In the German system, there is the concept of overhang seats. In the German system, the seats are allocated on a proportional basis considering the total size of the Bundestag and then the constituency results come into play. To look at this, consider the result for Scotland Central at the May 2003 elections to the Scottish Parliament. This has 10 constituencies and 7 top-up Members of the Scottish Parliament.
In a Germanic system, we would look at what would happen if 17 seats were allocated on a proportional basis:
- Labour - 8
- Scottish National Party - 4
- Conservative - 1
- Scottish Socialist Party - 1
- Scottish Senior Citizens Unity Party - 1
- Liberal Democrat - 1
- Green - 1
In this region, Labour won 9 of the constituencies, with Falkirk West won by an Independent, Dennis Canavan. Under the German system we subtract the number of constituencies to determine the number of regional MSPs:
| Party | Entitled to | Constituencies | Regional |
| Labour | 8 | 9 | -1 |
| Scottish National Party | 4 | 0 | 4 |
| Conservative | 1 | 0 | 1 |
| Scottish Socialist Party | 1 | 0 | 1 |
| Scottish Senior Citizens Unity Partye | 1 | 0 | 1 |
| Liberal Democrat | 1 | 0 | 1 |
| Green | 1 | 0 | 1 |
| Independent | 0 | 1 | -1 |
Hmm, so we need to take 1 constituency from Labour and Canavan. Under the German system, Scotland Central would be expanded from 17 to 19 MSPs, 9 of them regional, and Labour and Canavan each having an overhang seat.
Under the British system, AMS is an additive system, and begins with the number of constituencies won before deciding how many regional MSPs each party gets. It is possible - as for Labour in Scotland Central - for a party's vote to "saturate" (for want of a proper psephological term), where it does so well in the constituencies that it gets equal to, or more than, the number of seats it would be entitled to if all seats were allocated proportionally. If we look at Lothians in 2003, we get:
| Party | Constituencies | Regional | Total |
| Labour | 6 | 0 | 6 |
| Scottish National Party | 0 | 2 | 2 |
| Conservative | 1 | 1 | 2 |
| Green | 0 | 2 | 2 |
| Liberal Democrat | 2 | 0 | 2 |
| Independent | 0 | 1 | 1 |
| Scottish Socialist Party | 0 | 1 | 1 |
Here, the Independent is Margo MacDonald, who ran a "party list" with just her name on it.
In 2003, Labour lost two constituencies compared to the May 1999 election - Edinburgh South (to the Liberal Democrats) and Edinburgh Pentlands (to the Conservatives). What effect did these have on the result?
First assume Labour held Edinburgh South:
| Party | Constituencies | Regional | Total |
| Labour | 7 | 0 | 7 |
| Scottish National Party | 0 | 2 | 2 |
| Conservative | 1 | 1 | 2 |
| Green | 0 | 2 | 2 |
| Liberal Democrat | 1 | 1 | 2 |
| Independent | 0 | 1 | 1 |
In reality, the Scottish Socialist Party won the seventh and final regional seat. If Labour - rather than the Liberal Democrats - had won Edinburgh South, then it would be the Liberal Democrats picking up that seat. In total, it would appear that Labour has gained a seat at the expense of the Scottish Socialist Party.
Now assume that instead, Labour held Edinburgh Pentlands:
| Party | Constituencies | Regional | Total |
| Labour | 7 | 0 | 7 |
| Scottish National Party | 0 | 2 | 2 |
| Conservative | 0 | 2 | 2 |
| Green | 0 | 2 | 2 |
| Liberal Democrat | 2 | 0 | 2 |
| Independent | 0 | 1 | 1 |
Again, a similar effect. This time the Conservatives (rather than the Greens) pick up the sixth regional seat, while the Greens (rather than the Scottish Socialist Party) pick up the seventh and last regional seat. So, again we have the overall effect of Labour gaining a seat at the expense of the Scottish Socialist Party.
But now we consider what if Labour held both Edinburgh South and Edinburgh Pentlands. This then gives us:
| Party | Constituencies | Regional | Total |
| Labour | 8 | 0 | 8 |
| Scottish National Party | 0 | 2 | 2 |
| Conservative | 0 | 2 | 2 |
| Green | 0 | 2 | 2 |
| Liberal Democrat | 1 | 0 | 1 |
| Independent | 0 | 1 | 1 |
So, what we see is Labour gaining 2 seats at the expense of the Liberal Democrats and the Scottish Socialist Party. Hence, if you were a Scottish Socialist Party supporter in Edinburgh Pentlands, the best thing you could do to help your party would be to vote Conservative in the constituency section.
This might seem counter-intuitive. But suppose the House of Commons used AMS, and you were a minor party supporter in North East England. It would be reasonable to assume that Labour will pick up the overwhelming majority of the constituencies, so much so that their vote would saturate and they would get no regional MPs.
You would then reasonably assume that the Conservatives and Liberal Democrats will be picking up the regional seats. So, if you lived in a marginal, the best thing you could do would be to vote for whichever one of these would oust Labour, and hope that this reduces their number of regional MPs by one, and that your party would pick up a regional MP in return.
If we look at allocsting all of Lothian's regional seats in a Germanic style, we get:
- Labour - 4
- Scottish National Party - 3
- Conservatives - 2
- Green - 2
- Liberal Democrat - 2
- Independent - 2
- Scottish Socialist Party - 1
Hence, even after losing Edinburgh South and Edinburgh Pentlands, Labour is still doing 2 seats better than it would on a fully proportional system.
Note that MacDonald would have been entitled to 2 seats. This would be a problem as she is the only person on her list. In the German system, this is a rare underhang seat, and the size of the Bundestag is reduced by 1 for each underhang seat.
We could look at taking what became Labour's most marginal constituency - Linlithgow - and considering what would happen if Labour lost it to the Scottish National Party:
| Party | Constituencies | Regional | Total |
| Labour | 5 | 0 | 5 |
| Scottish National Party | 1 | 2 | 3 |
| Conservative | 1 | 1 | 2 |
| Green | 0 | 2 | 2 |
| Liberal Democrat | 2 | 0 | 2 |
| Independent | 0 | 1 | 1 |
| Scottish Socialist Party | 0 | 1 | 1 |
It would do what it says on the tin - the number of Scottish National Party seats would go up by 1, and the number of Labour down by 1. It has no effect on the number of regional MSPs.
If we then take the next most marginal Labour seat, Edinburgh Central, and allocate that to the runners-up, the Liberal Democrats, we get:
| Party | Constituencies | Regional | Total |
| Labour | 4 | 0 | 4 |
| Scottish National Party | 1 | 2 | 3 |
| Conservative | 1 | 1 | 2 |
| Green | 0 | 2 | 2 |
| Liberal Democrat | 3 | 0 | 3 |
| Independent | 0 | 1 | 1 |
| Scottish Socialist Party | 0 | 1 | 1 |
It is now the Liberal Democrat vote that saturates, and we see no change in the number of regional MSPs.
Finally, if we take Labour's next most marginal seat, Livingston, and allocate that to the runner-up, the Scottish National Party, we get:
| Party | Constituencies | Regional | Total |
| Labour | 3 | 1 | 4 |
| Scottish National Party | 2 | 1 | 3 |
| Conservative | 1 | 1 | 2 |
| Green | 0 | 2 | 2 |
| Liberal Democrat | 3 | 0 | 3 |
| Independent | 0 | 1 | 1 |
| Scottish Socialist Party | 0 | 1 | 1 |
Effectively, the Labour losing a constituency to the Scottish National Party is balanced out by gaining a regional seat from them. And this is what happens when party's votes aren't saturating - namely that winning or losing constituencies doens't impact on the final result.
No comments:
Post a Comment