Czechia: Constitutional Court rules lower house electoral system not proportional enough

The Czech Constitutional Court has ruled that the country’s current electoral system does not adequately fulfil the constitution’s requirement of being in accordance with “the principles of proportional representation” (article 18 of the Czech constitution). The 200 members of the Chamber of Deputies are currently elected under Flexible List-PR in 14 districts ranging in magnitude from 5 to 26, with a nationwide threshold of 5% for parties and 10% for alliance lists. The Constitutional Court struck down the districting scheme on the grounds that it disadvantages small parties, as well as the 10% threshold for lists of more than one party.

As an election is scheduled for October, Parliament will have to agree fairly quickly on a new districting scheme to replace the one the Court has struck down. Unusually, since the Senate usually only has a suspensive veto the Chamber of Deputies can override immediately by absolute majority, article 40 of the constitution requires the electoral law to be approved by consent of both houses.

What is somewhat ironic is that the case was brought to the Supreme Court by a group of 21 members of the Senate, a house which is not required to be elected by PR and is instead elected by runoff in single-seat districts (with elections to the Senate being fairly low salience and *very* low turnout, it has seen some success by minor parties despite the system’s lack of proportionality).

11 thoughts on “Czechia: Constitutional Court rules lower house electoral system not proportional enough

  1. Oh my, I hope the US is having legal fights like this in 20 years, some countries get all the good politics. Are there visible coalition benefits for the suing members of the Senate?


    • They are all from smaller parties, especially TOP 09, KDU-CSL, and the Mayors and Independents Parties, which are the smallest three parties to pass the 5% threshold and win seats in the last Chamber of Deputies election.

      Parties which are on the small side in lower house terms currently make up most of the Senate thanks to 1) their success in recent Senate elections and 2)defections.


      • To be more specific, TOP 09, KDU-CSL, and Mayors & Indies won respectively 5%, 3.5% and 3% of seats on 5.8%, 5.3% and 5.2% of the vote, respectively in 2017. The Pirate Party, whose two Senators also joined in the lawsuit, won 11% of seats on 10.8% of votes, so it’s actually slightly overrepresented.


  2. I’m confused here, was the threshold abolished here or was the the 14 districts (5-26 district magnitude) abolished and replaced with 1 200 member district?

    If the electoral threshold is abolished and the 14 districts with a 5-26 district magnitude continues, because the district magnitude isn’t that permissive for excessive fragmentation. Why isn’t there nation wide leveling seats? What European countries use List PR, but don’t have nation wide leveling seats?

    If it is 1 200 member district with a 5% threshold, how is that any different when it was regional 14 districts with a 5-26 magnitude a 5% threshold? Why a higher threshold for coalitions? How common is this under list PR?


    • Czechia uses a single-tier, fully districted PR system. There are no compensatory seats (although earlier in the post-communist period there were). The threshold, however, is applied nationwide. So even if a party has votes to elect a member in some district, it is ruled out if it does not pass the national threshold.

      Other districted single-tier PR systems would include Portugal (no legal threshold) and Spain (district-level threshold that matters only in the few high-magnitude districts). Poland and a few other ex-communist countries that follow the Czech model (districted, but threshold applied nationally). Some do so without such variance in M. Croatia, for example, has all districts of the same M, but it recall that it also applies its threshold nationally.

      I am understanding “leveling seats” to mean all the literature I can think of would call a “compensatory tier” or “supplemental seats” or “remainder-pooling” (although not all compensation is done through remainder pooling).

      The other issue in the Czech ruling seems to be the recent change in the threshold. My recollection is that it used to be 5% for a party, but 7% for a list consisting of two parties in alliance. The latter was raised to 10%.


  3. I checked Michael Gallagher’s database to find the figure on disproportionality in the past few elections. The Least Squares indices for the past four elections (2006, 2010, 2013, and 2017) were 5.7, 8.8, 6.1, and 7.2, respectively. I would call that fairly mild disproportionality. A major part of it is caused by the 5% threshold. From my own calculations, the disproportionality in the same elections when only taking parties over the threshold into account was 3.9, 5.2, 3.5, and 5.9, respectively.

    I will admit I have not read the full ruling on the matter, but my feeling is that the court was wrong to intervene, at least in the manner that it did. The Czech constitution merely requires the electoral system to be “proportional”. It does not specify how proportional it has to be, or any other detail. The Czech electoral system clearly qualifies as “proportional”, with the possible exception of the 10% threshold for alliances, which is the only part of the ruling I wouldn’t question. But even if the system had to be more proportional to meet the requirement of the constitution, the biggest single element causing disproportionality is probably the threshold, and it would be a far simpler matter to adjudicate a solution for without throwing the electoral code into disarray less than a year before an election.


    • Thanks for that information, JD. Yes, this is pretty modest disproportionality, in the wider scheme of things. On the other hand, it is high for a “proportional” system. It is on par for the bonus-adjusted system of Greece, for example, where the average from 1990-2017 is 6.94. The value for Spain, a districted PR system with fairly similar mean magnitude, in the same time period is 6.18.

      Yes indeed, it would be much easier (and a less extreme court intervention) to toss out the threshold for alliances, or to say that it is inappropriate to apply the threshold nationally instead of district-by-district, or some such solution.

      You might consider asking Kieran Williams (on Twitter) for his thoughts on this, as he knows the Czech context deeply.


      • That’s odd. Using Gallagher’s figures, I got an average of 8.4 for Greece and 5.7 for Spain (1990-2017). Spain obviously doesn’t have a nationwide threshold – if it did, its figures would be much higher.

        In Czechia, the last four elections saw between 6-19% of votes wasted due to the threshold. It would be easy to say the threshold should be no higher than 4%, and/or that no more than 10% of the total votes should be excluded from the count on account of the threshold, and the upcoming election could simply be held under that amendment to the system.

        Since, as Shugart & Taagepera (2017) find, while district magnitude affects the effective number of vote-winning parties (Nv), thresholds waste votes without moderating Nv very much, I think a future election under a revised districting scheme could actually see even higher disproportionality than now.


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 )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.