Originally uploaded by JohnConnell
The electoral system used for elections to the Scottish Parliament, the so-called additional member system, is often misunderstood by many voters.
All the other electoral systems used here (FPTP for Westminster, STV for the councils, and d’Hondt for Europe) have the property that each voter gets only one ballot paper, but AMS gives you two, which leads to confusion about their relative importance.
The answer depends on where in Scotland you live. In Glasgow and the West of Scotland, Labour are dominating and thereby winning the majority of constituencies.
As an example of this situation, let’s look at the Eastwood constituency within the West of Scotland region at the last elections back in 2007. The result was that Labour’s Ken Macintosh won Eastwood, and that Labour got 0 list seats (but 8 out of 9 possible constituency seats), SNP 4 list seats (in addition to 1 constituency), the Tories 2 list seats and the LibDems 1 list seat, for a total result of Lab 8, SNP 5, Cons 2 and LD 1. The list MSPs included two candidates from Eastwood (the SNP’s Stewart Maxwell and the Tories’ Jackson Carlaw), so this constituencies in effect saw three of its candidates elected.
Now let’s look at some different scenarios:
- What if the Tories had won Eastwood? The regional results would have been Lab 7, SNP 5, Cons 3 and LD 1. This would have been achieved by Labour’s Ken Macintosh being replaced in the Scottish Parliament by the Conservative Philip Lardner from the Cunninghame North constituency.
- What if the SNP had won Eastwood? The regional results would have been Lab 7, SNP 5, Cons 3 and LD 1. Yes, that’s right: The over-all effect would have been another Tory MSP, and the people involved would have been the same as above.
- What if the LibDems had won Eastwood? Guess what? The same again – the only difference being that the LibDem MSP would have been Gordon MacDonald instead of Ross Finnie.
It’s therefore clear that the only important question was whether Labour won Eastwood or not, and that it would have made sense for SNP and LibDem supporters to have voted Conservative if they wanted to hurt Labour.
So what about the effect of the list vote? Let’s again look at some scenarios, assuming that Ken Macintosh had won the constituency. Remember that the baseline scenario is Lab 8, SNP 5, Cons 2 and LD 1.
- What if Labour had got 10,000 list votes more (without taking votes from the other parties, i.e., through an increased turnout)? No effect whatsoever. What if they had got 10,000 votes less? No effect, either.
- What if the SNP had got 10,000 list votes more? No effect (they would have needed quite a lot more to take a seat from the Tories). What if they had lost 10,000 votes? They would have lost a seat to the Tories.
- What if the Tories had got 10,000 list votes more? They would have taken a seat from the SNP. And if they had lost 10,000? No effect.
- What if the LibDems had got 10,000 list votes more? They would have taken a seat from the SNP. And if they had lost 10,000? They would have lost their seat to the Tories.
What we are seeing here is the effect of the SNP having won the 7th and last list seat, and the Tories being due the next list seat, so when somebody wins a seat, it comes from the SNP, and if they lose one, it goes to the Tories.
However, the overall picture is clear: A vote on Labour is completely wasted when it comes to the list votes, whereas it makes a difference for all the other parties.
The conclusion is therefore clear. If you were a voter in Eastwood back in 2007, in retrospect these were your choices: For the constituency vote the choice was between Labour and the Conservatives, and for the list vote the only relevant choices were Conservative, SNP, LibDem or perhaps Green.
Given the notional results for 2011, it’s very likely that Labour will win so many constituencies that a list vote for them will again be completely wasted. This again means that the only real decision for the constituency vote is whether to vote for or against Labour. The voting recommendation for 2011 is therefore the same as for 2007.
On the other hand, in most Scottish electoral regions, one party does not win the vast majority of constituency seats, and the effect of this is that the constituency vote never matters. As an example, let’s look briefly at the results from the North East Scotland region.
The results in 2007 were SNP 6 constituency seats + 2 list seats = 8 seats in total, Labour 1 + 2 = 3, LD 2 + 1 = 3, and Cons 0 + 2 = 2. Here are the different alternative scenarios:
- If the SNP had taken one constituency seat from Labour, the result would still have been 8/3/3/2.
- If Labour had taken one seat from Labour, again the result would have been the same.
- The LibDems taking a seat from the SNP? No change.
- The Tories doing it? Again the same.
This is actually what the AMS system is supposed to do: The constituency vote is supposed to allow people to decide who they want to represent them (as opposed to which party), whereas the list vote is supposed the allow people to decide which party the want to represent them (as opposed to which person).
It’s just that it doesn’t quite work like that when one party is winning too many constituencies within a region, which is why the constituency vote in Eastwood is important for deciding how many MSPs Labour will get in total, rather than just on deciding on the personal merits of Ken Macintosh.
The vast majority of members would remain elected in single-member constituencies using the Australian form of preferential voting the British call the alternative vote in which voters rank candidates in order of preference with only 10-15 of the total MPs elected from party lists. The compensatory regions would consist of very small multi-member clusters formed from a group of constituencies and one or two additional list seats.
in favour of AMS is that it gives the advantages of proportionality ..