9

With the election results in the UK election finishing up, a lot of people have pointed out that the number of seats some of the parties got in parliament is far disproportionate to their vote share. Thus, ukip got around 10% of the vote (or around 4 million votes), but only 1 seat, whereas the SNp got 56 seats with millions fewer. In response to this, various people have called for electoral reform. What reasons are there not to change the system?

5

I think the positive arguments are really a mix.

  • The current UK system guarantees geographical representation. It is more difficult to get this with PR.
  • The current UK system has a clearly identifiable person who represents a clearly specified set of people. Those people can fire (not re-elect) their MP if they think they are not doing a good job.
  • The current UK system has historically given a majority to one party. This has various advantages over coalition politics. In my view perhaps the most important is that the electorate knows exactly what they are voting for. In coalition politics the policies of the new government are decided after the election.
  • Some people simply don't want extreme minority views to be represented at all in parliament. For example, PR would have given MPs to the BNP historically (although not in the 2015 election I believe). The first past the post system ensures that only views that are strongly held by a significant local population are represented in parliament.
4

In addition to the argument Michael Broughton has put above, that the MP is the clear first choice of his or her local area, another argument for FPTP is that it usually (not always!) gives a decisive result quickly. Under more proportional systems there is often complex haggling between several parties before a government is formed, involving dramatic concessions to smaller parties in order to get their support.

  • That actually makes sense, especially seeing the amount of trouble that just went into the Israeli coalition – ewkochin May 8 '15 at 15:48
3

We have that same argument in Canada every election. The argument FOR the system is that, from a regional perspective, the representative did indeed win the most local votes to represent their community. To usurp that by any proportional system removes true local representation. the question then becomes whether local representation is more important than overall proportionality. The answer to that depends on how much local representation can truly affect things like local payments from public coffers, and how independant your local voice is within their party. In Canada it has got really bad where the parties almost always vote as blocks - no matter what. In that case, first past the post fails miserably.

  • 2
    But even in just the local sense, you can end up with a majority of the voters voting for parties that don't get anything.... – ewkochin May 8 '15 at 15:47
  • That is true, but democracy does not automatically imply majority except in two-party elections. It implies a plurality, which is not the same thing. – Michael Broughton May 8 '15 at 18:57
  • An approach which would combine the advantages of both PR and FPtP, but which might be hard to "sell" people on, would be to use use a probabilistic method to select among the top vote getters, such that the relative probability of someone winning would be a function of their vote total. Thus, if one candidate gets 60% of the vote and another candidate 40%, the first candidate should have a higher probability of winning, but the second candidate's probability should be non-zero. Such an approach would have the advantage of giving candidates an incentive to get as many votes as they can... – supercat May 8 '15 at 19:46
  • ...even if they had no fear of getting less than 51% of the vote or no hope of getting more than 49%. In a large assembly, some seats would be swayed by chance, but the likelihood of chance skewing things particularly badly would be remote. Otherwise, the statistical average results would combine the locality advantages of FPtP with distributions closer to PR (how close they are would depend upon the mapping between vote percentages and probabilities). – supercat May 8 '15 at 19:50
2

One argument missing from previous answers is the simplicity of the system. How individual votes contribute to the final result is transparent and you only need one round of voting (as opposed to the French system which is still simple and strongly majority-oriented while giving smaller parties a bit more space but requires two rounds of voting – with one or two weeks in-between – for each election). Many alternative systems require complex calculations (e.g. the Irish STV) or lend themselves to strange paradoxes (in Germany, there have been elections in which supporters of a given party in a specific district could have ensured an extra seat for their party of choice by not voting at all).

Note that the almost complete lack of representation for UKIP and other small parties are the clearest evidence for the issues with FPTP. Because the SNP has a local stronghold, it would be very successful in any system. Based on its proportion of the national vote, it could expect at least 30 seats, which is also roughly half the seats for Scotland where it had about 50% of the vote. That's less than the 56 seats it was able to secure in the current system but still quite a lot.

Note that there is a common counter-argument to the simplicity argument, namely that it does not matter whether voters understand how the seats are assigned, as long as the parliament ultimately reflects their preferences.

  • How does the German issue work out? – ewkochin May 9 '15 at 22:16
  • 1
    @ewkochin It's a bit difficult to explain. First a bit of background. On the whole, the system is mostly PR but half of the MP are elected in a kind of FPTP system, in a constituency. Voters have two “votes” and a ballot looks like that. The second column represents what's called the “second vote” (the most important one, incidentally) and determines the overall share of seats a party gets. The first column is for the constituency's directly-elected representative. – Relaxed May 9 '15 at 23:31
  • Directly-elected MP are first assigned to their party's overall contingent (which is determined by the PR part of the system). But if there are too many directly elected MP in a province (compared to their party's share of all the seats for that province), they are elected anyway and get what's called an “overhang mandate” (i.e. an extra seat, the exact size of the parliament is not fixed before the election). – Relaxed May 9 '15 at 23:39
  • Now, the trick is that the number of seats for a province isn't fixed in advance either. It's assigned based on the total number of valid votes in each province. A party's number of seats is determined at the national level and then shared between the provinces, to be filled first with the directly-elected MPs and then by an extra list of candidates. If you live in a province with many directly-elected MP, your vote does not help elect them (they are elected anyway). But it can mean your province gets more seat to the detriment of other provinces where your party could win seats. – Relaxed May 9 '15 at 23:51
  • Many things need to happen together for this to happen but it has, several times, and the law was changed to address it, making the system even more complicated. I think it can still happen in other elections. – Relaxed May 9 '15 at 23:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.