Entangled Alliances

It is a weak point in the theory of representative government as now organized and administered, that a large portion of the voting people are permanently disenfranchised.

So said U.S. President James Garfield, back in 1881. At the time, he was talking about an American voting system that had only recently been opened up to African-Americans and was still denied to women. But if Garfield was alive today, he might well feel compelled to say the same thing about the effects of  the first-past-the-post system on the majority of voters in the United Kingdom. In Part One of this in-depth look at electoral reform, I examined why Britain’s current system is so bad and also tackled some of the criticisms frequently directed at the main alternative: proportional representation. Now I want to take a close look at what kind of electoral reform the UK could have and in doing so answer the question: just how proportional do we want our representation, anyway?

(By the way, as I was was almost finished writing this Gordon Brown announced (admittedly not unsurprisingly) that  he was planning a consultation on voting reform. Thus this post is transformed from a look at a complex, brain-achingly dense topic into a hot-off-the-press must-read. No, really!)

Party List PR

“Party list” is the most common form of proportional representation. It’s fairly simple: the country is divided up into multi-member districts. Each party draws up a list of candidates to present to the electorate in each district. That party gets allocated seats in the district in proportion to the amount of votes it gets. So, in theory, if a district had ten seats up for grabs and a party got half the vote there, then that party will get five of the district’s seats. They’ll fill them by taking their top five candidates from that district’s list. Depending on the type of party list PR, voters are either allowed to vote for the party only -”closed list” or for the individual candidates – “open list”.

Party list is an excellent system in that it’s as close to a completely proportionate system as you’re likely to see. The division of seats really does reflect the will of the electorate, and most of the time no-one’s vote is wasted.  Plus, as far as voting systems go it’s really simple and as I noted in part 1, simplicity matters.

Of course, it’s extreme proportionality means that it gets criticised for leading to weak, coalition-based governments which have to contend with dozens of small parties getting their say. This criticism clearly hurts, because a pure party list system is certainly not the favourite to get the nod in the near future – its advocates in the UK are few and far between. Critics point to Israel as an example of its failures- Israel’s use of party list PR has led to a series of shaky, paralysed coalition governments that suffer from the inclusion of small extremist parties. The current government, which includes an awkward pick n’ mix of small right wing parties, is a classic case in point. However, those same critics tend to ignore the fact  that countries like Norway, Sweden, Finland etc use party list without such problems, resulting in broadly successful coalitions.

It also matters whether you have a closed-list or an open list. A closed list system  could be criticised for allowing the party, not the people, to choose their candidates. Those selected thus become beholden to the party for their position on the list, making it pretty hard to dissent from the party line.  On the other hand, picking your own candidates is a good opportunity for a party to ensure that women and minorities are well represented. An open list, meanwhile, leads to the exciting prospect of voters being able to vote for a junior party member over a senior figure in the same party.  Great for political drama, less good for party unity, It’s worth taking a moment to imagine the, uh, political thrills that such a system would create in the current British climate: imagine, for example, if disillusioned voters were able to stick with their usual party but vote for  some new blood at the expense of senior figures caught up in the expenses scandal.

STV

STV stands for Single Transferable Vote, which isn’t very helpful. It’s a voting system the intricacies of which require  degrees in political science and maths to fully understand, so here’s the Cliff Notes version. Under STV, a country is drawn up into multi-member constituencies. Voters are presented with a list of candidates – there may be more than one of the same party – and asked to order them by preference – 1st choice, 2nd choice, 3rd choice etc. To be elected, a candidate has to get over a certain vote threshold, which is normally decided by, more or less, dividing the number of votes by the number of seats available in that constituency. Still with me? Good. Let’s say there’s 6 seats up for grabs in the constituency. If six candidates get the magic threshold on the first vote count, then great. If not, then the second preferences on the ballots for those who did make it are taken into account. Then third preferences… and so on, until six candidates have got enough votes. It’s slightly more complicated than that, but you get the picture. Or not.

photo credit: artesea

STV is clearly very similar to an open party list system and thus attracts the same kind of praise and criticism. The amount of seats is extremely proportional to the amount of votes, but because of this critics claim that it results in weak, unstable coalitions. However, a quick look at the electoral history of Ireland, which uses STV, shows this to be untrue. If STV governments were always unstable and weak, then you’d expect frequent general elections as coalitions collapse. Yet in the past sixty years, Ireland has had seventeen general elections – which, believe it or not, is only one more than the UK has had in the same time period.

The crucial difference between party list and STV, however, is in the latter’s use of preference voting. The likelihood that at least one of your preferences will help elect a candidate means votes are never wasted, which is a real advantage. However the preference system sometimes reduces the proportionality of the results, since big parties can benefit from the voter’s second and third preferences. A good example of this occurred in Ireland’s 2002 general election results, which were the least proportional ever: major party Fianna Fail won 41% of the vote but got 49% of the seats. Doesn’t sound very proportional to me…

Furthermore, a good argument can be made that people don’t put as much thought into lower preferences as they do for their first. This is simple psychology: ask someone  what their third choice of anything is, and you’re not going to get a very well thought out answer.  (Third favourite city? Thought so…) Therefore ensuring a voter’s second, third or even fourth preference make a difference is arguably not as  important in terms of electoral legitimacy as making sure someone’s first choice counts.

Another obvious criticism of STV is that it’s confusing. Really, really confusing. The ballot itself is not that confusing -  after all, it’s not really difficult to list your chosen candidates in some kind of numerical order. But to understand how your vote will be counted is arguably a challenge of Herculean proportions that’s unfair to place on the head of your average voter (if you don’t believe me, check out the examples on Wikipedia. Your brain will melt). Thus, the complexity of STV is a real problem.

Additional Member Systems

The final option on the table is an “additional member system”, which combines two very different types of electoral model in an attempt to get all the good things without any of the bad, a sort of “best of both worlds” option. This generally involves a majority of the Mps being voted in through a non-proportional, single member constituency system, while a minority are elected with party list PR. To take an example from the UK, 57% of members of the Scottish Parliament are elected through the normal process of first-past-the-post, while the remaining 43% are voted for using a regional party list. This allows for the benefits of first-past-the-post – the strong constituency link to the elected representative, a fairly stable government – but compensates for its complete lack of proportionality by the addition of members elected through PR. The Scots therefore have two votes: one for their constituency MSP and one for a party list.

Then there’s the Alternative Vote Top-up, or AV+. AV+ is an untested model invented in 1998 by members of  the Jenkins Commission, a commission set up by the UK Labour government of the time to investigate the best alternative to the first-past-the-post system. It’s similar to the Scottish model described above, in the sense that a majority of MPs are elected via a single member, constituency based system while a minority are elected through a regional party list system. The crucial difference, however, is that the constituency based system is not first-past-the-post, but the “alternative vote”. This is a preference-based system that requires an MP to gain over 50% of the vote to be elected. If no- candidates achieve this by first preferences, then second preferences are taken into account, and so on until one candidate is elected. Another difference with AV+ is that only 10%-20% of MPs would be elected via regional party lists which, critics argue, is nothing more than a faint nod to the idea of proportionality.

AV+ is thus a model preferred by  many British politicians, since it would allow the current constituency system to stay in place  – thus protecting their seats – while at the same time being more proportional than first-past-the-post and would allow MPs to point out that they now have to have the support of a majority of voters in their district. Current Home Secretary Alan Johnson has come out in support of it and leader of the Liberal Democrats Nick Clegg, who like most Lib-Dems would prefer STV, says that he could live with it. It’s probably the front runner to replace first-past-the-post.

However, though AV+ is better in one sense than the Scottish system – because a constituency MP would require a majority to win, rather than simply be the one with the most votes – the requirement of  a majority would only slightly make the number of votes cast more proportional to the seats. The Alternative vote is semi-proportional at best.  Moreover, the fact that a preference system would be used to give MPs a majority could lead to some safe seats becoming more safe. Strong Labour safe seats, for example, would become more so since Lib Dems would be likely to give Labour their second preferences. Vice versa for the Lib Dems. Then there’s the question marks over a preference system itself, which I mentioned while discussing STV. Finally, the party-list section of AV+ – the “top-up” -  is much worse than Scotland’s version. A mere 15%-20%, compared to 43%? This would be coating a flawed system with the thin veneer of proportionality.

Poor reasoning?

So which system is best? Party-list, STV, or one of the mixed “additional member” systems? The crucial problem with a mixed system is that its aim is to sacrifice (more or less) complete proportionality in favour of maintaining a balance of the status quo. Thus we are told that a mixed system ensures there will be a strong government as opposed to the weak coalitions of, say, a party list system. The trouble with this argument isn’t just, as I’ve mentioned before, that it isn’t factually true – plenty of party list PR countries have strong coalitions – but that it’s deeply offensive.  It involves accepting an equal balance of considerations like “strong government” against the consideration of ensuring that the seats won accurately represent the votes cast. But is this right? Surely allowing every one’s vote to be proportionally represented is an absolute, basic right – someone that can’t simply be placed on one end of a set of scales and bargained down. Otherwise, as I argued in part one, you end up defending the idea that if  representing the will of the voters more accurately will lead to a weak, gridlocked government, then we must make sure that the voters aren’t more accurately represented. This is a fine argument for a philosopher-king, but not a great argument for a democracy.

The same is true of the argument that only mixed systems ensure constituency representation. Again, this argument isn’t factually true – under the single transferable vote, (and indeed party list) you can choose how big your multi-member districts are, so they can in effect act as constituencies, albeit much larger ones than we are used to. Plus you can still get to vote for an individual rather than a party.  Indeed, one of the criticisms levelled at the use of STV in Ireland  is that it makes their MPs beholden to their constituents, leading to them obsess over local issues rather than spend any time in Parliament. But the argument that only a mixed system would protect the constituency link also doesn’t make any sense: it’s a bit odd to say that an MP’s connection to his constituents is vital when many of the votes of these same constituents – who may not have voted for their MP – are being unfairly represented. Watering down the proportionality of electoral reform to uphold such a paradox would be a ridiculous thing to do.

In summary, then, there is something unsettling about the kind of logic behind a mixed, “additional member” system where fair, equal democracy is bartered down to an acceptable level designed not to unsettle the status quo. For this reason, any sensible discussion of electoral reform options must surely come down to a choice between party list and  the Single Transferable Vote. Personally I favour party list – though strictly of the “open” variety – since I have serious doubts about the complexity of STV, not to mention the value of a preference system. But either would obviously be a great deal better than the current system, or indeed a mixed system.

But just as it can be safely predicted that either of those two would lead to a fairer voting system, so it can be safely said that neither stand a great chance of getting the green light.  Even in these times of public anger and national crisis, the fact remains that most politicians don’t want to see their world up-heaved and the safety of their seats challenged. There will no doubt be those who argue that any improvement over the current system will do and that going down the path towards the ideal voting system requires baby steps. Perhaps – but sometimes I just wish we’d all grow up. PR to the people? We’ll have to see…