Myth Busting: The winner should win

A frequent objection to systems other than the old First Past The Post system is that “the winner does not always win”.  I think we need to make sure we understand what we mean by “winner” and “win”.

This issue arises at two levels: the constituency level and the national level.

The Constituency Level

In First Past The Post “the winner” is the candidate who gets the most votes.  Where there are only two candidates, the winner will have more than 50% support and is indisputably the winner.  However as the number of candidates increases the level of support at which you could win drops.

  • In a three candidate election the victor could have as little as 34% support (the other two having 33% each).  In many English parliamentary elections MPs are elected with around 40% support.
  • In a four candidate election the victor could have as little as 26% support (the other three having a little less than 25% each).  In many Scottish parliamentary elections MSPs could be elected with around 30% support – except they don’t use First Past the Post.

There is an issue of legitimacy.  To take an extreme example that could be recognised by most (but which might alienate a few), imagine the following result:

Candidate Share Count1
Candidate A 25% 10,000 votes
Candidate B 24%  9,600 votes
Candidate C 23%  9,200 votes
Candidate D 20%  8,000 votes
Candidate E  8%  3,200 votes
Total           40,000 votes

Candidate A is “clearly the winner”? Well if Candidate A was Labour or Conservative many may wish to argue that.  But, what if Candidate A was the BNP, Candidate B the Conservative and Candidate E UKIP?  The BNP has topped the poll, but there is clearly a non-BNP majority, and arguably UKIP has split the Tory vote.  So what do we mean by “winner” and should the “winner” of the poll, “win” the seat?  Which candidate best represents the voters in that constituency?

The Alternative Vote System addresses this particular issue.  Under this system voters express preferences and for a candidate to be elected they have to achieve 50% of the vote.  If no one achieved 50% the bottom candidate (in this example candidate E) is eliminated and their votes transferred to the other candidates according to the second preferences on the ballot papers initially cast for Candidate E.  So the example above may evolve as follows:

Candidate Share Count1           Transfer        Count2
Candidate A 25% 10,000 votes +   100 2nd prefs = 10,100
Candidate B 24%  9,600 votes + 2,700 2nd prefs = 12,300
Candidate C 23%  9,200 votes +   150 2nd prefs =  9,350
Candidate D 20%  8,000 votes +   250 2nd prefs =  8,250
Candidate E  8%  3,200 votes eliminated
Total           40,000         3,200 transferred 40,000

No candidate yet has 50% (20,000 votes in this example), so Candidate D (who is now bottom of the poll) is eliminated and his votes transferred according to the next preference on each ballot paper.  How the next two counts may go is shown below.

Candidate Share Count1  Transfer Count2  Transfer Count3   Transfer Count4
Candidate A 25% 10,000 +   100 = 10,100 +    50 = 10,150 +    550 = 10,700
Candidate B 24%  9,600 + 2,700 = 12,300 + 5,000 = 17,300 + 12,000 = 29,300
Candidate C 23%  9,200 +   150 =  9,350 + 3,200 = 12,550 eliminated
Candidate D 20%  8,000 +   250 =  8,250 eliminated
Candidate E  8%  3,200 eliminated
Check Totals    40,000   3,200   40,000   8,250   40,000   12,550   40,000

One’s view of the “fairness” of this method of determining the “winner” (Candidate B in the above example) might depend on which party candidates A and B are drawn from.  In effect the supporters of Candidates B, C, D, & E have ganged up to prevent the election of Candidate A saying that they would prefer someone other than Candidate A.  If Candidate A was the BNP, many might accept this, but if Candidate A was either Conservative or Labour, supporters of those parties might object.

This is relativism!  But it does force the question: which candidate is the most representative, and is “most representative” a suitable criteria for election – or should the biggest minority take all?

The National Level

At national level, there are similar issues.

  • Should a party that wins 36% of the popular vote be able to take total power, particularly if they do not even have a majority of the seats?  According to some of the press, yes, the biggest minority should be allowed to take absolute control – and this view was repeated by some people in vox pops conducted by the BBC who said “they have a majority – they should be allowed to govern”. (See May 2010)
  • Should a party that came third, be allowed a sniff of power by combining with the biggest minority to create a majority voting block but with a compromised programme? (See May 2010)
  • Should a governing party that gains most votes in a General Election but ends up losing seats, hand over power to a party that has fewer votes? (See Feb 1974)

The current First Past The Post system can produce some very odd results, and it is necessary to address the above sort of anomalies when trying to consider who “won” (if indeed we feel that we can declare a “winner”).

Possibly it is a national characteristic.  How does the British “Sense of Fair Play” actually react to these sort of “hung situations” – particularly if we can strip out tribal loyalties from such considerations?

Post a comment or leave a trackback: Trackback URL.


Leave a Reply

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

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

Google photo

You are commenting using your Google 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 )

Connecting to %s

%d bloggers like this: