Trust Not the People

Aengus Collins wrote an excellent piece last fall on the democratic flaws inherent in Turkey's current system of closed-list proportional representation. As Collins notes, though the 10% threshold required of parties to enter parliament gets most of the attention (including recent mention in a 2010 Council of Europe report), considerable less attention is paid to the actual method by which parties and voters choose candidates. From Collins:

There are problems at every stage of Turkey’s electoral process. As I highlighted in my most recent post, parliament’s 550 seats are badly misallocated among the country’s 81 provinces. Next, the processes used to translate individual votes into seats for parties are deeply skewed. The 10 per cent threshold that parties need to clear before they can enter parliament deservedly gets the most attention, but it’s not the only issue here. Once the threshold has been passed, the d’Hondt method is used to distribute seats among the remaining parties. Of the many variants of proportional representation, d’Hondt is the least proportional, systematically favouring larger parties.*

We reach a further set of problems when it comes to filling the seats that have been allocated to the various parties. Turkey uses a closed-list proportional representation system. This means that voters vote for the party of their choice, but there is no mechanism for them to express a preference for one or more of the party’s individual candidates. Instead, a list of candidates for each province is drawn up by the party leadership and any seats won in that province are automatically assigned to the names on the list, starting from the top.

It’s not the list per se that causes the problems here. List-based proportional representation is an extremely widely used electoral approach. But in most cases an open list is used, which allows voters to influence the rank-order of the names on the list, and therefore the order in which seats will be allocated to party candidates. Turkey’s closed-list variant is less common. It has tended to feature in countries where democracy is relatively novel and/or shallow. The reasons for this should be clear. Closed lists produce an authoritarianism-friendly form of democracy, keeping power and control in the hands of party elites rather than individual voters.

Consider some of the negative effects that flow from the use of closed lists. First, as noted above, voters have no way of influencing the identity of the person who will represent them. Accordingly, the link between citizen and representative is weak at best. Second, parliamentarians are particularly strongly incentivised to bend to the will of their party leadership rather than to act in the electorate’s interest. Third, the calibre of party candidates is likely to be weaker than it could be, because leaders are free to promote weaker candidates on their lists in an effort to prevent rivals from emerging. Fourth, voters are left with no easy way to hold an individual politician to account by voting them out of office

Electoral systems are not the sexiest of business, but do not get enough attention. One of Turkey's chronic problems is the sheer strength that political parties wield over the system. Political leaders in Turkey are famed for staying around forever: once you are at the top of your party, there is little going away. This is, of course, also a problem of internal party democracy, but once leaders ascend to the position of party leader, they can virtually choose who will and will not be elected. Of course, this creates plenty of incentive for corruption, weak candidates with little popular appeal (think Deniz Baykal), and most importantly, a low level of democratic responsiveness to the people. The whole piece is worth a thorough read, and kudos to Aengus for writing it.

Prime Minister Erdogan has announced plans to draft a new constitution after the June 12 election, including a move to a presidential system. Why not an open-party list in addition?

Aengus also gives attention to parliamentary immunity, referencing Simon Wigley's equally cogent 2009 HRQ article on the topic. That one is also well worth the read.