Approval Voting and the UK General Election
A whole weekend has passed since the election and the British public still await conclusive decisions on the nature of the cabinet that will be formed. I note at this point that the discussions in the press imply that we're pending a decision on who will "run the country" which is simply not the case. With 649 out of 650 seats decided, I would say we already have a pretty good idea of who will be debating and voting on legislation.
Overall Majority
That's not to say the cabinet's role isn't important, but the real decision is whether the party that forms it commands an absolute majority of the house or not. With 326 or more seats held, a party all voting in unison (either by conscience or party whip) will always be successful in passing laws through the House of Commons.
This was particularly evident while Labour lead a very strong majority and was able to pass several items that were opposed by the other parties. These include:
- 42 day detention of suspects without charge
- Foundation hospitals
- Third runway at Heathrow
- Digital Economy Bill
The Case for Minority
There are myths circulating that a minority government is somehow weak. This ignores the fact that opposing parties don't necessarily vote against everything. Labour would still have executed the occupation of Iraq even with a minority due to support from the Tories.
Also, it could be argued that a majority government is more dangerous in that it need only declare a three line whip and it is able to pass anything it likes through the house.
Of course, we've not even touched on the fact that all legislation then has to go through to the unelected House of Lords. The primary case for retaining said assembly of unelected peers at all is that it counterbalances the House of Commons by making sure it can't pass unreasonable laws. I would suggest that this counterbalance is only needed because of our obsession that a party should absolute majority in the first place.
A parliament made up of parties with no overall majority would only pass legislation that was aligned with the consciences of over half the MPs. Accepting the premise that MPs are representative of the will of the people, this indirectly means that all passed legislation is backed by over half the population.
First-past-the post does not reflect approval
The First-past-the-post system simply fails when there are more than two candidates. This is the core truth behind the Liberal Democrats' policy to replace it. It is entirely possible that a party whose candidates consistently lose by only 10 votes will get no more seats in parliament than a party that gets zero votes in all but one seat in which they win by a small margin.
It is also the case that the party that goes on to form an overall majority has a relatively poor mandate from the public. Consider the following two elections:
- In 1979, Thatcher became Prime Minister with the Tories getting 43.9% of the popular vote.
- In 1997, Blair formed a government after popular support for the Labour party from 43.2% of the public.
These two results were considered major landslide victories, yet
the party that went on to form the government
Where this problem actually arises is in the fact that some people might well have supported Labour, say, in the 1997 election, but preferred the Liberal Democrats. This means the amount of people happy for a Labour prime minister may actually have been higher than as indicated at the polls -- albeit as second choice rather than first.
Now, here is where it becomes a real problem. For the sake of demonstrating the numbers, let us assume that all those that voted for the Liberal party (as the Liberal Democrats were then called) in the 1979 election would also have support Labour as a second choice, but would never support a Conservative Government. This is overly simplistic, but not too much of a stretch given their political leanings.
In this case, that means 50.7% of the public would have
This is all due to the basis of first-past-the-post being
that the candidates are in some ways politically orthogonal -- they
are somehow all equidistant in some political hyperspace.
Less mathematically, it suggests that if I support the Liberal
Democrats, I disagree equally with Labour and the Conservatives.
In essence, my vote for one party is an implied
This is very evident in the USA where the Green Party is said to "steal" votes from the Democrats more than it does from the Republicans. The fact that a large proportion of Green supporters would choose Democrat over Republican (because the former is closer in the political spectrum) is not taken into account. Thus a Republican candidate can win the race despite a lower approval rate than the Democrat one.
Alternative Vote
The solution to this proposed by Labour in their 2010 manifesto and recently offered in the form of a bait-and-switch to the Liberal Democrats by the Tories, is the Alternative Vote system.
The gist of this system is that you rank your candidates by preference. A candidate must then secure 50% or more of the "first choice" votes else the one with the fewest is eliminated, with the second choices of those ballots then applied candidates still in the running. This is repeated until someone commands support from more than half the electorate.
My primary objection to this system is the added complexity. As simple as it sounds to mark candidates in order of preference with a number, I know from much experience in the IT industry that adding complexity to a system will bring user error. Even user error that would seem very unlikely happens more than it should in any system.
Approval Voting
Some of the issues behind the "one man, one vote" mantra are discussed in The Arithmetic of Voting. The article demonstrates that just about any system whereby people vote for only one candidate is flawed and concludes that best solution is Approval Voting.
Approval voting allows a voter to mark as many candidates as they like, with no more than one vote for each candidate. This essentially means that you are indicating the candidates you would approve being elected. Some numerical examples are presented in the aforementioned article and I recommend that those curious have read through them.
The elegance of this system amounts to two things. The first is that it is a very simple extension of the current ballot paper. There is no confusion over employing different markings such as numbers. The other is that it simulates the approval rating measure applied to a leader in office. Rather than ask "do you approve of the job the prime minister is doing?" it asks "would you approve of the job this MP would do?".
The winner is then determined by the most number of crosses received. The end result may still be arrived at by a first-past-the-post approach or proportional representation. Thus I am a proponent of PR for use in the UK as well as approval voting; they are not mutually exclusive. One thing to note is that candidates' votes will not sum to 100%.
Testing the System
After my first read of "The Arithmetic of Voting", I immediately
pictured the process as like being presented each candidate in turn
and asked "Do you like this candidate?". My imagination then ran
with the idea and thought it would be interesting to run such
a voting system on
The result was building the website Do You Like it? on which you can browse over one million things and mark your approval or lack thereof. The results so far allow me to make official declarations on what is the most liked thing in the world. This is currently stick figures according to the list of Top 100 things.
I also categorised things to some extent and in keeping with election season I went as far to include a list of most approved UK political parties. Perhaps if enough people voted, it would prove its worth as a better electoral system.