Pages Menu
Categories Menu

Posted by on Jan 12, 2018 in Featured, Germany | 9 comments

Against Proportional Representation

Against Proportional Representation

Europe’s political world was shocked to learn that Christian Lindner, head of Germany’s Free Democratic Party (centre-right), had walked out of coalition negotiations with the Christian Democrats of Angela Merkel and the Greens. “It’s better not to govern, than to govern badly,” Lindner claimed. Critics allege he had planned this move from the start, and accused him of opportunism. Germany usually manages to hammer out strong coalitions relatively quickly, but now the way forward is unclear.

Christian Lindner, leader of the Free Democratic Party, the man who single-handedly threw German politics into turmoil.

Solutions are few as the only other possible coalition is to continue the last term’s grand coalition between Merkel and the main centre-left party, the Social Democrats. Yet, the latter declared on Election Day that they would stay in opposition, as their recent stints governing with Merkel have robbed them of a distinct identity and led to catastrophic electoral results. 

There could be another election, but polls suggest it would yield the same result. Finally, Merkel could form a minority government, but it would be an exceptionally weak one, controlling only one-third of the seats in parliament and unlikely to stem the tide of anti-establishment radical parties. She is currently negotiating with the Social Democrats, so that the latter might shield her government from being overthrown without formally joining cabinet. Suffice it to say, everybody involved is unhappy with this.

Every aspect of this crisis is tied to a major flaw of proportional representation (PR), which Germany uses from the Social Democrats’ anguish over grand coalitions, to the mainstream’s desperate scramble to keep extremists out, to how all of this is negotiated behind closed doors. These problems would be avoided if Germany had preferential voting.

Elections are about choosing governments

At the most abstract level, a government should only be sworn into office and be empowered to make decisions if it has the support of over 50% of the people. Especially in large, diverse polities with a wide spectrum of beliefs, any attempt to create a majority involves making compromises, both in issue positions and in priorities. In majoritarian systems, the major parties make these compromises before the election, assembling heterogenous “big tents” that hope to secure 50% of votes. In proportional systems, the numerous parties make compromises after the election in the form of coalition negotiations. Its logic is that since seat share mirrors vote share, any government that commands a majority in Parliament must also have the support of a majority of the general public. However, reality does not necessarily bear this out.

Winston Peters, leader of NZ First and “kingmaker” following the country’s September elections.

An example of this may be taken from September’s election results in New Zealand. Due to its PR system, election results were also inconclusive on election night – neither the centre-right National Party nor the Labour-Green coalition had a majority and both needed the support of the nationalist New Zealand First party. Thus, for three weeks, the country’s political future lay in the hands of Winston Peters, who had created NZ First as a personal party and could make whatever decision he wished. For three weeks, commentators attempted to pry into his mind, looking through his political history, personal grudges, and desire for a legacy to figure out who would end up being the new Prime Minister.

Left out of this process was the Kiwi electorate; their opinion could only be divined through opinion polls. What government a majority prefers would seem like an incredibly important measure in a democracy, but PR offers no way of measuring it, much less ensuring it. Who became Prime Minister of New Zealand was instead a function of which side was willing to concede more principles behind closed doors, who offered better Cabinet positions, and the peculiarities of Winston Peters’ psychology.

In other words, in PR, voters can choose between parties but have no say in the coalition agreement that actually decides how the country is governed. This is not a trivial distinction; New Zealand voters were fortunate in that Peters’ arbitrary decision process happened to result in a Labour government, the outcome that a majority probably wanted.

Others have been less lucky. 25% of Dutch voters chose Labour in 2012, only to watch them then form a coalition with the centre-right Liberals. The fact that Labour’s polling quickly collapsed to 8% showed that Labour voters were really not happy that their vote had led to continued right-wing government – but they had no recourse to correct it for another five years. Similarly, the UK Liberal Democrats’ polling immediately dropped from 25% to 10% after their 2010 coalition agreement with the Conservatives. These coalitions, then, never had the support of a majority of the population, even if the individual parties did.

In the same way, the coalition-formation process in Germany has been insulated from democracy. Those who voted for the Free Democrats likely expected them to be friendly to Merkel (as the two had been natural allies through most of postwar Germany); Lindner’s decision to torpedo talks (likely) for personal gain would have come as a surprise. Meanwhile, within the Social Democrats, the ultimate decision to negotiate with Merkel or rebuild in opposition is being made through the party’s arcane factional struggles.

The Peril of the Grand Coalition

The grand coalition is another feature unique to PR, and it poses a significant problem for democracy. PR countries are more likely than majoritarian ones to see far-left and far-right parties enter parliament, and their very existence makes the formation of a majority government more difficult. At a theoretical level, PR supporters might argue that this forces mainstream parties to negotiate with those discontent with the status quo, and this has indeed happened in some countries. But other countries instead see mainstream parties band together and erect a cordon sanitaire against extremists.

Social Democrat leader Martin Schulz with Christian Democrat leader and Chancellor Angela Merkel. Whether they were alternatives or collaborators was never clear, leading to poor results for both.

Germany’s mainstream parties mostly refuse to negotiate with either radical-left or radical-right. For this reason, two of the past three federal governments have been grand coalitions, and talks between the two parties are again underway. But another grand coalition would be a terrible outcome, for any democratic government depends on the existence of a “loyal opposition” that could, at any moment, form an alternative majority as soon as voters tire of the government in place.

For this reason, Merkel’s main challenger in the elections was unable to convincingly criticize her, for his party had been in cabinet and bore responsibility for its record. This meant that voters were given options but no power – their choices were to vote for Merkel, a party that had just collaborated with her, two minor parties who were likely to collaborate with her, or radical parties. A credible alternative was clearly desired, but the nature of coalition politics meant that none was available.

PR In Decline

The collapse of German negotiations is all the more ominous, since Germany is the only first-world country to consistently marry stability, proportionality, and a population larger than Canada’s. (France and Italy abandoned PR due to instability; TurkeySpain, and Poland use systems with significant distortions).

Indeed, it is possible that the twin phenomena of dealignment and of mistrust of the “establishment” have fundamentally changed how (or if) PR systems function. Germany’s recent (and problematic) reliance on grand coalitions, for example, was not intended as a feature of the system; for decades they were not necessary, as one major and one small party could always form an ideologically coherent government. Party fragmentation has changed the calculus: this year saw the first negotiation involving three parties (four if the conservative “sister parties” are counted separately), as well as the first negotiation ever to collapse. Should the latest talks conclude in a Merkel minority government, this too would be an unfortunate first.

Similarly, the Netherlands has always functioned well because its three major blocs (Christian, socialist, liberal) consistently won 75% of votes, but this system has fragmented. 2017 saw the first-ever four-party coalition, following the longest negotiations in Dutch history. The Spanish election of 2015 was the first ever to require a repeat; only after ten months and the overthrow of the opposition leader was a very shaky minority government able to form. Ireland’s minority government is also historically shaky and requires herding a large group of independents. No majority is in sight following last month’s Czech elections, while Italy’s next elections are likely to plunge the country into a state of ungovernability. PR advocates often respond to claims of instability with “plenty of countries do it fine” – but perhaps this is an argument that only holds with pre-1990 voting behaviour.

The Alternatives

The Italian Chamber of Deputies. The return of PR in Italy is expected to make coalition formation all but impossible.

Meanwhile, last month’s elections in the Australian state of Queensland, which uses a preferential system, provides a better method of assembling majority coalitions. Polling showed that neither Labor nor the Liberal National Party (LNP; centre-right) was close to winning 50%, while the radical-right One Nation party held 15-18% of first-preference votes. Consequently, the LNP suggested to their voters to rank One Nation as their second preference while also adopting some of their rhetoric, in the hopes that One Nation would reciprocate. But this faced a backlash, for One Nation is reviled by a broad swath of the public; some cite the deal as a key reason for the LNP’s loss to Labor. As in any democracy, Queensland parties had to weigh the pros and cons of aligning with certain groups over others; the difference is that voters got to render their judgments on these alliances, as opposed to watching later negotiations from afar.

The preferential system and its cousin, the French two-round system, marry stability with numerous parties, while usually ensuring that the winning party is preferred by an absolute majority at a nationwide level. There are exceptions where the less-preferred party still wins more seats – but this could probably be fixed by the addition of leveling seats. (Variations exist within this category.)

Another possibility comes from Italy, whose 2015 electoral law featured a fascinating innovation: the election would be conducted by PR, but if no party wins 40% of votes, the top two parties compete in a runoff. The party that wins the runoff (and thus captures over 50% of preferences) is awarded a majority of seats, with the remaining seats allocated according to first-round vote share. The result is a government that is preferred to any other by a majority of Italians, but where smaller parties get a voice in parliament and a chance to become one of the major parties at a later point. Regrettably, the runoff provision was ruled to go against Italy’s constitution, but it provides an interesting possibility for other countries contemplating reform.

No such talk of reform has surfaced in Germany. To be sure, Germany will figure out an end to this crisis, and years from now, few will remember if coalition negotiations took one month or six. But over the long term, presuming all six parties are here to stay, the country’s political system has two options: one is to drop the cordon sanitaire and normalize negotiating with radical parties (as recently happened in Austria); the result is a two-coalition system, not very different from what would happen under majoritarianism. The other option is to stay the current course, where the major parties usually have no choice but to form grand coalitions in order to avoid utter chaos, depriving German voters of a meaningful choice and thus strengthening radical parties. Both options reflect poorly on the PR system’s ability to encourage stable, coherent governments.

Edited by Marissa Fortune.


  1. This is a thoughtful, evidence-based critique of party proportional voting and the problems it brings to otherwise fine countries. Iceland would have been another example. Seven parties in their tiny 63 member parliament. A year and a half and two elections to produce the current, still shaky government.

    If you prefer the stability, accountability and simplicity of BC and Canada’s current First Past the Post voting system, join our informative, no rant, non-partisan Facebook Group: FPTP …It Works for Canada.

    • FPTP can product stability, but it offers neither accountability nor simplicity. In Canada federally, the majority of voters voted for someone other than their current representative, and while the article may try to suggest otherwise, there is no special recourse to eliminate voter remorse after the election in majoritarian systems (look at Trump, or Brexit, or the current Liberal sweep of the Maritimes). Seat majorities are stable, but that very trait makes them largely immune to democratic input between elections. The knots we tie ourselves in over “strategic voting” show that there is no simplicity for the voter to understand who their vote is aiding.

      The claim that FPTP somehow protects us from extreme parties is shaky at best: we have seen how it gives disproportionate power to extreme parties like Reform and the Bloq, pitting different parts of the country against each other and siphoning power from more moderate broad-based parties. Even in the moderate major parties extreme Canadians end up with backroom or dog-whistle concessions under FPTP. Pretending extreme Canadians don’t exist doesn’t solve anything, and trying to deny them votes ends up costing regular Canadians their voice.

      “Proportional representation” works pretty well as a definition of democracy, and it’s messy because democracy is messy. Majoritarian systems like our current FPTP are stable by contrast, and it’s pretty easy to see why. Arbitrary power is stable power. It’s a question of what you value as a Canadian.

  2. This article has a lot of problems and plays a lot of gymnastics with our minds. Let’s consider some actual current realities with the few remaining OECD countries that still use FPTP: We got…Brexit. We got Trump. In the UK Theresa May’s complete dependence on a regionally based extremist anti abortion party…she would have had other options under a proportional system. Plus FPTP countries tend to have massive government debt. The author is quick to dismiss Germany as in peril…but consider that Germany now has a 28 billion dollar government surplus. Wouldn’t that be nice? Arend Lijpart, political scientist in a significant study compared 36 democracies over 55 years. He studied two systems: Majoritarian (this includes FPTP and AV/preferential vote in a single district riding) and Proportional Systems. Lijpart found that on 16 of 17 measures PR (proportional representation) countries outperformed majoritarian countries including in the areas of government effectiveness and control of corruption (i.e. avoiding being captured by elite interests). This is not surprising since PR governments represent a broader set of voices – often considerably more than 50% of voters have some representation in government. (Compare this to the 39% that in Canada makes a “majority government” under FPTP – the rest of the populations sits on the outside). He also found that countries with proportional systems have more effective environmental policies …again not surprising….there are more voices at the table and negotiation / cooperation has to occur- let’s not be to quick to consider this a bad thing. Proportional representation will not respond to all democratic deficits but for those of us living in relic FPTP countries it will be a significant step forward for our democracies. Please do your research before making a decision. Of the more than 12 national or provincial studies / commissions etc done in Canada on the question of electoral reform – all have recommended adding proportionality to our voting system. over 88% of OECD countries use some form of proportional representation. To be completely honest..I sometimes wonder how we even consider ourselves a democracy when more than 60% of voters vote for something other than the ruling government but have no voice where it matter – in decision making power.

    from a McGill alumni

    • Hi! Thank you for your comment! You’ve definitely given me a lot to think about, here’s how I would respond:

      1. Sure, I don’t disagree that the American and British political scenes have gone very wrong. But Hungary has Orbán, Turkey has Erdogan, and Poland has Kaczynski, each with majority support. Any country where a clear majority is trapped in a nationalist fever would have a nationalist government, regardless of the voting system. And I’m not convinced that it’s the voting system that *determines* the extent of nationalist sentiment.
      2. I don’t buy at all that May “would have had other options under a proportional system”. As I mentioned in the article (the section “PR in Decline”), in the de-aligned, personality-centred political world, the process of coalition formation has become extremely difficult, bordering on impossible, with Merkel being the most prominent example.
      3. And because I’m discussing how PR has *recently* been working (given the rise in radical parties and the necessity for grand coalitions) citing a study of the previous 55 years seems to be beside the point. There’s also a methodological question – Lijphart ran a regression analysis on the 18 factors you mentioned, which shows correlation more than it does causation.
      4. I agree that 39% should not make a “majority government”, which is why at no point in this article do I defend First Past the Post. As I say in the article, the very first condition that any electoral system should meet is that a government must always have 50% of the vote.
      5. I specifically made sure to address the “most other countries use PR” argument – this is why I brought up the fact that Germany is the only advanced economy to “consistently marry stability, proportionality, and a population larger than Canada’s”, as well as how recent performance has been quite different from historical performance.

      Koji Shiromoto

  3. Coalitions prevent the kind of rabid party-identity politics which have swept majoritarian systems. If you have to cooperate at a grassroots level in a big tent party, you have to identify with the Roy Moores and the Donald Trumps, even if you wouldn’t normally do so, and must rationalize voting for them to get your party (the only one with any chance of representing people like you) a win.

    Not so under Proportional Representation: under PR you can simply vote for another party of vaguely the same ideology if you feel the leadership of your old party has become corrupt, out-of-touch or extreme. So no, coalitions are not simply “big tent parties” formed after elections; they are different, far less-permanent and thus less-dangerous entities.

    It’s also worth noting that the Conservative Party is currently in a coalition with the DUP, while the Republican party is currently controlled by the populists. So it’s not like that “Cordon Sanitaire” against negotiating with radicals is particularly strong in majoritarian states.

    • Hi! Thank you for your comment!

      So I wasn’t arguing from the position that the cordon sanitaire must always be kept – mentioning that some do, some don’t, and the consequences of both – but from the position that elections should always be competitive, i.e. feature two responsible alternatives. I’m perfectly fine with radical parties being subsumed into to broad left-wing and right-wing coalitions, as has happened in Denmark, Norway, and now Austria – as this actually offers voters a choice, unlike Germany’s grand coalition. (Like, it’d be best if people didn’t vote for radical parties at all, but given that they do, this is the less bad option.)

      Secondly, it’s not as though the entire American conservative movement had been forced into voting for people like Trump and Moore against their will – Trump maintains 38% approval even after all that has happened, suggesting their ideas would be powerful in a PR democracy as well. There are some that held their noses to vote for him, of course – but even if they had a centre-right, John Katich-style party to vote for, it would face the choice after the election of forming coalition with Trumpists, or forming a grand coalition with moderate Democrats and leaving Trumpists as the main opposition.

      Lastly, I did make sure to advocate not for maintaining First Past the Post, but forms like preferential voting and the two-round system, which allow new parties to be formed without being spoilers, so that there *would* be competition between multiple right-wing parties, and between multiple left-wing parties, as you mentioned. (Such as the competition between UDF and Gaullists on the French right, between Macronists and the Socialists on the left/centre; the Liberal Democrats, who were within four points of overtaking Labour in 2010, might have done even better under this system, etc.)

  4. More proof Justin Trudeau will pledge to enact Alternative Voting in the Liberals 2019 platform.

  5. Koji, if you’re insisting that any majority government have the support of the majority of voters, you’re just describing proportional representation.

    Runoff systems (AV is just an instant runoff) that still only elect 1 person are not proportional and do not guarantee that a majority government. They only ensure that each elected candidate has the majority of the remaining votes, which is a much smaller number than the number of voters. They work fairly well for things like non-party municipal elections, gubernatorial, or presidential elections (where there is no functional way to make it proportional), but they do not guarantee that a majority government in a legislature has the support of the majority of voters.

    The systems you link to as “improving” preferential ballot are STV and MMP: these are the 2 main systems of proportional representation being advocated for in Canada. Almost nobody is advocating for “pure” PR.

  6. Koji Shiromoto makes a change from electoral reformers, who are star-struck with every party-proportional voting system under the sun. The fact is that they cannot democratically elect a government, for want of a transferable vote across party lines.
    It is a little unfortunate, for the sake of argument, as Gallagher pointed out in his comments to the Canadian Electoral Reform committee, that the 2016 Irish election did not demonstrate the unique potential of Single Transferable Vote for providing a democratically decided majority coalition, as by Fine Gael and Irish Labour.
    Ireland has become disenchanted with their once main party Fianna Fail, because of its financial mismanagement (“Ship of Fools”). The Irish also turned against Fine Gael, because of its austerity measures, in consequence.
    The 2016 centenary of the Irish revolution/rebellion revealed the two parties are still implicitly pro and anti Irish treaty with the UK, which is perhaps why they still would not coalesce in government.
    This left Fine Gael to cobble together a government, including an Independent.
    Popular distrust of the parties led voters to elect more Independents than ever. And this freedom that STV gives is unique in Europe. I once read, even before the (23 I think Independents) Ireland with STV had more Independents than the rest of Europe combined.
    In short STV is the democratic electoral system that gives liberty, equality, fraternity.
    Richard Lung. “Democracy Science.”

Share This
%d bloggers like this: