the results for each electoral region. I think the association between winning constituency seats and not winning list seats is obvious by inspection.

Region

Constituency seats won by SNP

Regional List seats won by SNP

  1. Glasgow

9 of 9

0

  1. Central Scotland

9 of 9

0

  1. North East Scotland

9 of 10

0

  1. Mid Scotland & Fife

8 of 9

0

  1. West Scotland

8 of 10

0

  1. Lothian

6 of 9

0

  1. Highlands & Islands

6 of 8

1

  1. South Scotland

4 of 9

3

 

In the first two regions – Glasgow and Central Scotland – the SNP won all the constituencies and none of the list seats. In regions 3 and 4 – North East Scotland and Mid Scotland & Fife – they won all but one seat in each, but again no list seats. In West of Scotland they won all but two constituencies, but still no list seats. It seems we need to get down to somewhere about winning ‘only’ two thirds or three quarters of the constituencies before winning list seats becomes possible, Its only once we get to the last region – South of Scotland – that the SNP win a significant number of list seats – in this case three. But they had won only four of the nine constituency seats.

There is an interesting comparator here to 2011 in Glasgow, where the SNP won “only” five constituency seats, while Labour won four. When it came to splitting up the list seats, the SNP took two while Labour took three, with the Tories and Greens getting one each.

The lesson from this must be, the more constituency seats you win, the fewer list seats you will win – something that seems to kick in by the time you get to winning two third – three quarters of the constituency seats.

It can of course be argued that this is what the electoral system we have is supposed to do – to give us a Parliament which reflects the overall vote, unlike the Westminster constituency-based system, in which it is “to the victor the spoils”, which produces an outcome where with 37% of the vote you can form a majority government. In Scotland in contrast, with 46.5% in the constituency vote, you won’t be forming a majority government. But as someone who has argued for an electoral system in which the outcome is proportional to votes cast, to be critical of this seems very deceitful.

But Ponsonby’s claim is stronger than that. His claim is that “the fall in the nationalist List vote cost the party List seats and ultimately a majority”.

It does have to be admitted that #BothVotesSNP did not hold up. In the constituency section nationally they won 46.5% of the vote, but on the list vote they won only 41.7%, so there was ‘leakage’ between constituency and list votes. But how much did this matter? Specifically, is Ponsonby correct when he argues “the fall in the nationalist List vote cost the party List seats and ultimately a majority”? In fact, the following figures show the fallacy in this claim.

This was how the allocation of list seats in Glasgow played out – the winning party as each seat is allocated is shown in red.

Party

List vote

1

2

3

4

5

6

7

8

9

10

SNP

111101

11110.1

11110.1

11110.1

11110.1

11110.1

11110.1

11110.1

11110.1

11110.1

11110.1

Labour

59151

59151

29575.5

19717

19717

19717

14787.75

11830.2

11830.2

11830.2

9858.5

Con

29533

29533

29533

29533

14766.5

14766.5

14766.5

14766.5

9844.333

9844.333

9844.333

Green

23998

23998

23998

23998

23998

11999

11999

11999

11999

7999.333

7999.333

LD

5850

5850

5850

5850

5850

5850

5850

5850

5850

5850

5850

 

Regions only have 7 list seats, but I have developed this to the (fictitious) 10th seat to show just how difficult it is for a party to win list seats in our electoral system, when they have cleaned up in the constituencies. With their regional list vote – more than Labour and Tory combined – there would have had to be 10 list seats in Glasgow before the SNP, even with that size of list vote, would have taken a list seat. The reason for this is that they won all 9 constituencies in the Glasgow Region, so, with the voting system, their vote, from round 1, was divided by 10 (9 constituencies +1).

But more importantly for Ponsonby’s argument, the 7th seat was won with a (modified) Conservative vote of 14766.5 (their list vote too being divided by the number of seats won +1). The SNP vote is 11110.1 (their actual list vote divided by 10 – 9 constituencies + 1). Thus, for the SNP to have won even the last regional list seat in Glasgow, they would have had to have achieved a list vote of 36,564 greater than it actually was (the Conservative modified list vote 14766.5 minus the SNP modified list vote 11110.1 times 10 to allow for the 9 constituency seats already won). Thus, their actual list vote would have had to be more than 147,665 to beat the Conservatives to the seventh seat. But their constituency vote was 128,443, so even if all their constituency voters had voted SNP in the list, they would still have been 19,222 short by the 7th round. To sustain Ponsonby’s #BothVotesSNP argument, therefore, the SNP vote would have to have been still higher in the constituencies than it actually was.

In West of Scotland, the final list seat was taken by the Conservatives with a modified list vote of 17,822. In that region the SNP won 8 of the 10 seats. Thus the SNP list vote would have had to be more than 9 times (8+1) the Conservative modified list vote for the 7th seat – 160,398. In fact, the SNP list vote was only 135,827. But as the combined constituency vote was 148,660 it is clear not all their constituency vote followed the #BothVotesSNP advice. However, to win even just that 7th seat, the SNP vote would have had to be 160,398, and the question can be posed in West of Scotland, just like Glasgow, where were these votes going to come from, as what was needed was more than the constituency vote?

Ponsonby’s contention is that it was “tactical voting groupies” urging tactical voting on the SNP voter who caused the SNP to fail to do better than the list, which they would have done had the #BothVotesSNP advice been followed. Taking both these regions as examples, were Ponsonby’s argument true, then the SNP constituency vote would have had to be higher than it was in both cases. To have picked up even the last regional list seat, the SNP would have needed to increase their vote by almost 15% in Glasgow and almost 8% in West of Scotland.

There are therefore two reasons why the SNP lost their majority in May

  1. The electoral system worked as it was designed to do, to balance up in the list system the excesses of the “winner takes all” constituency section in order to give a more proportional outcome. If we average out constituency and list votes, the SNP took 44.1% of the vote and won 48.8% of the seats. The system did its job.
  2. The SNP did not win enough votes to overcome the electoral system. As we show above, in Glasgow and West of Scotland regions, even if #BothVotesSNP had been adhered to, they would still have failed to win even the 7th (and last) regional seat.

Thus rather than continuing to fight a dispute that they have not, and cannot, win, perhaps luminaries such as Stewart Campbell, GA Ponsonby, Peter Bell and James Kelly would do better to use their undoubted skills, reputation and reach to support the development of a wider independence movement. However, as we shall explore in part 2, this seems unlikely.

 

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s