![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Entry tags:
Why I'm a Liberal Democrat part 4: electoral and parliamentary reform
'It’s called “democracy”, and I kind of like it.' – Nick Clegg, party conference March 2010
Electoral reform is surely a idea whose time has come, and Liberal Democrats are no longer lonely voices in the wilderness calling for it. The surge in Liberal Democrat support over the past fortnight has left the very real prospect of Labour getting fewer votes but more seats than the Lib Dems. The inequality had been apparent for some time before that, of course, but it seems as though as long as the parties came in in the right order, no one much minded.
The voting system isn't the only thing that needs changing, though. The British public is highly disenchanted with its MPs, mostly as a result of the expenses scandal. I admit that I think the disenchantment is out of proportion to the offence; certainly there were some egregious examples of abuse coupled with some ‘dog ate my homework’ levels of excuses, but the majority of MPs did little or nothing to criticise, and our government is not vastly more corrupt than others in Europe. That said, there are still severe problems inherent in the system, and we the electorate are right to be angry about the abuses that did happen and sense of entitlement that prevailed. Lib Dem blogger Mark Thompson was the first to spot that the likelihood of expenses abuse correlated well with the safety of the MP’s seat. Correlation is not causation, of course, but in this case it makes sense to me that the safer an MP's seat, the less that MP feels themselves under the scrutiny of the electorate.
The Liberal Democrats' proposal to instigate a recall scheme for MPs will help with this. But far better is to reduce the number of safe seats in the first place. And we'd do that by getting rid of the current First Past The Post (FPTP) system and replacing it with a proportional one.
Of the non-FPTP systems, my preferred choice is Single Transferable Vote (STV) in multi-member constituencies of between four and six MPs. In STV, you number all the candidates on your ballot paper in order of preference continuing until you no longer care. If your most preferred candidate isn't elected, your vote is redistributed to your second choice, and so on. This makes it more likely that one of the candidates you've expressed a preference for is likely to get elected, and means that even if you mark your first preference for someone who doesn't get elected, you haven't 'wasted' your vote.
(An STV count is actually a bit more complicated than this makes it sound, usually involving fractional votes being passed around; I can elaborate in comments if asked.)
The other key point of this system is the 'multi-member' aspect, which is the one that delivers some semblance of proportionality. STV in single-member constituencies is known as 'Alternative Vote' or 'Instant Runoff', and tends to provide poor proportionality (as, obviously, does FPTP). If you only elect one MP per constituency, then the electoral 'view' of that constituency will be of a single party, no matter how the votes split. With five MPs, there's more scope for reflecting the diverse voting intentions of the voters across the constituency. A solidly Labour constituency may still return a block of five Labour MPs, but one in a more diverse area might return three Labour, one Lib Dem and one Conservative.
From the point of view of proportionality, an ideal at one extreme would be to treat the UK as a single massive constituency with 650 members, but this has disadvantages depending on how you do it. If you just make it a vote for parties rather than candidates, then your ballot paper is a lot shorter, but that means that party MPs are chosen from a list that the public doesn't get to vote on. This encourages candidates to curry favour with the party hierarchy rather than the voters. And if you make it a vote for candidates themselves (note that nobody would seriously propose doing this UK-wide), the ballot paper gets unmanageably long, and voters are likely to recognise only a fraction of the candidates on the list.
I prefer the compromise (and it is a compromise) of the multi-member seat approach. This allows voters to vote for individual candidates rather than parties, but keeps ballot paper length down to a manageable size. A constituency of between four and six MPs seems about right to me; the constituencies would be proportionately larger, so we're not talking about suddenly quadrupling the number of MPs here. (As it happens, we want to reduce the number of MPs overall by 150 as part of this process, so the constituencies would be slightly larger still.) Furthermore, this keeps at least some semblance of a regional link between MPs and their voters, with the advantage that most voters are now likely to be represented by at least one MP with views that reflect theirs.
A further advantage of multi-member constituencies is that parties can – and usually do – put up more than one candidate. This means that the electorate can choose to vote out an unpopular MP from a party even if they support the party in general. I don't want to over-sell this, mind you. Parties still have strong control over who gets to stand as a candidate, and it's unusual for a party to field many more candidates than the seats they're expecting to win. However, choosing between candidates in the same party is still more possible with multi-member constituencies than with single-member ones, which makes 'safe' seats a lot less safe. Yes, you may be standing as a Tory in a seat that's elected a Tory since Victoria was on the throne, but you're probably standing against three or four fellow Tories. At least some of whom may not have used public money to renovate their gatehouse. As a bonus, this system also makes it more possible for well-liked independent candidates to stand and be elected.
If STV proves to hard to sell to a hypothetical hung parliament, we may end up with 'Alternative Vote Plus' (AV+) as a system. This is the system recommended by the Jenkins Commission in 1998 and appears to be the one currently favoured by Labour. In this system, the bulk of the MPs are elected by AV in single-member constituencies, but they are 'topped up' by a set of candidates elected in regional blocks to make the final result more proportional. The system under consideration doesn't guarantee proportionality (there are too few extra MPs for that), but their notional order on the party lists will at least also be chosen by the voters rather than the public. I strongly dislike AV+; I feel that it's overcomplicated and does nothing well that STV doesn't do better. However, it is a viable alternative with a lot of support, and I have a instinct that it may be easier to sell to the public. Still, STV is my favoured system, and that of the party, too.
One consequence of a more proportional voting system is that locally popular candidates from smaller parties will find it easier to get elected, and yes, that includes the BNP. It's not a happy thought, but I firmly believe that if a large enough number of voters turn out to vote to elect a BNP MP, then the solution is not to disenfranchise them, but to offer them a better alternative. By that, I don't mean 'another racist party that's a bit fluffier', but 'a non-racist alternative that addresses their concerns'. The vast majority of BNP voters - as opposed to activists - vote BNP as a form of protest vote; Lib Dems in Burnley have had great success fighting them on this basis. (Historically, we haven't always done this well; the Liberals in Tower Hamlets would be a poor example to follow, for example.)
Voting reform is a long-standing pillar of Liberal policy, but it's by no means the be-all and end-all. We're also proposing fixed-term parliaments, lowering the voting age to 16, and replacing the House of Lords with a smaller, fully-elected upper chamber. We're also in favour of capping donations to political parties, limiting the ability of any one individual or organisation to buy influence at the party level.
We would also seek to introduce a written constitution and, in a bold if impeccably democratic move, we wouldn't impose this from above. The constitution would be drawn up in a citizens' convention with a majority of people drawn from all across the UK and all walks of life, and once agreed the constitution would be subject to a national referendum. Democracy; I like it.
Electoral reform is surely a idea whose time has come, and Liberal Democrats are no longer lonely voices in the wilderness calling for it. The surge in Liberal Democrat support over the past fortnight has left the very real prospect of Labour getting fewer votes but more seats than the Lib Dems. The inequality had been apparent for some time before that, of course, but it seems as though as long as the parties came in in the right order, no one much minded.
The voting system isn't the only thing that needs changing, though. The British public is highly disenchanted with its MPs, mostly as a result of the expenses scandal. I admit that I think the disenchantment is out of proportion to the offence; certainly there were some egregious examples of abuse coupled with some ‘dog ate my homework’ levels of excuses, but the majority of MPs did little or nothing to criticise, and our government is not vastly more corrupt than others in Europe. That said, there are still severe problems inherent in the system, and we the electorate are right to be angry about the abuses that did happen and sense of entitlement that prevailed. Lib Dem blogger Mark Thompson was the first to spot that the likelihood of expenses abuse correlated well with the safety of the MP’s seat. Correlation is not causation, of course, but in this case it makes sense to me that the safer an MP's seat, the less that MP feels themselves under the scrutiny of the electorate.
The Liberal Democrats' proposal to instigate a recall scheme for MPs will help with this. But far better is to reduce the number of safe seats in the first place. And we'd do that by getting rid of the current First Past The Post (FPTP) system and replacing it with a proportional one.
Of the non-FPTP systems, my preferred choice is Single Transferable Vote (STV) in multi-member constituencies of between four and six MPs. In STV, you number all the candidates on your ballot paper in order of preference continuing until you no longer care. If your most preferred candidate isn't elected, your vote is redistributed to your second choice, and so on. This makes it more likely that one of the candidates you've expressed a preference for is likely to get elected, and means that even if you mark your first preference for someone who doesn't get elected, you haven't 'wasted' your vote.
(An STV count is actually a bit more complicated than this makes it sound, usually involving fractional votes being passed around; I can elaborate in comments if asked.)
The other key point of this system is the 'multi-member' aspect, which is the one that delivers some semblance of proportionality. STV in single-member constituencies is known as 'Alternative Vote' or 'Instant Runoff', and tends to provide poor proportionality (as, obviously, does FPTP). If you only elect one MP per constituency, then the electoral 'view' of that constituency will be of a single party, no matter how the votes split. With five MPs, there's more scope for reflecting the diverse voting intentions of the voters across the constituency. A solidly Labour constituency may still return a block of five Labour MPs, but one in a more diverse area might return three Labour, one Lib Dem and one Conservative.
From the point of view of proportionality, an ideal at one extreme would be to treat the UK as a single massive constituency with 650 members, but this has disadvantages depending on how you do it. If you just make it a vote for parties rather than candidates, then your ballot paper is a lot shorter, but that means that party MPs are chosen from a list that the public doesn't get to vote on. This encourages candidates to curry favour with the party hierarchy rather than the voters. And if you make it a vote for candidates themselves (note that nobody would seriously propose doing this UK-wide), the ballot paper gets unmanageably long, and voters are likely to recognise only a fraction of the candidates on the list.
I prefer the compromise (and it is a compromise) of the multi-member seat approach. This allows voters to vote for individual candidates rather than parties, but keeps ballot paper length down to a manageable size. A constituency of between four and six MPs seems about right to me; the constituencies would be proportionately larger, so we're not talking about suddenly quadrupling the number of MPs here. (As it happens, we want to reduce the number of MPs overall by 150 as part of this process, so the constituencies would be slightly larger still.) Furthermore, this keeps at least some semblance of a regional link between MPs and their voters, with the advantage that most voters are now likely to be represented by at least one MP with views that reflect theirs.
A further advantage of multi-member constituencies is that parties can – and usually do – put up more than one candidate. This means that the electorate can choose to vote out an unpopular MP from a party even if they support the party in general. I don't want to over-sell this, mind you. Parties still have strong control over who gets to stand as a candidate, and it's unusual for a party to field many more candidates than the seats they're expecting to win. However, choosing between candidates in the same party is still more possible with multi-member constituencies than with single-member ones, which makes 'safe' seats a lot less safe. Yes, you may be standing as a Tory in a seat that's elected a Tory since Victoria was on the throne, but you're probably standing against three or four fellow Tories. At least some of whom may not have used public money to renovate their gatehouse. As a bonus, this system also makes it more possible for well-liked independent candidates to stand and be elected.
If STV proves to hard to sell to a hypothetical hung parliament, we may end up with 'Alternative Vote Plus' (AV+) as a system. This is the system recommended by the Jenkins Commission in 1998 and appears to be the one currently favoured by Labour. In this system, the bulk of the MPs are elected by AV in single-member constituencies, but they are 'topped up' by a set of candidates elected in regional blocks to make the final result more proportional. The system under consideration doesn't guarantee proportionality (there are too few extra MPs for that), but their notional order on the party lists will at least also be chosen by the voters rather than the public. I strongly dislike AV+; I feel that it's overcomplicated and does nothing well that STV doesn't do better. However, it is a viable alternative with a lot of support, and I have a instinct that it may be easier to sell to the public. Still, STV is my favoured system, and that of the party, too.
One consequence of a more proportional voting system is that locally popular candidates from smaller parties will find it easier to get elected, and yes, that includes the BNP. It's not a happy thought, but I firmly believe that if a large enough number of voters turn out to vote to elect a BNP MP, then the solution is not to disenfranchise them, but to offer them a better alternative. By that, I don't mean 'another racist party that's a bit fluffier', but 'a non-racist alternative that addresses their concerns'. The vast majority of BNP voters - as opposed to activists - vote BNP as a form of protest vote; Lib Dems in Burnley have had great success fighting them on this basis. (Historically, we haven't always done this well; the Liberals in Tower Hamlets would be a poor example to follow, for example.)
Voting reform is a long-standing pillar of Liberal policy, but it's by no means the be-all and end-all. We're also proposing fixed-term parliaments, lowering the voting age to 16, and replacing the House of Lords with a smaller, fully-elected upper chamber. We're also in favour of capping donations to political parties, limiting the ability of any one individual or organisation to buy influence at the party level.
We would also seek to introduce a written constitution and, in a bold if impeccably democratic move, we wouldn't impose this from above. The constitution would be drawn up in a citizens' convention with a majority of people drawn from all across the UK and all walks of life, and once agreed the constitution would be subject to a national referendum. Democracy; I like it.
no subject
A great advantage of multi-member constituencies, especially if you have a wide variation of sizes, is that you can draw boundaries based on real communities rather than on population and just add an extra MP when a seat gets too big, or lop one off when it gets too small. At some point you would have to merge or split constituencies, but probably only a handful in each review, which could easily be made every 20 years, while number of MPs per seat could be adjusted each election based on the actual electoral roll on the day.
Some example constituencies, to show the advantages:
Birmingham
London, East End
London, West End
Inner London, South of the River
Cornwall
Bristol
Surrey
Highlands (of Scotland)
Glasgow
Borders (within Scotland)
Cumbria
The most difficult are the large towns / smaller cities which presently have 1-3 MPs and where county seats get impractical. Hampshire, for example, currently has 18 MPs but has no natural divisions other than Portsmouth (2 MPs) and Southampton (2.5 MPs).
no subject
no subject
.. because I understand why the Tories do: a) it's an excuse to disproportionally reduce the number of non-English constituency seats and b) it increases the Parliamentary power of the government - there will be far fewer back bench MPs in the ruling party/ies not covered by 'ministerial responsibility' to always toe the line.
no subject
no subject
no subject
no subject
(An STV count is actually a bit more complicated than this makes it sound, usually involving fractional votes being passed around; I can elaborate in comments if asked.)
You offered :) I'd be interested in this elaboration.
These posts are really interesting, I know I don't understand lots of stuff and you make it easy to understand while being clear what is your opinion and stuff. I am probably already a convert as I've usually been LibDem but it's useful to be able to learn more about the meaning behind their policies and how they reflect my beliefs or indeed those of my friends. You as someone whose beliefs I respect greatly even if I don't always agree!
Interestingly my family have historically been lib dem by preference, I think for the PR and honesty about 'taxing those who can afford it' as Scottish based people they resented "the English Government" for a long time and as 'Christians' believe rich people should pay proportionately more tax than the poor - iirc there was some increasing income tax by n pence in pound my parents favoured. Tax rises were never seen as a bad thing in our household even if it made us worse off as my parents were aware my dad had a good job and we benefitted from good public services e.g NHS.
no subject
1) Count up all the first preference votes for each candidate.
2) Establish the 'quota'. This is the number of votes that a candidate needs to get to be certain of being elected. It is calculated as follows:
(total number of votes cast)
------------------------------------- + 1
(total number of remaining seats + 1)
In the case of just a single seat, that simply reduces to half the total number of votes cast plus 1. In an election for four seats, it would be a fifth of the total vote plus one. A candidate who has achieved the quota (or higher) has enough votes that they cannot be overtaken by more candidates than there are seats left, so they are definitely elected.
3) If any candidates have achieved the quota, they are elected. There is a small fanfare (literally, if you're tallying the votes with with Colin Rosenstiel's STV software). If not, go to step 6.
4) If a candidate has achieved exactly the quota, no further action need be taken. However, if they have more votes than the quote, the extra votes - the 'surplus' - needs to be redistributed. Conceptually, these are votes that would otherwise be wasted, having been cast for a candidate who was already elected. The surplus is distributed by taking *all* the candidate's votes and distributing them according to their next preference down (which will be the second preference in this initial round). However, the votes at this stage are distributed with a fraction of their value. For example, if the quota is 100 and the candidate actually got 125 votes, then their surplus is 25. However, you will actually distribute all 125 next preference votes, but at a valie of 1/5 of a vote, to make a total of 25 votes from the 125 ballot papers.
5) Go to step 2 - you need to recalculate the quota each time, because (a) you now have fewer remaining seats and (b) some votes have no later preference and need to be removed from the total votes cast for the purposes of calculating the quota.
6) The candidate with the lowest number of votes is eliminated, and their votes redistributed according to their next preference down.
This continues until enough candidates reach the quota, or you have eliminated enough so that you only have as many candidates left as you have seats.
See here for variations:
http://en.wikipedia.org/wiki/Counting_Single_Transferable_Votes
no subject
I may ask the lazyweb some clarification questions if I still can't bash it into my head by the weekend! :)
no subject
Thanks, djm4, for explaining STV: I was wondering how second (etc.) preferences were weighted.
no subject
I also don't believe that electoral reform will go as far as STV in the form you propose. In which case the least-worst option is I am pretty sure Condorcet (in one of its clone-reducing instantiations).
But I don't imagine that'll find much favour either.
no subject
I could be wrong. I find it difficult to get a handle on how a Condorcet counting system would play out in the UK at the moment, and long-term, the political landscape might well adjust to suit the voting system.
Size of multi-member constituencies
(Anonymous) 2010-04-28 01:22 pm (UTC)(link)Re: Size of multi-member constituencies