Introduction

In this article, we are mainly thinking about the reform of elections to the UK parliament.  Other perspectives will be welcomed.

As you might expect, Wikipedia has a great deal of relevant background material [1]; the New Scientist has also returned this subject on numerous occasions.  The subject is a complex interplay between some highly technical analysis of different systems - of which there are many! - and the difficult question of what exactly we want the voting system to achieve.  We have an intuitive idea of what a fair voting system should achieve, but translating what we want into a concrete system in the real world is never easy and sometimes not possible.

As well as being technically complicated, this subject often generates a great deal of passion.  People - the people who talk about electoral systems - often have a strong emotional connection with their preferred system being the 'right' one.  But there is no single correct electoral system, only a variety of systems which each produce quite different outcomes [2].

The Problem

Since 1985, voter turnout has been reducing, not just in the UK but across most democratic countries [3].  And, increasingly, the poor are more disengaged than the rich: in 1987, the difference was 4%; by 2010 the difference had risen to 23%, and is likely to have grown since then; similarly, young people are increasingly less likely to vote than older people [4].   

Many people feel there is no point in voting, because 'nothing changes' - nothing important.  Which is not surprising, because the system we have is designed to resist change.  By definition, the current system works for the groups which get, and expect to get, power through using it.  Almost all those with power support the status quo - after all, those who have power gained their power within the current system, and they are unlikely to want to change the system which gave them success.

In the last UK election, the Conservatives won by a 'landslide', gaining an 80 seat majority.  But they achieved this massive success despite the majority of the votes being cast against them: they received less than 44% of the votes cast, over 56% of the votes were against the Conservatives, a difference of around 12.5%.  They repeatedly say that the people voted for them, but the people actually voted against them [5].

The failure of the current system is seen in several distinct ways.

• At the national level, many people feel the government does not represent them or listen to their needs and concerns.
• At the local level, many people live in 'safe' seats, which means that their vote - should they choose to cast it - is essentially wasted: they either voted for someone who didn't need their support, or for someone who was never able to benefit from their support.
• Also at the local level, it is often the case that people feel unable to vote for the candidate they want to win, because they care more about preventing a different candidate from winning, so they are compelled to engage in tactical voting.
• At a deeper level, the whole framework of democracy loses credibility when the winning party gains fewer votes than the losing party [6].

Tactical voting can be effective.  But it is both difficult and damaging.  Ordinary voting is hard enough - deciding on who you want to represent you in parliament, when many of the candidates will have aspects you like and aspects you dislike.  But choosing who you want, while difficult, is do-able.  Tactical voting requires you to do something much more difficult.

• Who do you like?
• Who do you dislike?
• What are their relative chances of getting elected?
• How motivated are you to stop the person you dislike getting elected?
• Who is best placed to get elected instead of the person you dislike?
• How likely is it that a vote for this other person will prevent the person you dislike getting elected?

Tactical voting requires you to decide what you think of the various candidates, and also to make a guess about what all the other voters in your constituency will do - which you cannot know before the vote.  And, even when it succeeds, tactical voting distorts democracy: if lots of people vote for someone they don't want because they think their preferred candidate has no chance, then the candidate they want has no chance, even if most of the constituency actually want to vote for them.  Tactical voting prevents you from knowing the true level of support of the different candidates.

Voting systems are complicated and obscure, and can appear to be an academic concern.  The system we are familiar with can feel like the obvious way to do things, and we are always suspicious of the unknown, but we need to try.

An Answer

This is a personal view, developed through many conversations in various different places.  But it is offered as a starting point for discussion.

Half the number of constituencies, and elect (on average) two MPs from each constituency, using two different systems.  Joining constituencies will have to be done very carefully, to avoid gerrymandering, but we have systems in place to oversee this [7].  The two MPs will have the same responsibility, obligations and rights, but the different methods of election will probably give them slightly different priorities.

All MPs will be elected purely on the basis of local votes.  All candidates will stand as an individual, and most candidates will also stand for a political party (independent candidates being the sole exception); each party can have, at most, two candidates standing in each election in each constituency; parties will be registered (as they are at present) on a national basis, and each party will have to adopt the individual as a party candidate before they can stand.  There is a very low barrier for registering as a political party [8]  - you basically need £150, a constitution, and two people.

All candidates will be on the local ballot paper, and one of them (the 'Local MP') will be selected using the 'Single Transferable Vote' system.  325 MPs will be elected this way.  Within each constituency, the total number of first votes for each party will be recorded, along with the percentage of first or second votes for each candidate.

All the first votes given for each party will be added together.  Any party receiving less than 5% of the total first votes will play no further part; the remaining parties will take part in the 'Party MP' allocation - the remaining 325 MPs will be allocated between the remaining parties in proportion to the total first votes they received - so a party receiving just over 10% of the national first vote will be allocated 33 MPs.

Each party will rank its unelected candidates by the percentage of first or second votes they received, and the candidates with the highest percentage will be elected, as many as the party has been allocated.

The effect of this system will be that everybody will have an incentive to vote:

• every vote will count when selecting your Local MP;
• every first and second vote will count when selecting your Party MP; and
• every first vote will count when supporting your party.

The relationship between the MP and the constituency will inevitably change, but I see no way to make everybody's vote count without introducing some change here.  Every constituency will have one 'Local MP', and most constituencies will also have one 'Party MP', but it is likely that a few constituencies will have no Party MPs, and a few will have two.  Whatever the number, every MP will represent their constituency, and every constituent will be able to contact any of their MPs, in the same way they do at present.

Other Thoughts

There are other, less radical, changes which could be made.  There are two obvious candidates.

• Make voting compulsory.  Many countries already do this.  On its own it will not restore people's faith in the system, but in conjunction with the next idea it might help.
• Add another option to every ballot: 'None of the above'.  If 'None' wins the ballot, the election has to be run again, and none of the original candidates may stand.

In the longer term, I believe that a secure electronic voting system is possible, that it could be trusted, and that it would be much faster, cheaper and more reliable than our current paper system.   This needs to be described and defended elsewhere.  But, if electronic voting is possible, then other voting possibilities arise: you could , for example decide to split your vote between different candidates or parties, with each one receiving a different percentage of your vote.  Or perhaps you could make your vote conditional upon the candidate backing a certain proposal, and the candidate then has to choose whether to accept those conditional votes or not.

There are probably many other possibilities which electronic voting would make possible, but the first priority is to get a voting system which people trust and which more accurately reflects their views.  Once that has been agreed, implemented, and shown to function, then we can start to think seriously about improving the system further.

References

[1]  Wikipedia (https://en.wikipedia.org/wiki/Electoral_system)

[2]  The New Scientist provides a readable description of the problem with voting systems (https://www.newscientist.com/article/mg20627581-400-electoral-dysfunction-why-democracy-is-always-unfair/), and Wikipedia provides the theory and mathematics which prove that no voting system can deliver everything we want (https://en.wikipedia.org/wiki/Arrow%27s_impossibility_theorem).

[3]  Wikipedia (https://en.wikipedia.org/wiki/Voter_turnout#Trends_of_decreasing_turnout_since_the_1980s)

[4] LSE: Low voter turnout is clearly a problem, but a much greater worry is the growing inequality of that turnout (https://blogs.lse.ac.uk/politicsandpolicy/look-beneath-the-vote/)

[5] How to solve a constitutional crisis (https://www.newsroom.co.nz/ideasroom/how-to-solve-a-constitutional-crisis)

[6]  New Scientist leader: Democracy needs an upgrade to ensure it keeps people on its side (https://www.newscientist.com/article/mg23531423-500-democracy-needs-an-upgrade-to-ensure-it-keeps-people-on-its-side/)

[7]  New Scientist: Wrong division: How maths can save democracy from gerrymandering (https://www.newscientist.com/article/mg23631520-700-wrong-division-how-maths-can-save-democracy-from-gerrymandering/)

[8]  The Electoral Commission tell you how to register a political party (https://www.electoralcommission.org.uk/who-we-are-and-what-we-do/party-registration-applications/how-political-parties-are-registered)