by Stephen Fisher

In the Scottish Parliament elections yesterday the SNP lost 6 seats and their majority because their share of the list vote fell by 2 points. They ended up with 49% of the seats from 42% of the vote. Meanwhile, at the Welsh Assembly elections Labour’s vote share dropped by 5.4% points but they lost just one seat, finishing with 48.3% of the seats from just 31.5% of the vote.

Why was the Welsh system so much more generous to Labour than the Scottish system was to the SNP?

The electoral systems for Scotland and Wales have the same formal structure. They are both additional member systems with first-past-the-post constituencies and regional list top-up seats that are allocated taking into account the outcome of the regional constituency seat tallies. The Scottish system is more proportional than the Welsh one because it has a higher ratio of list seats to constituencies and more seats overall. The Scottish Parliament has 129 seats, 43% of which are list seats. The Welsh Assembly has 60 seats, only a third of which are in the proportional top up section.

Labour won 27 of the 40 Welsh constituencies. That alone meant that they would end up being considerably over-represented relative to their 31.5% of the list vote.

The fact that the top-up seats are allocated on a regional instead of a national basis makes the system still more disproportional. Labour’s support is concentrated in the three South Wales regions and North Wales, helping them win all but three of the constituencies in those regions in 2011. By contrast they only won one constituency in Mid and West Wales this year and last time. This meant that they were eligible for 2 list seats (both this time and last) in the region despite being substantially over-represented elsewhere in Wales.

The only change for Labour’s seat tally from 2011 was due to the loss of the Rhondda constituency to Plaid Cymru’s leader Leanne Wood on a massive 24.2% swing. Since Labour again won all the other constituencies in South Wales Central they were still too over-represented in the region to be eligible for any list seats in compensation.

Rhondda was the only constituency to be won by a different party than in 2011. This is striking given that the scale of the changes in the share of the constituency vote (Con -3.8, Lab -7.6, LD -2.9, PC +1.3). The largest swing between any two of these parties was the 5.7 point swing from Labour to the Conservatives. This should have been enough for the Tories to take Cardiff North, but the swing there was just 2.3 points. Similarly the 1.9 Labour to Plaid Cymru swing nationally would have seen Llanelli change hands, but in fact it was 0.5 points in the constituency. The Lib Dems could reasonably have expected to take Cardiff Central based on the 0.5 point Labour to Liberal Democrat swing overall, but in the event there was a 1.5 point swing in the opposite direction. No other seats would have been projected to be lost by any party if the changes in the share of the vote had been the same everywhere.

Labour did well to hold on to the three constituencies they did against the national trend. Had they not done Labour would most likely not have been compensated with any additional seats on the list system. At best they might have picked up one more.

By beating the national swing in three constituencies and only being beaten in one, Labour ended the night with 29 seats out of 30 instead of 27. This difference has given Labour more governing options and so more power in forming an administration than they otherwise would have done.

However, the generosity of the Welsh Assembly electoral system to Labour came at the cost of the parties that would, perhaps, be most likely coalition partners. As usual small parties suffered most from the disproportionality of the system. The Liberal Democrats won just 3% of the seats from 6% of the votes while Greens got no seats despite winning 3% of the vote.