Turkey’s election 2011

Turkey has its parliamentary election Sunday.

It seems as though the only interesting questions are: (1) how big will the incumbent party’s majority be, and (2) how many of the sorts of electoral-system induced anomalies that we have seen in recent elections will we see this time?

A Council of Europe report criticizes the Turkish 10% national threshold, as well it should. The source of the anomalies is that the country uses a fully distrcted “proportional” list system, but parties are unable to win seats, regardless of their district-level strength, unless they surpass 10% nationwide. “Independents,” on the other hand, can win seats on their local support.

The Council of Europe report criticizes numerous other aspects of Turkish politics. I continue to be skeptical that Turkey deserves to be ranked among the democracies, other than in the most minimal sense.

11 thoughts on “Turkey’s election 2011

  1. I’ve got serious concerns about Turkey as a democracy on the human rights and civil liberties fronts but I’m not sure why I should be concerned about it on an electoral front.

    Are threshold requirements unusual? I recall Germany having them but don’t recall the number.

    And why is a threshold undemocratic? Many countries don’t allocate seats proportionally at all, requiring plurality wins in SMD setups or run-offs to achieve majorities. Isn’t a 10% floor more friendly to small parties than either of those alternatives?

    • Thresholds based on minimum national party-list votes that are set higher than 5% (Germany’s level) are extremely rare. (Off the top of my head, only Russia, also likely no longer a democracy, is higher, at 7%.)

      Thresholds in fully districted “PR” systems that are nonetheless based on national vote shares are likewise extremely rare. (Sweden is one of the few long-term democratic countries with such a threshold, but it is much lower, at 3 4%.)

      You are right, James, that single-seat districts impose high “effective” thresholds. However, in most actual plurality or majority systems, various small parties win seats due to regional strongholds (the recent Green wins in the UK, Australia, and Canada come to mind). There is no plurality/majority system that I am aware of in which a party could win one or more district seats and have them taken away for failure to reach some level of nationwide vote total. So, no, the 10% threshold would not typically be “more friendly” to small parties than a plurality/majority system based on single-seat districts–at least if found in a large country with a large assembly. Without doubt, if Turkey changed to FPTP, but kept its 500+ seat legislature, there would be many more parties represented in parliament.

      Each of Turkey’s barriers is thus unusual. Having these various provisions in one electoral law is thus especially unusual. I’ll leave the question of where (if anywhere) the line should be drawn on thresholds leaving a country outside the realm of electoral democracies up for debate. There is no clear answer, and for whatever it might be worth, a few years ago the European Court of Human Rights declined to rule Turkey’s threshold illegal.

  2. Matthew:

    Thanks; I see the issue. Having both districts AND a national-level floor is a perplexing system and it’s hard to construct a democratic rationale for it. It makes sense to have systems in place to force consensus (run-offs/double balloting) and not reward fringe parties (thresholds). But this may go too far.

  3. The goal of the 10% national threshold is to prevent regionally concentrated ethnic or religious minorities, like say the Kurds, from getting significant representation. Its working quite well in that respect-I suspect that’s the Council of Europe’s problem with it, rather than having a threshold in and of itself.

  4. If the threshold was modified to requiring 15% in each of the multi-member districts instead of the 10% nationwide?

    Would that be an improvement in Turkey’s case? Turkey does not have a nation-wide tier only multi-member districts.

    Does the Turkish system of PR function similarly like the Estonian system of PR which has a 5% nationwide threshold with multi-member districts, no nation-wide tier?

    The AK Party is going to win a majority even if the threshold was set at 5% nation wide. They are polling 48% which is probably a low estimate. I imagine that they would win 51% of the vote. The big question is how big a majority the AK party wins. I imagine that they most likely would be a few seats short of a two-thirds majority.

    • Suaprazzodi, Estonia uses nationwide PR. The districts affect which candidates win seats, but not how many seats each party wins. Nothing at all like the Turkish system.

      As for district-level thresholds, given an average M of 6.5, there are not many districts in Turkey where a party with under 15% of a district’s vote plausibly could win a seat anyway. However, as I noted at the earlier planting that I linked to, there are many districts where the largest party in a district wins no seat because it fell below 10% nationwide.

  5. I recall that the 10% threshold was set up in the 1980s by the outgoing military government, in an effort to channel votes to the three tame parties it had set up and to erect a barrier to keep new parties from emerging.

  6. Ed, your recollection is basically correct, although the primary effect (and intention) is to block regional parties. Read: Kurds.

  7. Why was the Turkish party system so fragmented in the mid-90’s?

    It seems like the electoral system is now doing it’s job. It’s now a three party parliament with indies. The election results are becoming more proportionate. It’s a bit ironic the AK party is winning more votes, but fewer seats in each election.

    What would the election result have been if there was no threshold, or a 5% threshold nation-wide?

    Should an at-large national wide tier be introduced with a threshold of 5%?

    I thought the Estonian PR system was a multi-district PR system, not a nation-wide system like Israel.

    • Estonia has a two-tier PR system. So, districted, unlike Israel. But nationwide proportionality, very unlike Turkey.

      Back to Turkey: DC is right that the (mostly Kurdish) independents “soak up some of the otherwise wasted vote.”

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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