28 May 2015

Flexibility Is The Key To A Good Voting System

My proposal is that every elector gets one vote for one candidate and the top candidates in votes are elected to the seats. So if there are 16 MPs in a county, the 16 candidates with the most votes are elected.

The more I look into the practicalities of this, the more I like it.

It is very proportional and very easy to understand. The candidates with the most votes are elected. No complicated formulas or party lists. No safe seats, expensive boundary reviews or 2 classes of MP. And most of all no silly results with parties amassing millions of votes for little or no MPs.

The clever part is how this system deals with "split votes". Which is at the root of the problem for most systems.

Rather than the smaller parties having the largest burden of split votes and thereby being completely shut out of parliament, every party has to manage the problem of potential split votes or face losing seats.

And it is easily managed by parties by them adjusting the number of candidates that they stand and by having candidates target local areas. Both good for the working of the system.

Thus providing ultra local candidates to those voters that vote for local candidates and keeping numbers of candidates on the ballot paper down to manageable levels.

Also, the voter knows that EVERY candidate on the ballot paper has a real chance of being elected. This is REAL choice. Wasted votes are kept to a minimum because all parties have to aim to minimise surplus votes and only stand candidates with a real chance of election.

No hoper candidates would harm other candidates in the party by wasting party support.

Also the proportionality and the fairly low threshold give independent candidates a chance, but only if they pass the strict entry criteria - 500 electors in the county having donated at least a pound to their campaign in the 12 months before the election.

This also makes it difficult for the larger parties to just "parachute" in candidates from elsewhere. They will need to have built up local support in the preceding 12 months to garner enough support for every party candidate to make the ballot paper.

Easy enough if you are a serious candidate with decent levels of support but this will stop joke and time waster candidates clogging up the ballot paper.

It cannot be right that all you need is a spare £500 deposit to stand. This purely money based system that we have now goes against the principle of democracy. By having to garner 500 real electors willing to donate £1 to a campaign, it will demonstrate really strong support for a candidate before they get on the ballot paper. There might even be an argument for making this donation threshold even higher, maybe 1000 donations. The idea of setting the minimum donation at £1 is to make it affordable for all electors to back a candidate.

Thereby everyone on the ballot paper will have a strong chance of being elected. To be elected you need to get approximately 4% of the vote or more (depending on the size of the county and how the vote splits this threshold might be as high as 8%).

It is the "self correcting" aspect that is the genius part of the system. It is the flexibility of the system that is its strength, allowing voters to decide the rules of the system rather than faceless bureaucrats. Voters decide how local the candidates will be and also the areas they represent within each county.

In Sussex, to guarantee finishing in the top 16 places and become an MP, you'd need 5.88% of the county vote. (1/17 Thanks for correction, Christine. See comments below)

This is the optimum vote. If you get above this you're elected. If a candidate gets far more than this they are safely elected. But whereas safe seats in FPTP are useful for a party, too many surplus votes in this system can damage a party's chances by wasting party votes that could have elected other candidates.

There will be wider choice for the voter with around 17 to 23 candidates on the ballot paper. This is still a manageable level. You can have this many candidates in current byelections.

Each party would likely stand one more candidate than they think they would win in seats. And there would likely be at least one independent stand.

In a 16 seat county, that might mean 23 candidates. Very unlikely to be more than that, because parties would avoid wasting votes.

Parties would judge support by opinion polls. I know polls infamously can be wrong, but they are accurate enough for this purpose (within 6% accuracy).

Voters would be able to choose from ultra local candidates from the more popular parties in the local area or for smaller parties maybe standing for a wider area in the county. Independents might be ultra local or wider. It all depends on size and concentration of support.

Officials wouldn't have to agonise over which streets to include in some arbitrary boundaries, it would all be decided by voters and parties responding to that and boundaries would be invisible and flexible.

All the officials would do is set the number of seats in a county according to the population. This would also mean poorer areas with low voter registration wouldn't be shortchanged on representation.

If we look at our Sussex example using party votes from the general election projected onto likely candidates we can see that the vote needed to be elected is lower than 6.25%.

This is because votes for party candidates are not going to be perfectly spread.

In Sussex the 1st place candidate would be elected with 6.75% of the vote and the 16th placed candidate elected with 3.82%.

The 6 candidates falling below this figure were unelected with combined totals of 18.4% of the vote. These votes elected no-one, so could be called the "wasted" votes under my system.

If we now compare these figures with the percentage of the county vote the current Sussex MPs have under first-past-the-post.

These varied from the highest vote for an MP of 4.07% (Peter Bottomley) to the lowest elected at 2.19% (Simon Kirby) of the Sussex vote.

The total votes not electing anyone in Sussex under first-past-the-post are 50.31%. So the majority of votes could be said to be wasted votes.

So, we can see that under my proposed system, MPs are elected with more votes and there are less wasted votes. We also get a highly proportional result. The Tories in Sussex with 48% of the vote, get 50% of the seats, (whereas they get 88% of the seats under FPTP). LABOUR 19%,19% (6%). UKIP 14%, 13% (0%). LIBDEMs 11%, 13%, (0%). GRNS 7%, 6%, (6%).

So, voters could vote for an ultra local Tory or Labour candidate, or maybe a UKIP, LibDem or Green candidate covering a wider area. The choice is for the voter. Their votes will determine how many candidates are local.

The election will always be close between the election of a candidate in 16th place and non elected place immediately below this - a fraction of a percent. So every candidate can be considered worth voting for.

But if you live in say Worthing, you are probably very likely to vote for a candidate covering your area rather than say a Brighton and Hove candidate and vice versa. Although you could vote cross county if you liked that candidate. There is that freedom.

It would be interesting to see this system in practice. I'm confident it would work. Even if there was a lot of cross county voting and wide disparities in votes between candidates in the same party. It would still deliver reasonable proportionality and legitimacy. If a candidate does get a low vote and fails to get elected, whose fault is that?

This system is fair, simple to understand and still provides local MPs if voters vote that way. In short there is still a direct link between elected candidates and voters. It would be cheaper to run than our present system and it provides proportional results without 2 classes of MP.

Indeed MPs are likely to be elected with more votes than now and there will be less difference between their support.

Nationally at the moment MPs are elected with anything from 12,000 votes to over 40,000. In my system it is unlikely any MP will be elected with less than 25,000 votes. In my Sussex example the lowest would be 32,265 votes to garner the 16th and final elected place.

So, spread the word. Lets elect our MPs county by county by placing an X next to our favoured candidate. Simple and easy. A "Fairest Past The Post" system.


  1. "In Sussex, to guarantee finishing in the top 16 places and become an MP, you'd need 6.25% of the county vote."
    NO - it would be anything over 1/17 of the total: >5.88235%

  2. Indeed, you are right. I stand corrected. In practise it will be lower than this, as the vote is unlikely to be evenly spread.