Don’t forget: Iceland is semi-presidential

Many comparative politics works lump the Icelandic system in with the “parliamentary” democracies.* Today we are reminded of why that is incorrect.

As part of the fallout from the “Panama Papers”, the Prime Minister, Sigmundur Davíð Gunnlaugsson, requested that the president, Ólafur Ragnar Grímsson, dissolve parliament and call early elections. The president refused the request.

Now, one could argue that this act alone does not prove the system is not “parliamentary”, because some presidents in strictly parliamentary systems might be able to refuse a dissolution request under some circumstances. But the event does show that the Icelandic president does take politically consequential actions, and on top of the direct election of the post, this surely qualifies the system as “semi-presidential” in practice as well as in formal rules. That is, rather than being purely ceremonial and accepting “advice” from the head of government, a president who is an agent of the electorate has exercised discretion.

The prime minister subsequently resigned.


* That is, those works that don’t just ignore it.

19 thoughts on “Don’t forget: Iceland is semi-presidential

  1. Iceland’s president also has a form of veto, in which he can send a bill to be voted upon in a referendum. The current president, Olafur Grimsson, has used this power on a fairly frequent basis, most notably in two separate occasions related to the Icesave bailout.

    • Well, I don’t know if I would call that ‘frequent’. He’s been in office since 1996, and the 2010/2011 referenda were the first ever use of the power.

      • No, he used it in 2004 on a media law, but that was the first time a president had used the veto power so I admit that “frequent” may not have been the best descriptor.

      • Well, I stand corrected. Was the bill withdrawn by parliament in 2004? I couldn’t find anything about a referendum in 2004.

  2. Presidential elections are also due in June, which will be the first non-incumbent election since Grimsson was elected in 1996. Unusually for semi-presidential systems (and increasingly unusually in general), Icelandic presidential elections are held under plurality rule.

  3. “some presidents in strictly parliamentary systems might be able to refuse a dissolution request under some circumstances. But the event does show that the Icelandic president does take politically consequential actions” From what I recall from when I was writing my bachelor’s thesis (on the effect of dissolution powers on party unity), dissolution provisions tend to be rather similar between semi-presidential and parliamentary systems.

  4. Why I love having this blog: while I knew about the referendum “veto” provision, and its use in 2010, I did not know it had been used in 2004.

    JD, I think you are right about dissolution powers in parliamentary republics, hence the caveat in the post about popular election along with the use of discretion. I am not sure how often unelected presidents exercise discretion in these cases (although maybe you do, from your research).

    • I was not able to find much information about when parliamentary heads of state refuse dissolution in practice. It’s difficult to measure, as PMs can usually make the request in private, and would often prefer not to go public if rebuffed; more importantly, though, if a PM believes he will be rebuffed, he would often just not ask for it at all – so the power to refuse has an effect, even if it is never actually exercised. Interestingly, the best study I could find of the occurrence early dissolution (“Strategic Parliamentary Dissolution” – Strøm and Swindle 2002), mentions the presence of coalition partners who oppose dissolution as a reason a head of state (elected or not) might refuse dissolution advice. I think they indeed found early elections to be somewhat less common in countries where coalitions are often needed, but I’m not sure.

      Looking at the powers table in ‘Presidents and Assemblies’ (Shugart & Carey 1992) page 155, you seem to have missed the Icelandic referendum-veto there (as well as Weimar, which I think had the same power). From your comments on Ecuador on page 149, am I right in saying you would score it a 3, as a veto power stronger than a 2/3-override veto?

      • JD, those are good points about the difficulty of assessing cases of dissolution requests. Apparently there is even some question in this Icelandic incident over whether a formal request was made.

        As to the powers, it looks to me as if Shugart and Carey have both Germany (Weimar) and Iceland scored “2” on referendum power, same as Ecuador. It looks like we did not adjust the veto score in cases where a referendum could be called to (potentially) veto a bill passed by the assembly. Instead, we just took provisions of this sort to be “restricted” authority to call a referendum. Here the restriction would be that there must first be an “objectionable” legislative measure.

  5. My understanding is that the PM went to the President without having consulted with his coalition partner or even his own party colleagues. The President felt that a majority still existed in the Althing and so declined the request.

    Though I”m sure the fact that the Pirate Party is in a commanding lead in the polls influenced his decision at least somewhat. The Establishment tends to have strong self preservation instincts.

    • Olafur is, from my limited understanding, not a particularly “establishment” figure, particularly for a head of state. He was a left-wing political science professor before he went into politics, and at that point he represented the left-wing People’s Alliance.

      As far as I can tell, the President simply didn’t feel an election was necessary as the government had not lost the confidence of the Althing. Unlike, say, the UK, in Iceland it’s not customary for the parliament to be dissolved at the whim of the PM.

      • I’m a bit late to the party but wanted to throw in my own knowledge here.

        The 2004 veto was highly controversial at the time. It was the first time any Icelandic President had used or threatened to use the veto power and there were accusations that it was done so in order to protect Olafur’s daughter. At the time she worked for a media company the government was attempting to break up. See link below;

        The fallout from that veto was quite interesting.

        If you look at Presidential elections before 2004 they were quite sedate affairs. An unspoken convention said that if the incumbent chose to run again no other candidate would stand against him. The election of 2004 after Olafur excercised his veto was contested, thus going against this convention. The other two candidates fared badly but 21% of voters decided to spoil their ballot (seen as a protest against the veto used by Olafur earlier in the month).

        Since then that old convention seems to have gone out of the window.

        The veto triggering a referendum used during the financial crisis was nowhere near as controversial (60% of voters upheld Olafur’s position). You could argue that by using the reserve powers that had previously been untouched Olafur has led Iceland to become a fully semi-Presidential system rather than a Parliamentary one.

        What will be interesting is that Olafur has – ostensibly due to the Panama Papers – decided to run again for office at the 11th hour this week.

        This Presidential election may very well end up as an indirect referendum on Olafur’s decision to deny a fresh legislative ballot…

  6. Given what both Andrew and Christopher say (a similar version of which an Icelandic contact also told me), I think this particular use of discretion over a dissolution request may be no different from the type that an unelected president may exercise. That does not change the point of the post–the president is elected and has some discretion, so the system is semi-presidential. It does, however, suggest the president was playing a more “regulatory” role typical of heads of state in parliamentary and semi-presidential systems rather than a “partisan” one as we would see in, for instance, France or Finland under the latter’s pre-2000 constitution.

  7. Would defining Iceland as semi-presidential also make Éire semi-presidential, as the President may (though to date has not) refuse a dissolution and also refer bills to either the Supreme Court or the People?

    • Semi-presidentialism is defined as any country where there is both a directly elected president and a Prime Minister and/or cabinet responsible to the legislature. No formal dissolution power is needed.

      It’s a broad definition, and one that Shugart resisted in ‘Presidents and Assemblies’, but it’s entered general use due to the subjectivity in judging presidential power. It does sweep up effectively parliamentary regimes, like Ireland, and probably some presidential ones (Korea? Taiwan? Colombia?), though more parliamentary ones than presidential ones in my view.

      • Samuels and Shugart reinstated the “semi-presidential” category, while maintaining the sub-categories of premier-presidential and president-parliamentary. In the intervening years, Robert Elgie and several others developed similar classification schemes.

        We (S&S) classify Ireland as semi-presidential (specifically,premier-presidential), although the case could be made that the presidency is just too weak to qualify. Iceland’s formal powers make it president-parliamentary, although aside from the cases discussed in this thread it seems Icelandic presidents do not exercise the powers that this classification-by-inputs implies.

        If there is no PM (as in Colombia) or there is a PM who is not subject to assembly-majroity confidence (South Korea), then it is not semi-presidential. I understand Taiwan to be president-parliamentary.

  8. Interesting point re: heads of state. Didn’t CDU President Carstens refuse to dissolve the Bundestag in 1982 when Helmut Schmidt requested it (in the aftermath of the Free Democrats flipping to the CDU & Kohl). Schmidt was evidently irate and thought the refusal was unconstitutional (or at least improperly “political” for a German president).

    • Bill, you might want to check out my series of posts on “mostly ceremonial” presidencies in parliamentary systems. I do not recall whether there was a refused dissolution request in 1982 in Germany. However, the early (1983) election was called under similar terms as the 2005 election–that is, with the government faking its own loss of confidence.

Leave a Reply

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

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

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s