A strong headline and one that can fast-track a blogger into embarrassing disrepute but I’ve checked, double-checked and triple-checked. (And, not only that, but I was informed that the super-reliable Will Paterson came to the same conclusion a full two months ago. Will’s discovery was entirely unbeknownst to me over this weekend as I familiarised myself with the Scottish Parliament boundary changes, but I’ve written this post now so I might aswell publish it too)
The result of a report conducted by David Denver of the University of Lancaster (as found on the BBC) had the notional result of the 2007 Scottish Parliament elections (per the new boundaries) as follows (difference to actual result in brackets):
SNP – 46 (-1)
Labour – 44 (-2)
Conservatives – 20 (+3)
Liberal Democrats – 17 (+1)
Greens – 1 (-1)
Margo – 1 (-)
So, the Conservatives are seemingly the big winners from the boundary changes and Labour are the biggest losers. The SNP hasn’t done so badly either, doubling its majority despite losing a seat.
These figures have been widely quoted by politicians and activists alike. The Herald ran the news with the headline “Boundary changes may help Tories at Holyrood” and the BBC’s Brian Taylor went into a lot of detail over what the report may mean for 2011, one of the main talking points being the increased Tory representation and “a slightly improved net lead for the SNP over Labour”.
However, there is an error in the workings for the allocation of the 7 South MSPs that overstates the Conservatives by 1 seat, understates Labour by 1 seat and results in a notional SNP majority of 1 MSP, the same majority as the ‘actual’ 2007 result.
The South of Scotland Regional MSP allocation per this analysis is as follows:
SNP – 3 regional seats
Lib Dems – 2 regional seats
Conservatives – 1 regional seat
Labour – 1 regional seat
I am not trying to show anyone up with the below but I do want to recalculate the above in order to clarify what the true notional result is
The South region’s First Past the Post seats, with ‘notional’ winners, are as follows:
Ayr – Conservative
Carrick, Cumnock & Doon Valley – Labour
Clydesdale – Labour
Dumfriesshire – Conservative
East Lothian – Labour
Ettrick, Roxburgh and Berwickshire – Conservative
Galloway and West Dumfries – Conservative
Kilmarnock and Irvine Valley – SNP
Midlothian South, Tweedale and Lauderdale – SNP
Conservatives – 4
Labour – 3
SNP – 2
The split of notional 2007 regional votes, as per the report on the BBC, is as follows:
Labour – 81,326
SNP – 80,668
Conservative – 62,972
Lib Dem – 28,001
Greens – 9,494
‘Other’ – 20,371 (breakdown is irrelevant as neither the BBC report nor my recalculation gives these Others a seat – these votes will therefore be disregarded from now on)
Applying the d’Hondt formula in order to allocate the 7 South Regional MSPs gives the result as per below (see Wikipedia page to understand the slightly complicated allocation method):
1st Regional MSP – Lib Dems (28,001 = 28,001/1)
2nd Regional MSP – SNP (26,889 = 81,326/3)
3rd Regional MSP – Labour (20,332 = 81,326/4)
4th Regional MSP – SNP (20,167 = 81,326/4)
5th Regional MSP – Labour (16,265 = 81,326/5)
6th Regional MSP – SNP (16,134 = 80,668/5)
7th Regional MSP – Lib Dems (14,001 = 28,001/4)
The Conservatives won 4 FPTP seats so their regional statistic of 12,594 (= 62,972/5) was too low to win a regional seat. However, the report on the notional numbers nonetheless gives the Conservatives a regional seat at the expense of Labour.
As I said earlier, I’m not looking to show anyone up with the above but given how often claims and counter-claims will be thrown around between the various parties over the coming election campaign, it is best to start with the truth.
That truth seems to be that that boundary changes give the SNP a notional majority of one seat over Labour, not the two that most believe from September’s headline and has been used in the comments section of this blog. Furthermore, the Conservatives only pick up an extra two seats, a significant increase in its own right, but not the more eye-catching three seats increase that many currently believe.
So, as I admitted at the top, Will got there first, but I’m sure he would agree that there is no harm in reminding those that are interested where it is we are starting from, and what election claims are and are not valid, as the 2011 Holyrood campaign approaches.
UPDATE – Note also that these notional results has the SNP to win the new constituency of Almond Valley by only 4 votes, their only notional victory in all of Lothian.
Were this seat to be won by Labour, and all other results remain equal, the SNP would not win an extra seat in the list so it would be a genuine loss of an MSP from SNP to Labour meaning it would be Iain Gray who would head up the largest party.
The SNP won the 2007 ‘actual’ election by 46 votes thanks to Kenneth Gibson’s slender victory in Cunninghame North and they win the 2007 ‘notional’ election by an even more slender 4 votes.
(Note that the Cunninghame North majority in the notional analysis is 40 votes)
#1 by Gaz on November 16, 2010 - 6:01 pm
Good analysis Jeff.
However, the real error in these notional results is that they are based on the council election results of that year and therefore contain huge assumptions about people following the ticket rather than respective candidates and projections as to how the vote in multi member wards break down.
Why the analysis did not make use of the actual ballot box results for the Scottish Parliament elections which have been available for some considerable time is beyond me. Doing this would allow an extremely precise estimate of notional results to be produced.
As an example, I have looked into the claimed notional SNP majority of 4 for Almond Valley. Using the ballot box results from 2007 gives the SNP a majority more like 200. This is still a good bit smaller than the 870 from the old Livingston constituency but a good deal more comfortable than David Denver’s figures suggest.