Change in the electoral system isn’t necessarily ‘reform’

A new post on LabourList provides a useful summary of the various alternatives to our current system of First Past the Post (FPTP) for elections to the house of commons currently being hawked around in much of the vaguely left-of-centre press, especially in the Guardian and the Observer.  Polly Toynbee in particular seems quite unable to write a column without including a commercial for Proportional Representation.  However, the advantages and drawbacks of each of the systems discussed in LabourList seem to me (and to at least one other reader of LabourList who has commented on it) to be somewhat skewed in favour of a change in the electoral system, over-stating some of the arguments in favour and omitting some of those against.

I have appended the following rather lengthy comment to the LabourList post in an effort, probably doomed, to help correct any imbalance.  This can usefully be read, if anyone’s interested, in conjunction with what I have written previously on the subject:

This otherwise useful summary of the various options omits any mention of a (to my mind) crippling defect in all the systems which involve transferring votes: in AV, the only votes that get to be transferred are the second preferences of those whose first preference went to a candidate who has been eliminated (because he/she has come bottom of the list at the first or subsequent recount). In many (most?) cases, the candidates who come bottom of the poll and are eliminated in the early counts are frivolous egotists and exhibitionists, other weirdos, or candidates of far-out extremist parties or causes. It’s not clear why those who give such no-hoper candidates their first preference votes should have their second preferences given so much more weight in the eventual result than those who gave their first preference votes to mainstream candidates, and whose second preferences will never even be counted if their candidate is never eliminated. Votes cast are thus treated unequally, those for eliminated candidates given more weight (by virtue of the redistribution of their second preferences) than those cast for candidates not eliminated. This hardly qualifies as ‘fair’. The same unequal weight objection applies, mutatis mutandis, to the Single Transferable Vote system.

There are in addition serious drawbacks to having more than one MP for each constituency (as STV would entail), notably that it breaks the invaluable convention that a single MP, once elected, represents the interests of all his constituents, not just those who voted for him (or her).

There is also the obvious illogicality in all preference vote redistribution systems of pretending that first and second (or even third) preference votes are of equal weight, simply in order to be able to claim that the winning candidate has had the ‘support’ of the majority of the votes cast, whereas (unless he or she won more than 50% of the votes at the first count, in which case no second preferences need to be redistributed other than to the other candidates in multi-member constituencies) it’s obvious that on the first preference count, which is the only one that accurately reflects voter choice, a majority of the votes were actually cast against the eventual winner. The fact is that nationally, and in most individual constituencies, no single party ever commands the support of 50% or more of the electorate, and this is inevitably reflected in votes cast at elections. Fiddling around with redistributions of votes from one candidate to another can’t magically transform this reality into an apparent overall majority for a particular candidate. A voter who has cast a first preference vote for candidate A can’t meaningfully be said to have voted ‘for’ candidate B, even if B is eventually given his second preference vote when preferences are redistributed. Thus the principal argument generally advanced in favour of redistribution systems such as AV and STV, namely that they ensure that every MP has had the support of a majority of the votes cast in the relevant constituency, is bogus.

The objections to any party list system, an essential element in the horrendous proposals by Roy Jenkins and his Commission, are obvious. They put even more power into the hands of the party apparatchiks than they already have, enabling them alone to decide which politicians are included in the list and which are excluded from it. Guess which independent-minded, maverick members of each party’s awkward squad are going to make it onto the list! The electorate has no say in the matter, even if the list is an ‘open’ one in which voters are allowed to express preferences as between the various names on the party list. This objectionable system is one of the many reasons for the abysmal turnout at elections in the UK to the European parliament, conducted inexplicably under the party list system. The last thing we should want is to introduce it as part of the system for electing members of the house of commons, and thus for electing governments.

But above all the objections to any system that will always deliver coalition governments (as true PR always will) are surely decisive. It will always empower a party which has won fewer seats and votes than either of the two main parties to decide which of those two main parties is to get the keys to No. 10 Downing Street. Both the parties which come in first and second will have to bargain, after the voting has finished, with the minor parties in order to negotiate changes or additions to their election manifestos sufficient to guarantee enough support from one or more of the minority parties to secure a majority for the coalition (whether formal or informal) in the house of commons. Thus the government taking office will have a programme that has been drawn up or finalised only after the election has taken place and for which not a single voter can have voted. The majority partner in the resulting coalition is then permanently at the mercy of its minority partner(s), who can bring the government down at any time, on a whim or following a personality clash, and put the other main party into office instead, without a single voter having any say in the change of government. It’s no good saying this doesn’t happen in practice: it has happened at least once in Germany since the war and is a permanent feature of the system in Israel, where moderate centrist governments are constantly held to ransom by right-wing extremist minority parties on whom they depend for their continuing majority support in parliament.

Finally, a change in the present electoral system (tendentiously referred to as ‘electoral reform’!) is far too momentous a constitutional change to be introduced, or even to be submitted to a national referendum, without the broad agreement of all the main parties across the political spectrum. It certainly should not be imposed by one party on the rest without their agreement, for the temporary political advantage of the party in power. If a referendum is to be held, it should be preceded by a lengthy period of consultation and information in which the pros and cons, especially the cons, of each of the options can be extensively debated and publicised. Simply to ask if one is in favour of electoral ‘reform’ is almost to guarantee a Yes vote, and that’s liable to land us in an even worse mess than we’re in already.

It’s easy to pick holes in First Past the Post, but few commentators in the current febrile atmosphere seem to be willing to point out the even bigger holes in every one of the alternatives. As Churchill said of democracy, FPTP is a terrible system; it’s just that all the others are even worse.

Apologies for the length of this.


15 Responses

  1. Laughing Gravy says:

    The simplest ‘electoral reform’ which retains the advantages of FPTP but also gives legitimacy to every elected candidate, is the run-off election. First election as now, run-off between top two candidates if the top candidate does not gain 50% of the vote on first election. Each elector is empowered on the first election, BUT on the second election, where required, those electors have full information about how their fellow constituents voted and also how the whole country voted. Any other system of vote transfer works on less than full information to the elector. I think this is still used in France and elsewhere.  

    Brian writes:I agree that this has a great deal to commend it — not least because it generally presents voters in the final stage with a simple binary choice between left and right which is the essence of virtually all political choices, however much the romantics of the right may yearn for lots of little splinter parties representing every possible shade of opinion, allowing the ultra-fastidious to satisfy their consciences and think they can evade the brutal choice that in the end has to be made. I suppose there are two possible drawbacks to the second-stage run-off election: the system once, IIRC, came perilously close to landing France with a neo-fascist President, when the Socialist candidate was unexpectedly eliminated in the first round (also see this criticism of the French system): and of course it’s liable to be twice as expensive as a single-round election, generally a decisive objection in our penny-pinching country. All the same, I would certainly include it as an option in any referendum on electoral system change [please let’s not beg the question by calling it ‘reform’].

  2. Neil Harding says:

    Brian, PR advocates need to persuade Labour members like you of the merits of PR, thats if we are going to get this referendum. I feel after 13 years in power we are going to really regret not changing the system when we had the chance – 2 more decades of hard-right Tories having complete control of the country with just 30 something percent of the vote on a 60 percent turnout is scary. In the Fifties, when the main 2 parties could garner 90% of the vote on an 80% turnout, a case could be made for the present system, but not now. Unlike some other supporters of the status quo, you are at least willing to debate, so I am hopeful you can be persuaded.

    As it happens, I share your dislike of preference systems like AV and STV which is why I prefer list PR, but we have to put their flaw into perspective. People having their second and third preferences considered is unfair but it is a minor flaw when we consider under our present system most votes are ignored altogether.

    You say the cons of PR are underplayed, but actually PR is unfairly maligned with plenty of myths in the tabloids and beyond and the actual facts are ignored.

    The facts are, that countries that have had PR continuously for well over half a century now, have better healthcare and education systems, better public transport and infrastructure, less inequality, higher political engagement and a higher standard of living and quality of life. Isn’t this what matters?

    How this supports your assertion of indecisive and poor governance is anyones guess.

    You boldly assert that PR brings instability, but Germany has had less post-war general elections and less post-war leaders, as indeed have most countries that use PR in Western Europe and beyond. Look at the German cabinet and you will find only 2 or 3 ministers posts changing in the last three years, whereas in the UK virtually every cabinet post (including PM) has had different occupants, some ministers have had 4 jobs in the last 3 years. Where is the stability there? If you want to see really unstable government, then go to Canada where they have had 5 general elections in the last 8 years using the same Westminster first-past-the-post system as us.

    As for coalition government not being what people want – giving all the power and big majorities to one party that only gets 35% of the vote is certainly not what people want but thats what fptp delivers.

    Isn’t it better that voters have 5 MPs to choose from that includes who they actually voted for, rather than socialists having to rely on a Tory MP installed for life, trying to making their case to an MP they will never vote for.

    The party has a bad name in the UK because they are so undemocratic here, with just a handful of party members deciding policies – this is a direct result of an electoral system that limits choice for voters. People do not read manifestos and parties rarely abide by them anyway. PR makes parties more democratic and open because people know they go elsewhere and their vote will count. Policies are actually more likely to represent what the majority want. Like I say, explain to me what is so disastrous about German or Scandanavian public services? In fact Harvard studies have demonstrated that PR delivers more of what the majority want and not surprisingly parliaments that are more representative. Not all minority parties are extremist and they only have power in proportion to their support – so the FDP had 20% of cabinet posts because they made up 20% of the coalition. There is nothing wrong with that. It only sounds wierd in the UK, because we are used to small parties having no power whatsoever.

    If you really are left of centre, you have to ask yourself why you are so happy with the present bunch of public school boys running everything?

    Brian writes:I don’t know what makes you think that I’m “so happy” either with our current political leaders (although I don’t think it matters where they went to school — Attlee was a public schoolboy, as many other great liberal reforming leaders have been) or with a great many other things about the way our country is governed. This alleged happiness of mine isn’t reflected, for example, in this blog. I simply don’t see any grounds for claiming that our numerous problems, or the glaring defects in our constitutional arrangements, are connected causally or otherwise with the electoral system that we use for electing MPs and, through them, governments. Similarly, the many successful aspects of the way Germany has been governed for the last few decades seem to me as likely to be in spite of its PR system for electing the Bundestag and government as because of it. After the last German elections, for example, there was no effective government for several months while the parties horse-traded among themselves to try to put together a coalition that could gain majority parliamentary support — and they came up in the end with a Grand Coalition between the two major rival parties which has prevented Angela Merkel from doing anything radical or controversial, because of the extreme difficulty of getting agreement within her government — never mind in parliament — on anything remotely controversial, much as one might expect if we here were to be forced to endure a Conservative-Labour coalition government formed of parties with mainly irreconcilable policies and led by Cameron as prime minister and Brown as his deputy. Shades of Ramsay MacDonald! It’s fortunate for Germany (and the rest of us) that there was no sudden national or international crisis during the months of German government paralysis after the election; and even now, with the Grand Coalition, Germany’s response to the banking and general economic crises has been notably slow and contradictory, certainly compared with our single-party government’s. But coalitions with all their drawbacks and weaknesses are a direct consequence of proportional representation in situations like ours and Germany’s in which no single party ever commands the support of anything like 50% or more of the electorate or of their votes. (Incidentally, I don’t understand your statement that I “boldly assert that PR brings instability” when I don’t remember asserting any such thing, ‘boldly’ or otherwise: perhaps you can quote chapter and verse? Nor do I agree that our damaging habit of shifting ministers around every few weeks in an endless game of musical chairs results from FPTP; it probably reflects the weakness and insecurity of the prime minister.)

    I freely acknowledge the defects in First Past the Post, and agree that it would be unacceptable and inappropriate for elections to a legislative chamber whose main functions were debate and scrutiny of legislation — such as an elected second chamber ought to be — but whose primary fiunction is not to choose the party of government and then hold it to account and if necessary to eject it. Where that’s a chamber’s principal role, as is the case with the house of commons, FPTP seems obviously superior for that role to any alternative. Even then FPTP has its obvious faults and often causes anomalies and distortions. But any form of PR or alternative voting has even more when it comes to producing governments.

    And finally (!), I see that you share my distaste for “preference systems like AV and STV” and would prefer “list PR”. Personally I think any list system would give us the worst of all worlds, by actually increasing the power of the big-party apparatchiks when one of the principal objectives of any sensible reform ought to be to diminish it. And a proportional list system, like any other proportional system, takes us back into the quagmire of coalitions. I just don’t see anything left-of-centre about such a change. Innumerable other things in our system are crying out for progressive reform: why pick on the one thing that works reasonably well, give or take some boundary changes to remove or reduce the disadvantage currently suffered by the Tories (and suffered for years in the past by Labour)?

  3. Neil Harding says:

    The following site has some excellent links that will give you some more facts and answer some of your other points.

    Brian writes:Neil, many thanks for this invaluable link: it’s indeed an extremely useful resource, well worth bookmarking (or Favoritising, as the case may be). One of its links leads eventually to a wonderful summary (here) of the results of the 2005 elections in Germany, under a system of PR that combines both constituency representation and a list system in order to produce a proportionate overall result — much as the Jenkins Commission recommended in its misbegotten report. You need to be a Senior Wrangler in mathematics to follow the various numerical tables explaining how the results are manipulated to achieve this outcome, and the summary of its political implications once the election result has been calculated and published perfectly illustrates the way PR leads to situations that no single voter can have voted for or even envisaged, and which after prolonged horse-trading between the parties will produce a government and a government programme which it’s impossible to predict and also impossible for any voter to influence:

    The ruling coalition of SPD and Greens thus win 273 seats, a net loss of 33, and lose their majority in the Bundestag. But the centre-right
    coalition of the CDU-CSU and FPD win only 287, a net loss of eight, and
    short of a majority (308). The Left thus holds the balance of power. The
    SPD-Green coalition can retain office if the Left supports them, or if
    they can entice the FPD into a three-party coalition. An SPD-CDU-CSU
    “grand coalition” is also possible.

    Grist to the mill for academic political scientists, no doubt, but a grim warning for the rest of us!

  4. Neil Harding says:

    Brian, you make a big deal of coalition government, but seem content with the massive defects of first-past-the-post. That is why I said you were a defender of the status quo.

    The fact remains that most countries in Western Europe function very well with PR systems, indeed most have higher standards of living, higher quality of life, better public services and less inequality. Are ALL these countries doing well despite having PR? Seems very unlikely.

    Harvard published a study on PR which came to the following conclusions on left-of-centre government:-

    “The details of actual tax and spend policies for the purpose of redistribution are complex, but the explanation for redistribution in advanced democracies is probably fairly simple. To a very considerable extent, redistribution is the result of electoral systems and the class coalitions they engender.

    Electoral systems matter because they alter the bargaining power and coalition behavior of groups with different interests. In majoritarian systems, parties have to balance the incentive to capture the median voter with the incentive to pursue the policy preferred by their core constituencies. Because the median voter is closer to the distributive interests of the center-right party, any probability that parties will defect from a median voter platform once elected will make the median voter more likely to vote for the center-right.

    This result contrasts to multiparty PR systems where governments are based on coalitions of class parties. In this context, center parties will tend to find it in their own interest to ally with parties to the left. This result follows because the middle class can use taxation of the rich to bargain a tax rate and benefit level with the poor that is closer to its own preference. There is no opportunity for a coalition of the center and right to exploit the poor in the same manner”.

    As for ‘horse trading’, the Labour party is a coalition that includes Jeremy Corbin and Alan Milburn – the horse trading that goes on behind closed doors in the Labour and Tory party is far more secret and distant from the populace than what happens in the few days (sometimes weeks) after an election result under PR – people can see the joins and generally the negotiations are very public and transparent. Rather than having to vote for an MP that could be well to the left or right of what they want and they have no say on this under fptp. 

    The handover is not a problem either, as happens in the US where the handover takes 3 months, the current incumbents continue in power for a few weeks. Most people in Germany expected a grand coalition and indeed this was supported by a majority in polls both before and after the election. It has actually worked quite well. Germany are likely to emerge far stronger from this recession that the UK will. Germany has actually instigated a bigger stimulus to its economy that the UK.

    I intend to do some research on party manifestos, both in this country and abroad that I am sure will demonstrate that people get more of what they voted for under PR than under our present unaccountable system. Brian start quoting some facts at me, rather than just repeating tired old myths about PR. The big question is; what is so wrong about Scandanavian and German government? You talk as of the world would end.

    As for the power of parties – what you get under PR is far higher party membership, far more power to the members on selection of candidates and anyhow open lists allow the public to order these candidates as they choose – far more choice for the electorate than under fptp. The ‘party’ is not such a dirty word abroad and this is not by coincidence because parties have to listen to their members because it is so easy for members and voters to defect to other parties and know that their vote WILL COUNT and not be wasted as it is under fptp. Maverick MPs are far more common abroad than here, because they can start their own party if they can garner enough support and maybe end up as a junior party in government. This is far healthier than the few accidental mavericks we get here that are now in terminal decline as party leaderships increase their grip on the two main parties. There is no more closed list a system than fptp where even in a landslide election i.e. 1997 only 186 MPs were replaced out of 659. The vast majority of MPs sit in safe seats for life – this is not accountability. The party leadership install their favoured candidates in these seats often outwtting or over-riding a dispirited membership.

    You think a few boundary changes will solve the problems of fptp, but what the Tories are proposing will make things worse – bigger constituencies, more frequent reviews and abolishing geographical and administrative considerations that will make the idea of a ‘constituency link’ a joke. Voters will not know who their MP is from one election to the next, let alone be able to hold them accountable for anything. The problems with fptp run deep, as this gerrymander wheel demonstrates – it is impossible to draw boundaries that are fair, even if you try and the big parties don’t.

    Brian writes:Neil, thanks, but I can’t reply to all your arguments here without repeating myself, so I’ll refer you back to what I have already written in my original post and in responses to your and others’ comments on it — as well as to my earlier Ephems posts about electoral systems. I’ll just mention here that (1) I don’t recognise any of the views that you attribute to me, almost none of which I hold: that (2) I don’t think most of the analysis in the Harvard study which you quote is applicable to the UK, where the class system and group economic interests are radically different from those in the US: and that (3) I see no evidence that either the good things enjoyed by people in countries that use PR or the problems that we experience in the UK are attributable to their or our electoral systems. PR countries experience plenty of problems and we have had plenty of successes in the past, irrespective of the ways in which we and they cast our votes and elect our governments. For the very last time, please stop telling me (and anyone else reading this) that I’m “content with the massive defects of first-past-the-post” when I have repeatedly said the exact opposite.

  5. Neil Harding says:


    Brian writes:I have said and written nothing at all about the effect of different electoral systems on turnout, to the best of my recollection. Do stop misrepresenting my opinions, will you? And these capital letters, equivalent to SHOUTING, are disturbing. I suggest that you wait until you get home in future, if you can’t write in lower case on your mobile!

  6. Laughing Gravy says:

    Thank you for your comment. No system of voting is fault free, so one must ask what are the objectives that wish to be achieved and to what extent does any system achieve them? To some extent this depends on personal preferences. In my own case, in a general election, I want:
    – my representative to have a high degree of legitimacy within the constituency,
    – a government that is stable and not beholden to being blackmailed by minority parties after the election (that is, coalitions before not after the election),
    – a parliament that is, is far as possible, representative of the whole electorate’s opinion in terms of votes cast
    – a simple, easy to understand, system,
    – and one in which electors have as much information as possible before they cast their vote.
    To my mind the run-off election with FPTP gets closest to satisfying these objectives. Of course, there are also objections. But every system has objections – in some cases very strong indeed!

    Brian writes:Thanks again. I agree broadly with your statement of objectives for the electoral system, except that I would put your second point first. I also agree that a two-stage election with a run-off after the first if the first doesn’t produce an overall majority for anyone is probably the best way to achieve them.

  7. Neil Harding says:


    Brian writes:It’s more important that an election result permits the speedy formation of a government able to be held to account for keeping its pre-election manifesto and other policy promises, and able to govern without being blackmailed by minority coalition partners for concessions and compromises as a condition of being able to remain in office, than that the composition of the house of commons should exactly mirror the distribution of votes cast on a particular day up to four and a half years earlier. (And please no more comments in capital letters!)

  8. Neil Harding says:

    lg, 2 round system is better than fptp, maybe even brian could agree on that. A many rounds system would be even better. This is what the tories and labour use to elect their leaders. AV is an attempt to emulate this without need for many elections. I think brian’s defence of fptp is pathetically weak. The so called major flaws in PR are nothing compared to the flaws if fptp. Is it really a bearable problem that the poor and minorities have so little representation. Political science and the evidence from countries that use PR is categorical in suggesting higher turnouts and more representation and equality. Brian chooses to ignore this. Brian seems to not want people to have a say in a referendum that was promised in labour’s manifesto. Lets have the referendum, if we lose, we will shut up, but until then, lets see what the people really want.

    Brian writes: Neil, this repeated misrepresentation of my position, completely ignoring what I have written in I hope reasonably clear language, has gone far enough. No more, thanks. If I can respect your views without agreeing with them, I don’t see why you shouldn’t make the effort to respect mine. And for the very last time: neither Labour nor anyone else ever promised a referendum on the Lisbon treaty, and endlessly repeating the contrary doesn’t make it true. End of story. Finis!

  9. Laughing Gravy says:

    This will be my last comment on this matter. It is true that transfer or preference systems try to simulate multi- elections, BUT the fundamental flaw is that, at the time the preferences are ranked by the voter, he or she has incomplete knowledge about what other voters are doing. The run- off allows those voters still involved to know what others have done and to cast their vote in that knowledge. To make a prospective majority larger or smaller for example. One run-off is sufficient, though theoretically you could have more than one. On a personal note, I first voted in the 1959 election and I have worried about electoral systems ever since. At various times finding logical reasons to support every sytem under the sun, including the weighted vote in parliament system (in this, each MP gets a weighted vote in parliamentary divisions depending on their majority at the election:  50%+ gets one vote, less than 50% a lesser vote). I have come down eventually in favour of the run-off as the least worst system. I doubt very much there will be a referendum on this. It will be completely bogged down in meaningless wrangles about systems, and fairness, and stability and so on. But if I were David Cameron I would offer the run-off as my alternative to AV or AV+ if any referendum were to be held.

    Brian writes: LG [resonant initials!], I think you have made an impregnable case. I agree that an early referendum on such a complex question would be a nightmare. One only needs to read the comments on this blog and over at LabourList to realise just how intricate the arguments are — and how much some people get worked up over the issue. In my view any change in the system would require a Royal Commission to set out the pros and cons of the numerous options and come up with a specific recommendation for parliament (after the next election) to approve or reject. (The Jenkins Commission totally failed to do this, incidentally.) If parliament approved it, it should only then be put to a national referendum for a simple yes or no. But as Cameron is likely to win the election and as he’s against any change from FPTP (no doubt for all the wrong reasons), none of this is at all likely to happen, in which case this whole debate is probably academic. Interesting, though!

  10. Neil Harding says:

    I posted to apologise if I had been confrontational, but that comment has not appeared. I assume you have banned me from the site. Very sad. You surely must know I meant the promise of a referendum on electoral reform not Lisbon, and you did say that PR was the cause for the poor turnout in the Euros when turnout was abysmal before PR and already falling. PR has reversed this. Anyway, you are clearly not interested in the facts and you demonstrate your democratic credentials by banning me from this site, if indeed that is what you have done.

    Brian writes: I treated your post as a private message (it was sent from your mobile and was quite difficult to decipher) and I replied by e-mail, saying that I too apologised if I had been ratty in some of my responses to your comments. I’m sorry that once again you have reverted to personal invective, a substitute for debate from which the great majority of those contributing comments to this blog seem able to refrain. It doesn’t strengthen your case. As you say, it’s “very sad”. Goodbye!

  11. Alun lloyd says:

    I very much agree with many of the points you have made against PR, although I do believe there is a place for it. However, that place is definitely not in electing a government. My father-in-law has a pithy riposte to PR that sums up your objections: PR gave us Hitler. A nice simplification of a complex time that will probably have AJP Taylor spinning in his grave. Nevertheless, a lesson in the worst case scenario of extremists holding the balance of power. There needs to be winner and everybody must be clear who that winner is. FPTP is the only real way and the choice usually comes down to Right or Left and the shading is irrelevant.
                I understand your objections to vote redistribution but they presuppose that people actually vote for their first choice candidate in the first place. Tactical voting is the norm nowadays. How many voters who see themselves as Socialist Worker supporters vote Labour as they are the closest to their view with a chance of winning, how many UKIP vote Conservative? How many Greens will pick the party with the most environmentally aware policies? People are already redistributing their votes. Furthermore, you appear to lean towards ‘run-off’ elections as a possible alternative, but surely that is just AV by another name. Those who did not vote for the top two last time around now get to choose which they prefer (how many who voted top two before will change do you think?). Why bother with the expense? Why not ask them in the first ballot. I take on board your point about the candidates who turn out bottom tending to be of a dubious provenance, but they will stand somewhere on the right/left political spectrum. Vote distribution would coalesce the votes into a true reflection of where a constituency stands on that spectrum. Most of these candidate’s supporters are well aware of their chances and might actually value a chance to register preferences to indicate where they are on the spectrum. In simplified terms if in a three way constituency the votes go Con 40% Lab 35% Lib 25% the Tories get elected even though 60% of the voters are clearly to left of them. With AV they may still win if the Liberal voters believe them to be closer policy wise to themselves and the constituency will be shown to be right leaning; a fair reflection.
                I favour AV but only in the context that the first ballot should actually mean something, otherwise it is totally pointless. The trick is in finding that meaning and making it relevant. The only real way to make it relevant is to use the vote in a PR context. I appear to have made a circular argument here: PR should not be used for electing governments, AV is only relevant if there is a PR context therefore, AV should not be used for electing governments. However, what if you used the results of the first ballot as a basis for electing something else; the second chamber maybe. I can already see the dreaded party list system galloping over the horizon of your mind. So let me set out my full 10 point electoral reform plan.
    1. Compulsory voting for parliamentary elections (see the way I slipped that in without any notice) with the inclusion of an abstention.
    2. Voters are free to register as many preferences they like but an abstention stops the vote. (So you vote for the BNP candidate 1st then abstain 2nd and second time round your vote goes into the abstention pile)
    3. All candidates polling less than 10% of the vote are eliminated first time around and their votes transferred, from then on the lowest remaining in the poll until an outright winner is found.
    4. If abstentions win the candidate coming second gets returned but is constitutionally bound to abstain in any vote and a by-election is held within 6 months with an all new set of candidates (I’d possibly give the candidate returned a second chance after having 6 months to make a good impression). This should introduce a bit of jeopardy and hopefully encourage the parties to try and engage the electorate much more.
    5. A second chamber of 200 will be appointed 3 months after the general election.
    6. All the votes in the initial ballot are totalled up and each party awarded a percentage of the vote which is rounded down to the nearest half a percent. Each party is given one appointee to the second chamber per half percent. Owing to the rounding down process, the abstentions and others below 0.5% the number will not be 200. The balance will be appointed independently (not sure how).
    7. There should be tight rules on who can be appointed to the second chamber. Former MPs can make up only 30% of the appointees up to a maximum of 20 but cannot include any MPs voted out of office at the election. There should be a fit and proper test for appointees conducted by an independent body. There should be quotas for expert appointees like lawyers, medical professionals, economists, educationalists and scientists which the chamber as a whole must conform to. The trick will be for the parties to appoint people that satisfy the electorate and ensure that they are actually able to do the job. This should not be jobs for the boys.
    8. A minimum standard will be set for participation any member not achieving the standard will be rejected at the next appointment period.
    9. The second chamber will be free of the whip but its main purpose of debating and scrutinising legislation would remain.
    10. The members of the second chamber should be term limited.
    So what we end up with is a first-past-the-post parliamentary election plus a PR election for the second chamber. A two in one deal that will save on cost and alleviate the dreaded election fatigue. The best of both worlds surely?
    Brian writes: Absolutely: PR is inappropriate for electing governments (the main function of elections to the house of commons) but well suited to an elected second chamber. However I would prefer the second chamber to become a House of the Four Nations, i.e. a federal Senate, with an equal number of members from each of the nations of England, Scotland, Wales and Northern Ireland, elected by PR by each of them. The US and Australian Senates provide good models, mutatis of course mutandis. Equality of representation regardless of population size would protect the three smaller nations from domination by England. At present even devolution to the smaller three doesn’t give them adequate protection. But that’s really for discussion in another post.

  12. Peter Harvey says:


    I can’t let you get away with your comments about Germany. The country has been remarkably well, stably, and competently governed for the last sixty years under a constitution that was, be it remembered, imposed by the Allies after WWII.

    Yes, they sometimes take their time to get a new government in office. Or to put it another way, the UK alone in Europe acts with unseemly haste in sending in the removal vans to Downing St the day after the election on the grounds that Her Majesty must at all times have a strong and powerful one-party Government to keep the peasants under control. Other countries spend a little time to digest the election results and think about what to do; some even have a minimum time after the election in which the government can be formed. The previous government remains in office in an acting capacity, with a clear definition of what it can and cannot do. And what it can do includes responding to international financial crises. It works, in Germany and many other countries.

    The present grand coalition is not the first in the Federal Republic; there was one in the sixties. It is not fair to liken it to a Labour-Tory coalition. The German SDP is a perfectly sane Social Democratic party that accepts the social market economy without a qualm, while the ultramontane advocates of Stalinist state control are to be found in a different party entirely. The same goes for Spain, and for France in theory but it is impossible to say anything sensible at all about the French Left nowadays. Portugal, too. In fact, it is said that the disputes in Portugal after the 1975 carnation revolution were the moment in European politics when the Mensheviks finally beat the Bolsheviks. And the Tories by their own admission do not wish to be associated with the European Christian Democrats, who are indeed most certainly not Tories anyway: by and large they show signs of having been in contact with civilisation.

    Brian writes: I don’t have any quarrel with anything you say, Peter, apart from the implied suggestion that I have failed to recognise Germany’s impressive record of good government over the last many decades. I just don’t think German successes can be attributed to its electoral system. The action of Germany’s then third party, the FDP, for decades holding the balance of power after every Budestag election, in withdrawing from its coalition with the SDP in the autumn of 1982 and replacing Chancellor Helmut Schmidt with the CDU’s Helmut Kohl without the benefit of an election to legitimate the change of government, was hardly a good advertisement for PR and its necessary concomitant, permanent coalition government:

    The 1976 election revealed the first decline in SPD support since 1957. The SPD government would not be in power much longer. As chancellor, Schmidt worked well under pressure, making tough decisions about the recession in the early 1980s and the increasing oil prices. Although Schmidt was a great policy maker, he did not cooperate with the leadership in his own party. Fed up with Schmidt’s independent dealings, and weary of the SPD’s declining popularity, the FDP withdrew its coalition agreement with the SPD in the fall of 1982. No longer holding a majority in the Bundestag without their coalition, the Schmidt government fell apart. The CDU/CSU, in the decade preceding the 1983 election, had been rebuilding their party leadership and were able to grasp this opportunity whole-heartedly. The FDP then helped the CDU/CSU replace Schmidt with Kohl and by the 1983 election, the CDU had all the benefits of an incumbent government without the accountability.

    I agree that changes of British prime minister take place with unseemly haste. On the other hand, you might agree that the 3-month delay in forming a coalition and hence a government after Germany’s 2005 elections was really too long for comfort. In the UK, with very little experience of negotiating with minority parties to form a governing coalition, the delays after each hung parliament emerged from every election under PR might well be even longer, with government effectively paralysed during the interregnum. In short, I think a good case can be made for saying that the good and stable government which Germany has enjoyed for most of the time in recent decades has been in spite, not because, of its electoral system. Stability has also been reinforced by Germany’s fully-fledged federal system, with the Lander effectively governing themselves without constant intervention from the federal government and Budestag in Berlin (or previously in Bonn). As I have boringly and persistently argued elsewhere, we could usefully emulate German decentralisation by completing the devolution process and moving to a full federation of the four nations of the United Kingdom. PR would then be fine — for the new federal Senate!

  13. Neil Harding says:

    Alan, saying the Nazis came to power because of PR is as silly as saying Labour came to power BECAUSE of first-past-the-post. Labour went from being a minor party in 1920 to goverment in 1929 and it would have happened under any system. Labour had a faster rise to power than the Nazis AND on a lower percentage of the vote. These were turbulent times and not surprisingly there was radical social change.

    The Nazis were a product of economic meltdown coupled with resentment at unjust war reparation payment and conditions. The German bourgoisie and US banks bankrolled the Nazis through fear of communism.

    Even then, the Nazis on 40% of the vote needed the support of the German Conservative party to achieve power – so PR probably made it MORE difficult for the Nazis to achieve power. Under fptp they would have won every seat in the German Parliament.

    PR has been operating in most western european countries for over 50 years now, and of the major countries it is the UK and France that have the biggest problems with the far right. Neither country uses PR for their main parliament.

  14. Peter Harvey says:


    Yes, the German Länder do govern themselves, but they also are elected by the same proportional system and thus often have coalition governments.

    It is unusual for the lead party in a government to change without an election. About as unusual as it is for the UK to change its PM without an election.

    Brian writes: You really can’t equate a change of prime minister where both are succesive leaders of the same party — technically no more significant than a change of home secretaries or ministers of energy where both predecessor and successor are from the same party — with a change of government from a government of the left to a government of the right, with no election to legitimise it. We elect a party to form a government, not a specific prime minister. Under the German system the minority party which holds the balance of power under PR and thus can decide which of the two bigger parties is to form the government may only rarely use this power actually to kick out one elected party from government and put in the other instead: the point is that they can credibly threaten to do this at any time as a means of blackmailing their senior coalition partner into making policy concessions to a party with perhaps half the popular support that the senior partner had at the preceding election. That seems to me manifestly undemocratic, and irreconcilable with the claim that PR gives the voters greater power over decisions about who governs them.

    As for the Länder, my point was simply that some of Germany’s record of mainly successful government in recent decades was probably due at least in part to the fact that with the Länder in effect self-governing, Germany escapes the toxic over-centralisation that plagues us in Britain. The electoral system in the Länder is irrelevant to that point.

  15. Laughing Gravy says:

    I did say I would not comment again, but the discussion has become too interesting to stand back from. At the end of the day, you have to decide what objectives you wish to attain and your priority on those objectives. I think it was Professor Arrow who won a Nobel Prize for showing the impossibility of the general reconciling of competing objectives, when the rankers of those objectives have different priorities in their rankings. The issue of voting systems is a perfect example of how this is true, and no matter how many examples of good governance with PR system are produced, it is always possible to produce examples of bad governance. I would cite Israel as a country in which PR has served it badly (a subjective view I readily admit). Now then of course, the argument moves to the form of PR. For these reasons, I am pessimistic that there will ever be much change in what we have now.  I support the run-off because it closest to my objectives. It is a relatively simple step to take from where we are now  and, therefore, might just possibly might make it through the mass of competing political interests. Anything more daring or more complex has no chance. Sorry Alun, I admire the ingenuity of your scheme but it has no chance of being enacted