DPEV: absolutely the best single-member voting system for the UK and England – honest

OK, I admit it: I’m a voting-system geek, if not obsessive. I really dislike AV, for a number of reasons, not the least of which is that it leaves England short-changed: nothing done to address the West Lothian Question or the broader English Question, to say nothing about the unaccountable nature of executive power in the UK, which relies on a disproportional voting system and a subservient parliament to run the country virtually as an elective dictatorship.

During the course of the last few months, in this blog, I’ve discussed a number of alternative single member-constituency voting systems that are better than the Alternative Vote, in my view, despite the fact that AV is the only alternative on offer. It seems to me I’ve been fishing around for a ‘killer’ system: one that is simple, fair and transparent but which also addresses the two main failings of First Past the Post and AV – that they 1) produce disproportional results and 2) bring about governments with no real mandate that can basically get away with whatever they want (a simplification, I know, but it sometimes feels that way). In addition, if a voting system passes the ‘English parliament test’ – or at least a fairness-to-England test – then all the better. Basically, if I’d be happy if the system in question were used to elect an English parliament, then it must be OK.

I now think I’ve come up with such a ‘killer voting-system app’, so to speak. It’s called DPEV: ‘Dual Parliamentary and Executive Voting’. How it works is as follows:

  1. DPEV is a single-member system. There are two parts to the voting process: a First Past the Post ballot of individual candidates and a separate ballot listing the parties standing in that constituency. Voters must select the individual candidate they’d like to be their MP by marking a cross next to their name – exactly as under the present FPTP system. Again, just as with the present system, the winner is the candidate obtaining the most votes. Voters must also mark a cross next to the name of the party or parties they’d like to form the next UK government. Here, they can vote for more than one party, thereby expressing a preference for a coalition government of the parties in question.

     

  2. The individual-candidate vote is used to determine the composition of Parliament, whereas the party vote is used separately to determine which parties have a mandate to form the next government. Basically, if one party wins an outright majority of the party vote, they are deemed to have a mandate to form the government. Otherwise, the strongest multi-party combination is considered to have a mandate to form a coalition government so long as the overall total of votes for both or all of the parties involved adds up to more than 50%.

    For example, let’s say that 30% of voters want the Tories only to be in government; 25% want just Labour; 10% want only the Lib Dems; 16% want a Lab-Lib coalition; and 11% want a Tory-Lib Dem coalition. Here, the system decides there is a mandate for a Lab-Lib Dem coalition because this was the most popular party-combination vote (i.e. 16% vs. 11% for a Con-Lib combo) and the total number of voters wanting either a Labour or Lib Dem government (single-party or coalition) adds up to a majority (51%). However, if the percentage of voters wanting a Lab-Lib Dem coalition had been only 14%, those parties’ combined vote would add up to only 49%. In that case, despite having the largest coalition vote, there would be no implied mandate for a Lab-Lib Dem coalition, and it would be down to the Conservatives and Lib Dems to try to reach a coalition deal, as the combined total of individual and joint votes for those parties would be 51%.

One other point of detail here: to form a government, a party or combination of parties must garner over 50% of the party vote in England as well as across the UK as a whole. This is because, without devolution for England, the UK government is also an acting English government and so must have a democratic mandate in England separately. Obviously, the best solution would be a completely separate English parliament and executive. But in the absence of that, this does address the West Lothian Question, if not the English Question: it wouldn’t matter, on one level, if English laws were passed by parliamentary majorities that included non-English MPs if the government enacting those laws had a legitimate democratic mandate from English voters. This is what this ‘England lock’ on the government is designed to ensure.

So basically: in the absence of an outright majority for any party in the party-vote bit of DPEV, parties must try to form majority coalitions, beginning with the multi-party combination vote (i.e. where people vote for two or more parties together) that obtained the largest share of the votes across the UK, so long as those coalitions command an overall majority of the votes across the UK and England.

What would happen if neither a Lab-Lib Dem nor a Con-Lib Dem coalition could muster a block of votes of over 50% across the UK or in England only? In this instance, the parties would have to bring in a third coalition partner that had won a sufficient percentage of the party vote across the UK and England, as applicable, to constitute a majority mandate, so long as that third party had won some MPs. However, even so, it is possible that a majority coalition could still not be formed, either because there was insufficient political will among the parties involved or because there would be parties for which people had voted that were without MPs, such as – for example – UKIP, the Greens, the BNP or the English Democrats. In this instance, the parties would have to try to form a coalition or single-party government based on the majority of available party votes. E.g. if only 80% of the UK-wide party vote had been for parties that succeeded in winning MPs, then the government (single-party or coalition) would have to command over 40% of the party vote across the UK, and whatever the corresponding majority percentage would be in England: probably higher than 40% given the lower share of the votes won by nationalist or sectarian parties in England compared with the UK’s other nations.

To prevent this rule becoming a get-out clause allowing the Conservatives or Labour to form governments that were without a true majority mandate, one of the major constitutional innovations of DPEV would be that any government commanding the support of only a minority of voters (as determined by the party vote) would have to be ratified by the electorate in a snap referendum following the conclusion of coalition negotiations. And that means a UK-wide referendum if the government in question was based on minority support across the UK as well as England, or a referendum in England only if the proposed government enjoyed a majority across the UK but was backed by only a minority in England.

If such a referendum failed to win the endorsement of either the British or English people, then the parties would have to go back to the drawing board and try to find a majority coalition or alternative minority government. If the latter were the outcome, this too would need to be ratified in a referendum. Then, if this in turn failed, a new election would have to be held – but not a whole general election with all the new constituency MPs needing to seek re-election, merely a new party election, in which all of the parties that had won MPs would be standing. This would determine a new clear majority mandate, as now the choice of parties would be greatly cut down. In practice, in the absence of an outright majority for any single party in this second party vote-only election, it would be pretty obvious which combination of parties (e.g. a coalition between the Lib Dems and one of the other major parties) had a clear mandate, and the parties concerned could have a constitutional obligation to work together.

These provisions for majority coalitions or – in the absence of majorities – ratifying referendums and follow-up elections, as required, would ensure that any UK government had a clear majority mandate from the people across the UK and England, irrespective of whether the party or parties in government commanded a majority of MPs in Parliament. This relates to another original feature of DPEV: MPs are elected using a disproportional system (FPTP), but the executive is elected using a perfectly proportional system – so the government takes its democratic mandate directly from the people, not from parliament. This does mean that the parliamentary majority could be at odds with the executive majority, and governments could well find they commanded only a minority in Parliament. However, this could be a good thing, in that Parliament would be able to hold governments to account more effectively. Equally, the system for electing MPs could be changed separately, without altering the perfectly proportional method for electing the executive. E.g. you could introduce AV, STV or some other system for electing Parliament that would make it more proportional and representative, and make the majority in Parliament less likely to clash with that of the government.

Not only the executive but also MPs would be more directly accountable to their voters, in two main ways:

  1. Separating out the vote for an individual MP from the vote for a government, as DPEV does, restores the direct accountability of an MP to his or her constituents. Voters can now choose an MP on the basis of their individual suitability for the role – their experience, character and values as well as political opinions – without prejudice to the party they want to be in government, which they vote for separately. By contrast, under the present FPTP system, and the proposed AV system, if you vote for the man or woman, you are also voting for the party and are presumed to be endorsing that whole party’s programme for government as set out in their manifesto. This is rolling up two distinct choices in one, and it’s what helps make MPs subservient to their party apparatus in Parliament, because they are presumed to have been sent to Parliament to fulfil that party’s programme. Under DPEV, each individual MP has been personally chosen by the plurality of their voters: having been elected independently, they are empowered to act independently.
     
  2. This independence from party, and accountability to voters, would be reinforced by another constitutional innovation that would be associated with DPEV. Let’s say a Conservative MP has been elected into Parliament, but a majority of constituents had voted either Labour or Lib Dem (or for both Labour and Lib Dem) in the party vote, and a Lab-Lib Dem coalition was in fact formed. Then, if that Tory MP persistently votes against government bills at their third and final reading (e.g. in 50% or more of cases), constituents should have the right to demand a by-election to hold that MP to account. If the MP is re-elected, they could be said to have received a mandate to continue opposing government bills. The smart thing for the government parties to do in this instance would be to field only one candidate to ensure a government majority in the by-election – but whether they’d have the wit to do that or not is moot.

    I would envisage that voters would be able to call a by-election on this basis one year after the general election, and then again after another year, by means of, say, more than 10% of the electorate turning up at polling stations on a designated day to sign a petition for such a by-election. Once more than two years have elapsed after the general election, there should be no further by-elections of this sort, in that – ideally – there would be four-year fixed-term governments, so that holding a large number of by-elections at the close of the third year of the parliament would be somewhat excessive. (Incidentally, if a coalition collapsed before the fixed term had expired, it could be made mandatory for the parties to try to form a new coalition – but this would also have to be ratified by referendum. If it was rejected in the referendum, then a general election would have to be held.)

    This method of holding MPs to account could be applied to any MP that persistently voted against the party majority in their constituency. For instance, if a candidate from one of the government parties had been elected as MP but a majority of constituents had not voted for the party or parties of government in the party vote, those MPs could also be held to account and forced to fight a by-election if they persistently voted with the government. This means that MPs would truly have to respect the opinions of their voters and take them into consideration in their work in Parliament, alongside party loyalty.

All these aspects of DPEV would bring about much greater popular sovereignty and political accountability: the government taking its mandate direct from the people; MPs directly accountable to their voters and expected to act independently of party dictates. And, as I said above, it provides a solution to the West Lothian Question, if not an answer to the English Question – but I would say that it’s still an excellent voting and constitutional system for any English parliament and so passes my English parliament test.

How does DPEV perform in relation to the six criteria I’ve been using to assess the merits of different single-member voting systems? My first criterion is: Does every vote count, and is every vote counted? Here, I’d give DPEV four out of five. Every party vote, under DPEV, counts in the sense that the right to form a government depends on every single vote cast. However, as many parties for which people vote would still not win MPs, those parties could not participate in government. Similarly, the constituency vote counts for more than it does presently under FPTP, in that MPs are elected independently of their party affiliation and are expected to act accordingly. However, as the system used to elect MPs – at least in my initial version of DPEV – is the disproportional FPTP, many constituency votes will count for little.

In terms of the second criterion – Is the system proportional? – I’d again award DPEV four out of five: it’s perfectly proportional in terms of the party vote that is used to determine the shape of the government,but disproportional with respect to the constituency vote. However, the fact that the government derives its mandate direct from the people, and the fact that accountability of MPs to constituents is built into DPEV, makes it less critical to achieve a perfectly proportional parliament.

The third criterion is: Does the system foster accountability? Here, I’d modestly give DPEV five out of five: it embodies a very high degree of accountability of MPs to their constituents, and it also makes the government directly answerable to the whole electorate, a majority of whom have to give it their backing, either in general elections and referendums to ratify minority governments or coalitions.

Fourthly: Does the voting system allow voters to express the full range of their political and personal preferences, and send a message to politicians? Here, DPEV scores four out of five. It enables voters to support individual candidates that can be of an altogether different political persuasion to themselves, on the basis that whichever candidate you vote for (based on their personal qualities), you can vote for different parties to form the government. In addition, you can vote for as many parties as are on the ballot paper in the party vote. Not all of those votes will be effective, however, in the sense of resulting in representation – and in fact, it would be silly to vote, say, for four parties, as it is unlikely that such a vote will be rewarded with a coalition of all four parties. But all party votes are nonetheless recorded, so that voters can send a message to politicians. In addition, the more people voted for parties such as UKIP and the Greens, the more people would feel emboldened to vote for UKIP and Green candidates, too – with the added incentive that even if only one or two MPs from those parties were elected, they might go straight into government as part of a coalition.

The fifth criterion is: Does the system mitigate / obviate tactical voting? Here, I’d give DPEV four out of five. Under DPEV, there is virtually no incentive for tactical voting, other than to try to defeat the candidate of a particular party in the constituency vote. But the reward for doing so is considerably less than under FPTP or AV, in that MPs of any hue are supposed to act independently and can be held to account if they put party interest above that of constituents by opposing legislation that the constituents have by implication supported (by voting in a majority for the parties that are in government).

Finally, How easy is the system for voters to understand, trust and use effectively? Well, DPEV should be easy for voters to understand and use to their best advantage. You just vote for an MP in the same way as now; and you can vote separately for the party or parties you want to form the next government. However, some voters might find the separation of the candidate and party vote confusing, and also might not use the option to combine votes for multiple parties very effectively (i.e. they might select several parties or not understand that selecting multiple parties means you’re expressing a preference for a coalition of them all). So I’d give DPEV four out of five here.

So here’s how I rate DPEV in comparison with the other single-member systems I’ve discussed, including several I’ve ‘invented’ myself, as I have with DPEV. For the sake of comprehensiveness, I’m also rating the variants of AV I’ve discussed recently (FMT (First Past the Post Majority Top-UP) and AV 2.0); and the method I evoked in my post yesterday, whereby you just have two preferences, and if there is no majority of first preferences, the second preferences of all voters are added to all candidates’ totals, and the winner is the candidate obtaining the most votes (let’s call that ‘TPP’: Two-Preference Plurality!):

Criterion FPTP AV AppV ARV TMPR AV+ NetV 3CV Bucklin
Does every vote count?

3

2

3

4

4

3

4

3

3

Is the system proportional?

1

2

2

3

3

3

3

3

3

Does the system foster accountability?

3

3

4

4

4

4

4

4

3

Does the system let voters express their views?

1

2

3

4

4

3

4

3

2

Does the system mitigate tactical voting?

1

2

2

3

4

2

3

3

3

How user-friendly is the system?

3

2

4

3

3

3

3

3

4

Total scored out of a maximum of 30

12

13

18

21

22

18

21

19

18

 

Criterion DPEV FMT AV 2.0 TPP
Does every vote count?

4

3

4

4

Is the system proportional?

4

2

2

2

Does the system foster accountability?

5

3

3

3

Does the system let voters express their views?

4

3

3

2

Does the system mitigate tactical voting?

4

2

2

3

How user-friendly is the system?

4

3

2

4

Total scored out of a maximum of 30

25

16

16

18

 

Clearly, different readers will rate these voting systems differently according to their own perspectives: my ratings are to an extent subjective. My scoring system is, however, based on an attempt to think through the main implications of the systems in terms of the degree to which they provide an accurate and user-friendly means for voters to record their opinions, and how they might influence voter behaviour.

But DPEV is the clear winner: a single-member system that would produce more accountable government and MPs, and would be fairer to England. I know I would say so, but it gets my vote!

Advertisements

7 Responses

  1. A comparison between DPEV and DPR Voting would be interesting.

  2. In DPR Voting MPs (independents and party sponsored) are elected much more on personal merit, and not on Party label. This aspect is shared with DPEV because the voting process is similar. This gives MPs a measure of personal electoral legitimacy. It makes them more accountable to their electorate and less beholden to the Party. (An MP can still be elected on personal merit when the party is very unpopular in the constituency. Similarly an unpopular Party nominee parachuted into an apparently safe seat may well not be elected because the Party vote is separate).
    The system would encourage MPs to be more independently minded and to vote accordingly because they will seek re-election on the basis of their track record.

    The DPR Voting Party Vote is simpler than the DPEV Party Vote.
    Can you show how the DPEV Party vote works – the choice between a single vote and a dual vote – that sounds like quite a complex tactical choice with significant implications. Can you run some examples and show how it is counted?

    You have included some modifications such as the right to call a by election. This proposal could be considered separately on its merits.

    DPR Voting differs from DPEV in that it is intended as a straight replacement for the existing Parliamentary electoral system, rather than a scheme to facilitate English devolution.

    Please can you put a link to the DPR Voting website on your website?

    • Thanks for the clarification, Stephen. I still feel that DPR Voting (there’s your link: I don’t put in permanent links other than to blog sites I like) adds to the pressure on MPs to tow their party line, as their weighted vote makes it even more crucial that they vote the way their party leaderships want them to. Plus I think all MPs should be equal, because if the vote of one MP (e.g. a Lib Dem) is worth 2.5-times the vote of another (e.g. a Tory), that doesn’t just rectify an imbalance in party representation but gives that MP more clout in representing his / her constituents, e.g. (s)he could use his / her voting power to extract concessions for the benefit of constituents, which an MP with less voting weight couldn’t so easily do.

  3. 1 I don’t understand why a weighted vote makes an MP more in thrall to their party (and can I challenge you on this point?).
    The dual vote has exactly the opposite effect as you argue quite rightly in relation to DPEV. I can’t see how you can draw a distinction between DPEV and DPR Voting in this regard.

    2 It is easy to say that all MPs should be equal, but it is not one of those unarguable facts. One man one vote is about universal suffrage. It does not apply to elected representatives.

    Clearly not all MPs are equal. They represent different sized constituencies with different numbers of electorate voting, with different majorities, and overall different distribution of votes between the different parties.
    Give them all equally one vote and you ignore all this (and incidentally that’s how you introduce the problems of disproportionality of other electoral systems.
    Another way in which MPs are not equal is that some become ministers in the Government or members of select committees, for example, which also varies their influence / clout in representing their constituents.

    DPR Voting makes a distinction between party sponsored MPs and independent MPs. Independent MPs gain their support purely from the constituency.
    The Party wins an overall nationwide level of support (quantified by the Party vote) for its manifesto. This level of support equates by simple ratio to a number of Parliamentary votes. These Parliamentary votes are shared out equally amongst the Party’s sponsored MPs to exercise when a vote is a party policy matter.
    If it is not a party political matter – a free vote – each MP has an equal vote.

    • I would say a weighted vote makes an MP more in thrall to his or her party because the value of their vote to the party is greater. For a Lib Dem MP whose vote is worth, say, 2.5 times more than a Labour MP’s vote, this is obvious: for the party to exercise the influence their share of the vote gives them under your system, then each individual MP needs to vote with the party, so a premium is placed on their loyalty. This also works for parties whose MPs’ vote is worth less – e.g. the Tories with, say, votes worth 0.75. Then, in order for the Tory vote in parliament to be maximised, it would be even more critical for every MP to vote as the whips decreed.

      As for your second point, and I quote: “These Parliamentary votes are shared out equally amongst the Party’s sponsored MPs to exercise when a vote is a party policy matter.
      If it is not a party political matter – a free vote – each MP has an equal vote.” However, you could say that the ideal of parliamentary democracy in the English / British tradition is that each MP is, or should be, free to vote with their conscience, and not along party lines or as a delegate sent to parliament to represent partisan interests. Therefore every vote is in principle a free vote, and there’s no such thing as a ‘party vote’ in that sense.

      I appreciate what you say about constituency sizes (although the present government is trying to address this issue), and about different grades of MPs. But if each MP truly represents their constituents, then they do this as a free individual, equal before his peers, expressing the equality and freedom of the ordinary citizen. So the principle of one person one vote applies in just the same way to MPs as it does to citizens.

  4. In analysing the extent to which the MP is in thrall to the Party and whether this is greater with DPR Voting compared to the current situation we need to consider the relative power and motivation of the various parties in the equation – the Party, the MP and the Constituency electorate.

    What power does the Party have over the MP? I assume it is the power of selection or de-selection of the MP, or advancement as a Minister in the Government, or the promotion of some bill where the MP has a particular interest.

    The motivation is about the collective power of the party. As a disciplined party, the full voting power of the party can be used to advance the aims of the party. If this breaks down the party will be less effective in achieving its aims. In the current system each MP has the same value vote and the party needs the votes of each MP equally. In DPR Voting each MP has a vote value equal to every other MP in the party so again the party needs the vote of each MP equally. However where vote values are greater than one, the value (or impact on party discipline) of each MP is greater.
    Thus far I think we agree.
    For a party with MPs with a vote value of less than one, the party still has the motivation to maximise discipline in the party, but the consequence of one MP breaking ranks is reduced.

    The motivation to maintain party discipline is unchanged in principle for each party. The consequence of an MP breaking ranks is mathematically different depending on the weight of the MP’s vote.

    But this is not the whole story. We also have to look at the situation from the viewpoint of the MP.

    For the MP, the party has the theoretical sanction to de-select the MP. In FPTP this is a draconian sanction because in many seats the party sponsored candidate will be safely elected, and no other party candidate or independent candidate has much of a chance, especially ‘safe’ seats. Thus regardless of popularity or personal merits, one MP can be deselected and thus have a career curtailed, and another candidate can in effect be given automatic election as an MP, based on Party sponsorship rather than merit.

    IN DPR Voting, this is quite different. The electorate in a ‘safe’ seat can continue to give their party vote to their chosen party but they now have a free choice as to which candidate to choose. Thus the MP is elected much more on personal merit and track record as judged by the electorate than by party label.

    Another aspect is that a well known popular figure has a better chance of being elected as an independent MP under DPR Voting for similar reasons.

    An interesting by-product in DPR Voting of the election results by constituency will be the comparison of the percentage and numbers of votes achieved by the party in the ‘party’ vote, and the percentage and number of votes achieved by the candidate of that party. In some situations a candidate will be seen to do relatively better than the party and in others much worse. It is also possible for a very popular candidate to be elected when his/her party gets less than a plurality of votes, or to fail to get elected even when the party does get a plurality of the votes.

    The key aspects are:
    The election results will give a measure of how successful the MP is compared to his/her party.
    There are no longer any safe seats.
    The election of the Candidate is dependent on personal merit as judged by the local electorate, not party label.

    This curtails the party’s freedom to deselect candidates and impose their own choice on the constituency. They can still do it but there is a risk of the imposed candidate failing to be elected. The threat of the deposed candidate standing and being elected as an independent is increased.

    The pressure the party can exert on an MP is thus reduced. The power that the local electorate has is increased.

    Second point – Votes on party political issues in the Commons.
    You argue that in the ideal world each MP acts and votes as an independent free spirit, but this negates our system of politics being organised along party lines.

    In practice some matters clearly are party votes in that parties set out their objectives in their manifesto. There are other matters that are judged to be matters of party policy regardless of whether or not they were in the manifesto. DPR Voting doesn’t change this aspect of our political system.

    In practice there are very few ‘Free Votes’. Parties have a view on most matters, even if it is to abstain. That is why a Free vote under DPR Voting has the narrow definition of a vote that every party agrees is a free vote. If they don’t all agree, party voting rules apply.

    Constituency Boundaries:
    How constituency boundaries are drawn affect the result of any constituency based electoral system. With our current FPTP system constituency boundaries have a significant effect on the overall result. Removing this ‘boundary’ bias is theoretically possible by redrawing the boundaries. In practice this is technically impossible if boundaries are still intended to define community and geographical areas (and thus exceptions have to be made).
    Furthermore demographic and political changes mean that this has to be a frequent and ongoing process, and because the process takes time, boundaries are often out of date by the time they are introduced.

    Thus constituency boundaries introduce an element of unfairness to the parties in constituency based elections.

    DPR Voting removes the element of unfairness to the parties because the Party votes are aggregated on a nationwide basis and so it is irrelevant where or in which constituency the votes are cast. Boundary revisions are not necessary to make the system fair.

    Constituency boundaries do have a local impact as regards the election of the MP. However this is an entirely local matter, and while how boundaries are drawn may favour a particular candidate in particular circumstances, there is no systematic unfairness, and no unfairness to the constituency electorate.

    On representation of the Constituents:
    The key issue is that where an MP acts as a party sponsored MP ie on matters where the party is exercising party discipline, partisan issues, the voting power has been vested through the voting system with the party, and thence to the MP. The MP has an equal share of his/her parties voting power.

    Where the issue is not a Party matter, ie a Free Vote, the MP does act as an individual with a vote weight of one, just as in the current system.

    Whew!

    • Indeed! Thanks for the clarification. I can see how DPR Voting creates checks and balances that help MPs to be more independent. I think DPEV does this to a greater extent; but it’s not a competition, and I think both systems have merit.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: