The Proportional Representation myth and the German elections

The German elections due tomorrow, 18 September 2005, are widely regarded as a contest between the free marketeers and labour market liberalisers under Angela Merkel and the more traditional welfare staters, more inclined to protectionism, under the present Chancellor, Gerhard Schroeder (whose attempts at reform have been largely resisted by his Social Democratic Party).  Ms Merkel is tentatively expected to gain a plurality of the votes, but not necessarily an overall majority.  In a British-type electoral system of First Past the Post (FPTP), she would probably win enough seats in the Bundestag to be able to form a government on her own, without the need to seek coalition partners, even if she had failed to win an overall majority of the votes cast.  This would enable her to carry out her programme of reform, which many observers of left and right alike regard as necessary if Germany is to climb out of its present malaise.  But under Germany’s system of Proportional Representation (PR), it seems increasingly likely that Ms Merkel, even if she has won more votes than any other party, will be forced to form yet another German coalition government, either with one or more of the smaller parties, or — increasingly often forecast in recent days — in a so-called Grand Coalition with Mr Schroeder’s Social Democrats.  Either way, the new Chancellor will be forced into so many compromises in order to attract whichever partner might emerge into the coalition that her reform programme seems likely to have to be fudged.  And a Grand Coalition between the two major parties, with ministerial offices shared out between them, leaving only the splinter groups as a rump opposition, would seem to many of us the worst outcome of all:  bad for clear and stable policy delivery, bad for open debate, bad for a government not disciplined by an active and credible opposition potentially able to replace it, bad for democracy.  All these unhappy prospects are the likely products of PR.  Of course the predictions may turn out to be wrong and the elections may yet produce a clear and decisive result for one side or the other.  But even the uncertainty can be ascribed to the PR electoral system which militates against a clear and decisive result in any country such as Germany (or Britain or the United States) where there are two fairly evenly balanced major parties, perhaps a third party some way behind them, and a group of much smaller parties bringing up the rear. 

Tim Hames, in his column in the Times on 13 September, provided a pungent account of the malign effects of PR on German politics and successive German elections:

Although Germany matters so enormously to Europe, German elections have not mattered much to Germans, so there was not much reason for anyone else to become especially interested in them. Although office has shifted between the Left and Right on three occasions since 1948, only once (1998, when Gerhard Schröder initially came in) has this been the direct consequence of an election result.  Otherwise, power has moved not after the voters have spoken but when political parties fall in and out of love with each other and acquire or discard coalition partners accordingly.

Broadly speaking, a political party has to win 5 per cent of the national vote to obtain seats in the German parliament (there is a partial exception which helped the ex-communists from eastern Germany in the 1990s) and it is rare for any party to be capable of seizing more than half the electorate, and hence a majority of seats, on its own. The small parties thus not only play the bridesmaid, but decide who will be wedded to power.

For the better part of five decades this led to what might be described, in deference to what was then the German capital, as the Bonn Paradox. All German adults were permitted to vote, but only 7 per cent of them appeared of any significance.

Those 7 per cent, or thereabouts, supported the Free Democrats — advocates of market economic principles, social liberalism and an Atlanticist foreign policy — and they in turn determined who would hold authority in Bonn. For the first 20 years of postwar German democracy, the Free Democrats considered the Social Democrats to be too socialist at home and too soft on the Soviet Union abroad to be trusted with government. German Chancellors were hence always drawn from the Christian Democrats. In 1969, the Free Democrats changed their stance and their 7 per cent ensured that first Willy Brandt, then Helmut Schmidt, took the highest political post in Germany for the Social Democrats. In 1982, with the activist wing of the Social Democrats clearly aching to shift to the left again, the Free Democrats dumped them and their 7 per cent of the electorate would sustain the Christian Democrat Helmut Kohl as Chancellor for 16 years.

This Bonn Paradox underlines why German election nights were mind-numbingly boring. Not only would the results be released in five minutes, but any citizen with a brain appreciated that, provided the Free Democrats could climb over the 5 per cent threshold to claim their share of parliamentary seats (which, despite scares, they always did), then whichever of the larger parties they opted to support would form an administration.

Almost everything in Hames’s analysis would apply to Britain under PR:  the LibDems, straining to win as much as 20 per cent of the votes, would almost always decide whether the occupant of No. 10 should be Tory or Labour.  The LibDems might, arguably, win more votes under PR, since more voters would be tempted by the idea of voting for the king-maker, and the LibDems would almost invariably be partners in the eventual coalition (voters like to vote for winners): but it’s unlikely in the extreme that they would approach the voting strength of either of the two main parties.  The result would be, as in contemporary Germany, that no individual voter would have any means of influencing the outcome of the election by his or her vote, since there would be no way of knowing what sort of coalition of what permutation of minority parties – i.e., all the parties – would emerge from the post-election bargainfest, still less what bundle of policy compromises negotiated to produce the coalition would emerge from the smoke-filled room.  Since PR might well cause both the Conservative and Labour Parties to break up into ideology-based segments (an anti-Europe segment breaking away from the Tories, perhaps, and a socialist break-away from Labour), it would become even more difficult to forecast how the resultant post-election coalition-making would work out, and consequently virtually impossible to decide how to cast a meaningful vote designed to assist a specific outcome.  The commentators are currently speculating about five possible permutations of party alliances coming out of the German elections, ranging all the way across the political spectrum.  True, fragmentation would allow each voter to select the fragment most closely conforming to that voter’s individual views and prejudices:  but this would be mere self-indulgence, since no voter would be able to influence the outcome.  Better Polly Toynbee’s nose-pegs.

There’s a certain irony, therefore, in the timing of the indefatigable Polly’s latest Guardian broadside yesterday (16 September 05), once again yearning for PR, this time not only in Britain but also, intriguingly, in the UN ("Must we wait a generation for democracy, here and at the UN?"), as usual tendentiously assuming an identity between democracy and PR, although just how either democracy or PR could be introduced at the UN, unless by the institution of world government, is a question that must await Polly’s answer on another occasion. 

Older readers may recall that before the Labour victory of 1997 and for a short time after his triumphal entry into No. 10 Downing Street, Tony Blair hankered after an alliance of some kind with the Liberal Democrats in order, as he hoped, to consolidate a left-of-centre bloc which could be expected to command a clear majority in British politics for a generation.  In this aspiration he was encouraged, for obvious reasons, by those LibDem grandees, Paddy Ashdown and in particular The (now) Late and Rt. Hon Lord Jenkins of Hillhead O.M., formerly Roy Jenkins, in his earlier incarnation a Labour Home Secretary of great distinction.  A few months after the 1997 election, Tony Blair, the new prime minister, invited Lord Jenkins to head a so-called ‘Independent’ Commission on the Voting System, whose 2-volume report, issued in October 1998 as CM 4090, majestically proposed, in rotund Jenkins prose, a novel and ingenious PR electoral system of such complexity that hardly anyone could understand how it was supposed to work, still less what effect its introduction would have on British politics.  Fortunately for us all, by the time the report appeared, Mr Blair’s enthusiasm for an alliance with the LibDems (and for paying the price, PR, which the LibDems were demanding for it) had waned with the realisation that the huge Labour majority in the House of Commons had made it unnecessary, and also because his allegiance to anything resembling the centre-left was anyway already beginning to slip.  No more was heard of Lord Jenkins’s majestic report (and very little more was heard of Lord Jenkins).  For a time, though, there was quite a lively debate on its pros and cons, of which I tried to sum up the cons in a series of messages on an internet Forum (fore-runner of blogs), later collated into a rather rambling article still available on this website.  It includes some unfavourable comments on the Jenkins prose style and its insufferable pomposity.  I think it also disposes of the only arguments for Lord J’s scheme worth discussing, and indeed of the arguments for PR in general.  But I would, wouldn’t I? 

PS (19 September):  As I feared, PR has produced a disastrous outcome from the German elections whose results in terms of leadership and policies will be in damaging doubt for, probably, weeks, and even after that we must brace ourselves for a key government at the heart of Europe half paralysed by coalition dissent and fudge.  Tom Watson, Labour MP, has summed up the lethal implications for the PR enthusiasts of the German elections with brilliant brevity on his website

17 September 2005

10 Responses

  1. Patrick says:

    FPTP may not be as unrepresentative as some people think and I offer this conclusion from a paper examining the psephology ofthe 2005 British Election:

    2 Labour does have a mandate to govern

    Labour secured just 36% of the vote across Great Britain, yet won 55% of the seats in the New House of Commons. No party in British history has secured a clear overall majority with such a small percentage of the vote.

    Yet this does not mean that most people wanted Labour removed from office. Throughout the campaign, YouGov tracked attitudes to a ‘forced choice’ question — would electors prefer a Labour Government led by Tony Blair or a Conservative government led by Michael Howard? Labour began the election campaign with a twelve point lead and ended it with a 17 point lead. By election day, a (narrow) overall majority of electors, 52 per cent, wanted Blair to remain Prime Minister; just 35 per cent wanted him replaced by Howard.

    Kellner, P. (2005) Clearing the fog: what really happened in the 2005 election campaign. The Political Quarterly 76(3):323–332.

    So with our ‘2 and a bit’ party FPTP system, the British electorate were sophisticated enough to give Labour a warning without removing them from office.

    Parties that have been in power in modern times in this country and have a significant enough following have no need or wish for PR.

  2. Tim Weakley says:

    Could I comment on one sentence of Patrick’s? “..The British electorate were sophisticated enough to give Labour a warning..”? Sorry, but we’re not quite that sophisticated: we don’t possess a collective consciousness, we’re not a single mind with over 40,000,000 bodies, and it doesn’t help us understand what goes on to suggest that we are. We may be influenced (but none of us in exactly the same way) by mass campaigning and by mass polls, and by discussions with our peers, but in the end we vote alone. We don’t all get together on the eve of polling day and hold a mass meeting, after which the convener says: “Well, it seems we want another Labour government, but they ought to get a fright first! So you, and you, and you go back to your constituencies and vote Labour, and Joe over there will hold his nose and vote Conservative, and so will Jane (let’s hear it for Jane!), and I’ll drop one for my Monster Raving Loony chap just for fun, and old Bill can support the LibDems or just take the day off, it won’t make any difference…”

    Brian comments: You are right of course that no individual voter can cast his or her vote in accordance with a centrally organised master plan designed to give effect to a collective consciousness and involving separate instructions to every voter. But in practice what happened last time came remarkably close to that. At this year’s general election there was an unprecedented volume of guidance available on websites and elsewhere in the electronic and print media on how best to determine one’s vote in the differing circumstances of each specific constitutency in order to maximise the prospects of (1) keeping the Tories out and ensuring the return of another Labour government, (2) punishing and where possible, consistent with keeping the Tories out, unseating tame Blairite MPs who had consistently failed to stand up for principle and had meekly obeyed the Whips in support of manifestly bad policies, and (3) reducing the Labour majority in parliament to a level at which Blair and other ministers would be forced to pay more attention to their back benchers and indeed to the Opposition, again to the extent that this could be done without prejudice to the overriding objective of keeping the Tories out. Reconciling all these objectives required a subtle analysis of the circumstances in each constituency and the likely implications of casting a vote in those circumstances for either of the two main parties or for the LibDems (or even for one of the still smaller parties). I think Patrick is absolutely right to say that the electorate (or at any rate that part of it which bothered to vote at all) showed great sophistication and maturity in following the advice given, and where necessary voting tactically, so as to produce, collectively, precisely the three outcomes sought — the outcomes that were widely wanted, as the opinion polls cited by Patrick confirm. (It’s worth noting, by the way, that none of this would have been possible under the German system of PR.)
    19 September 2005

  3. Patrick says:

    The ‘national mood’?

  4. John Miles says:

    If, as you seem to fear, the LibDems were to get too big for their boots what’s to stop Labour and Coservative getting together with each other?

    Brian comments (19 Sept 05): Well, I wouldn’t put it past Tony Blair to do exactly that, especially if the Tories were to have the wit to elect Ken Clarke as their next leader. This would be the British equivalent of the German Grand Coalition whose disastrous consequences I have described in my post above. Happily, though, a Labour Party led by Gordon Brown would be much less likely to be tempted by such a Faustian deal. The Labour Party has a longish memory and has not forgotten Ramsay MacDonald and his great betrayal of 1931. Gordon Brown knows all about Ramsay Mac, although I wouldn’t bet on Tony Blair ever having heard of him!

  5. Jarndyce says:

    Two points, Brian:

    1. How can you say the likely problems ahead in choosing a govt. in Germany are caused by the electoral system, when the two major parties couldn’t even manage 70% between them? Surely if the electorate shared your conviction that these reforms were needed for Germany, they would have voted in rather larger numbers for Merkel?

    2. Your analogy for the UK, with the LDs holding the perpetual balance of power, assumes that if we switched to PR the current party system would be held in stasis. Don’t you think that’s a rather unlikely scenario? In fact, as you mention immediately after, things could break up in an entirely different manner. But you also assume fragmentation. That hasn’t happened in, e.g., Spain or Malta.

    Brian replies (19 Sept 05): I didn’t say I favoured the Merkel reform programme, only that many commentators of left and right do, and that most Germans, whatever party they support, seem to recognise that reform of some kind is urgently necessary: hardly surprising, with more than 10 per cent unemployment. Both the SDP and the CDU promise reforms and the great majority of German voters have just voted again for one or other of those parties. My point is that PR has deprived either party of the overall majority in the Bundestag that would have enabled either to embark on a clear-cut reform programme, undistracted by the need to compromise with one or even several coalition partners. There’s no guarantee that First Past the Post (FPTP) would have given either party an overall majority of seats, but it seems quite likely: in Britain, no party since the second world war has ever won an overall majority of the votes cast, yet the hung parliament that this would always produce under PR has almost never happened under FPTP.

    I agree that there’s no way of knowing for certain what effect PR in Britain would have on the existing party structure. If the three biggest parties remained more or less united, the LibDems would probably increase their current share of the total vote (22 per cent), perhaps to as much as 30 per cent, and both Labour and the Conservatives would probably lose some of their shares to the LibDems and to other even smaller parties such as the Greens, so that we might well end up with three almost evenly matched parties with a host of others, either nationalist or single-issue for the most part, bringing up the rear. This would mean either the LibDems assuming the coveted role of king-maker and permanent coalition partner in government, as for years in Germany, or else a British Grand Coalition of Labour plus Conservatives, surely everybody’s worst nightmare, and virtually guaranteed to cause a major split in the Labour Party as in 1931. Either way, the voters would lose any say in the decision on which party occupied No 10 Downing Street.

    Enthusiasts for PR, however, generally predict that PR would encourage both (all?) the main parties to split up since socialists, for example, would probably vote for a left-wing breakaway segment of the Labour Party which, under PR, would have a good chance of winning separate representation in parliament on an explicitly socialist plaform (we should be so lucky!), and might be able to bargain its way into a left-of-centre coalition government. The same would apply to groups within the Tory Party, such as either Europhiles or Europhobes, and perhaps also the LibDems. This would make it even more unthinkable than it is already that any one party would ever be able to win an overall majority in parliament or to carry out anything resembling the programme for which it had campaigned, even if it had attracted more votes than any other. After every election we would face the messy, uncertain and unpredictable future now confronting the unfortunate Germans.

  6. Tim Weakley says:

    Brian: thanks for the long comment on my remarks! I’ll meet you halfway: the election results were as if the voters, or at any rate a sufficient number of them in those constituencies that are not irredeemably solid Labour or solid Conservative, had actually listened to the pundits regarding how to return Labour with a reduced majority. Whether this was really the case would require asking the voters, post-election, what influences had affected their votes – perhaps this has in fact been done for a meaningfully random and large selection of voters? Until then, we should remember that consistency (‘as if’) is not the same as proof: the Universe as we perceive it is consistent with having been created by some Entity some thirteen billion years ago, but this should in no way be construed as a solid proof of the Creator’s existence either then or now. Anyway, I just feel that attributing the populace with a collective will is unpleasantly reminiscent of the self-congratulatory way in which winning parties talk about ‘the people having spoken’ and ‘we have a popular mandate’. They mean “Sixty-three percent (or whatever it was) of the electorate voted, and of these 48 in every 100 thought our package deal was less revolting than the other parties’ package deals, although few people liked all of it and many had strong reservations about large chunks”.
    Sorry, a long way from the German elections.

  7. John Miles says:

    If, as you seem to think conceivable, Blair were to make a deal with Ken Clarke it would be interesting to see how they got on over Iraq.
    I’m not sure how relevant Ramsey Macdonald’s Great Betrayal is to any coalitions that might happen under PR.

  8. Brian says:

    An impressive retort by Jarndyce in The Sharpener (trackback above). My comments are in a new post of 22 September.


  1. 18 September, 2005

    Carnival of the Britblog Roundup # 31

    Yes, it’s here, the event of the week, what you’ve all been waiting for, the Britblog Roundup #31. Get your nominations for next week’s in early by emailing to britblog AT gmail DOT com. We’re interested in any subject, any

  2. 22 September, 2005

    […] From the distinguished to the idiotic, Germany’s election chaos results are getting kicked to death this week by electoral system-conservatives. The anti-PR brigade are sharpening their knives with glee. But why? The German result is irrelevant to the PR debate here in the UK, for at least three reasons: […]