When I title a post, like many people I want it to convey a sense both of what the post is about and how I feel about the topic. Today’s title should convey to you a sense of exasperation.
In evidence today to the Commission on Parliamentary Reform, the two living former Labour First Ministers of Scotland were invited to weigh in. One of the areas they touched on was reforming the electoral system. In Jack McConnell’s case, it was more about tweaking the existing system. He had a few criticisms of the party list aspect of the Additional Member System (AMS) – primarily to do with feeling that it perhaps gives too much power to parties rather than the electorate. He further suggested term limits for list MSPs and floated the idea of banning people from standing on both the list and constituency ballots – often referred to as “dual candidacy”.
Term limits is quick and easy to respond to: No. Just no. There may – and that’s a big may – be merit in the idea of putting term limits on parliamentarians in general. Applying them solely to list MSPs is utterly bizarre. A “safe seat” may not seem to exist in post-referendum Scotland, but we know single member constituencies generally tend to produce safe seats. In such cases, the same argument about power of party over electorate applies. For term limits, it should be neither or both groups of MSPs, not just one.
The criticisms of the list system require a bit more thought. These actually form part of a widespread grumbling (at least for someone like me who lives on Scots Politics social media) about how AMS works. For different reasons, many SNP supporters (and at least one MSP…) online have been casting doubt on the legitimacy of list MSPs. Despite the fact that almost every single one of the SNP’s 1999 MSPs (including a certain Nicola Sturgeon) would not have been elected had a ban on dual candidacy applied, some have been calling for this. The logic seems to be that failing to win in a constituency constitutes a “rejection” by the electorate that should therefore disqualify that candidate from office entirely.
This is bizarre on a number of levels. For one, it would have a serious effect on smaller parties. Any such party seeking to stand at both constituency and list level would need to stand a huge slate of candidates that it may not actually have the membership or the money to support. Given part of the benefit of proportional representation is allowing parties with significant but not concentrated support to win seats they couldn’t under First Past the Post, penalising them for standing on both ballots is ridiculous.
That point about FPTP is important, because to win a seat in FPTP you don’t need a majority of the vote, you just need to win the most votes. That means there have been cases of MSPs elected on the list winning a greater share of the vote in a constituency than a constituency MSP from elsewhere in the same region. For example, in 2011, Labour’s Sarah Boyack was narrowly defeated by the SNP’s Marco Biagi on a close 32.7% to 31.9% margin. However, Boyack was successfully elected on the list. In neighbouring Edinburgh Southern, the successful SNP candidate Jim Eadie only won 29.4%. So despite being defeated, Boyack actually had more support in her constituency than the successful Eadie did in his – how can you classify winning a greater share of voters as a rejection?
It also takes no account for relative performance. Here in Glasgow Kelvin, the Green’s Patrick Harvie came second. By the “rejection” logic, Patrick should therefore not have been elected on the list. But 24.3% is the best performance by the Scottish Greens in any single electoral area ever. And it represented us overtaking, in a Glasgow constituency of all places, the formerly absolutely dominant Labour Party, pushing them into third place. By any reckoning then, that result was an enormous success and endorsement of the Greens and of Patrick, not a rejection. By contrast, perhaps a place where the incumbent constituency MSP elected on 44% last time plummets to 15% yet is elected on the list, there’s an argument they’ve lost too much support to be legitimate.
There are other good reasons not to ban dual candidacies as well. We don’t often follow this logic here in Scotland, in part because of our continuing belief that constituency MSPs have greater legitimacy and in part because the past two elections have given a government with few list MSPs, but ministers are kinda rubbish constituency MSPs. It’s very difficult to both manage a national portfolio and provide effective local representation. List systems can therefore allow you to separate the two out – you can put your top “national” politicians on the list, whilst your backbenchers hold the constituencies and represent the most local concerns.
Failure to at least prepare for the eventuality of leading candidates not winning constituencies can have dire effects, as Labour found out in 2011. They used to have a policy of banning dual candidacies except (as in Sarah Boyack’s case) where an incumbent constituency MSP had notionally lost their constituency seat on boundary changes. This led to them losing a whole host of leading figures in their many crushing constituency defeats. This was so notably bad that when New Zealand was reviewing their electoral system (which is almost identical to Scotland, but with a single national list) a couple of years ago, Labour were a cautionary tale.
An example of banning dual candidacy can be seen in the 2011 elections to the Scottish Parliament. Only 19% of one party’s candidates stood as dual candidates (compared to 70 – 80% for the two other main parties). Many experienced members of this party lost their electorate seats and were thus unable to return to Parliament. The loss of so many experienced members has made it more difficult for the Opposition to effectively scrutinise the activities of the current majority government.
That’s not to say that the list system is perfect as is. It can be improved upon. And we can improve upon it quickly and easily in a way that we can’t with complete reform of the electoral system (my own preference). As they stand the lists are “closed”. In party list proportional representation terminology, that means parties select a list of candidates, they order them as they please, and voters simply vote for the party, who after winning n seats on the list will see the first n candidates on the list elected. Easy as pie, but this does indeed leave much to be desired in terms of accountability to the electorate.
Instead, we can “open” the lists. This allows voters to vote for a specific candidate on a party’s list, rather than just the party list in pre-approved order. Parties still pick the candidates for the list, and may still be able to present them in an order they prefer, but voters have a say over which candidates get elected. A vote for any candidate of a given party counts as a vote for that party when it comes to allocating the number of list seats. But rather than elect n candidates in order of appearance on the list, this system would elect the n candidates on that list with the most votes.
Let’s imagine then a party with six candidates on the list;
- Michael Mustard, 39 votes
- Wilhelmina White, 654 votes
- Patsy Peacock, 233 votes
- Peter Plum, 448 votes
- Gary Green, 127 votes
- Stacey Scarlet, 345 votes
Overall this party’s candidates have won 1846 votes. It turns out this entitles the party to two MSPs. Since Wilhelmina White (654) and Peter Plum (448) are the two most voted for candidates on the list, they become the MSPs, even though Michael Mustard was at the top of the list and Patsy Peacock was ahead of Peter Plum. So we can clearly see that opening the list would give voters far more say over their MSPs. If they don’t like a given candidate, even if that candidate heads the list, they can vote for someone else.
In that respect, open party list systems actually offer voters the most choice in terms of candidates within each party. In contrast to FPTP and STV, which either require or tend towards parties standing as many candidates as they think they can get elected, list systems almost always feature more candidates than can be elected. There’s still the constraint on who the party selects, but that’s the nature of party politics. Open lists crack open the electoral system enough to ensure list MSPs can claim a similar personal mandate to constituency MSPs without getting bogged down in the minutiae of more radical reforms.
TL;DR as this turned out about 400 words longer than expected – Just open the bloody lists and stop whinging about how undemocratic they are.