Northern Ireland assembly-size and magnitude change

A political agreement in late December in Northern Ireland includes a provision:

to cut the number of Assembly members from 108 to 90 and Executive departments from 12 to 9 and to allow for the formation of an opposition with effective speaking rights. In theory the creation of an opposition raises the possibility of a future alternative coalition. (Brian Walker, The Constitution Unit, 13 January)

Although Walker, in the cited piece, does not say so, it would seem this means district magnitude (M) will be reduced from six to five. There are currently 18 electoral districts, each electing six assembly members. Given a reduction of assembly size by 18, I will assume this means each district loses a seat.

The impact would be to make Northern Ireland’s elections somewhat less proportional, which is a goal consistent with Walker’s statement about the possibility of a future alternative coalition. That is, the political reforms, as a package, could be seen as representing a small step away from consociationalism (power-sharing) in the direction of majoritarianism (with its more defined government-opposition divide).

Based on the 2011 election results, one might surmise that the lone Green and TUV members, as well as the one independent, would be unlikely to retain their seats. The UUP would be unlikely to win two seats in Strangford on barely a fifth of the vote, as it did in the six-seat district, and might have trouble holdings its one seat in districts such as Belfast East, West Tyrone and Mid-Ulster. Sinn Fein likely would struggle to retain its seats in East Antrim (8.4% of vote) and Belfast South (12.5%). These are just examples of some potential losses, but one can’t be sure given that obviously the votes will not break down in the next election the same way as they did in the last, and I have not considered the distribution of preference flows (Northern Ireland uses STV). The general point is simply that larger parties, in any given district, should be somewhat favored by a reduction of district magnitude, particularly the move from an even number of seats in each district to an odd number. And this would tend to support, at least marginally, a move towards two alternative governing options, when combined with other changes.

15 thoughts on “Northern Ireland assembly-size and magnitude change

  1. This looks like a definite move away from consensus government for NI. Odd numbers of seats and an opposition will probably lead to majority government.

    Seeing as the Unionists have a majority, it is probably an idea that will irritate the Nationalists, and a Unionist dominated government could set off another round of fighting in NI.


  2. Northern Ireland gets a bad rep, but note that electoral politics in Mississippi has the same dynamic, voters line up according to racial (as opposed to confessional background), and the 60% majority racial group sticks together, so the opposition never wins. This is a particularly pathetic, but not uncommon, political dynamic.


  3. Looks less than revolutionary. Power-sharing rules continue. A party entitled to Executive (cabinet) seats may choose to form an opposition instead. Doubtful either of the two big parties would do so, unless they dismount in the run-up to an election. As for the reduction to five MLAs per district, this does not happen until the 2021 election.


  4. Doesn’t the STV allow of preference swamping between confessional backgrounds in Ireland? The problem with parliamentary democracy is that sometimes there is always a permanent majority government and permanent opposition especially if voting patterns are polarized. I guess there can never be a typical government party in power and opposition that is never in power.


      • I have read analyses (don’t have time to find them, sorry) on the number of moderate voters who prefer, for example, the SDLP over the DUP, or the UUP over Sinn Fein. And a significant number of women voters prefer cross-community women candidates over some men. Not to menton the voters for the Alliance and Green parties.


  5. “… the Northern Ireland case offers some evidence to the effect that STV assists the position of moderates in bicommunal societies, but this evidence is rather ambiguous. … Preference patterns are overwhelmingly single-peaked, and show certain striking traits: the centre tends to help and to be helped by moderate unionists, and has a similar relationship with moderate nationalists; and moderates within each communal bloc gain from radicals within the same bloc more than they give to the radicals.” – John Coakley and Jon Fraenkel (2010) : ‘Draft, not for citation’


  6. Matt

    You are correct – the proposal is to reduce the DM to 5 in each of the 18 constituencies. Which is a shame, since as you say it will be slightly less proportional. If there is a strong desire to reduce the number of MLAs they could have had fewer but larger districts. Another negative is that the Agreement does not proportionally reduce the number of MLAs needed to invoke a Petition of Concern (consociational veto powers) – it states that this will remain at 30 MLAs. So one-third of members would be needed.

    The Assembly still has to legislate for these changes and they are only envisaged for 2021, rather than the next scheduled election of 2016. So hopefully, it might never happen, especially once people likely to lose their seats figure out that ‘it might be me’!

    Text here:



  7. The consensus among electoral commissioners and academics in Australia is that it is desirable to have odd number magnitudes with STV. I am not aware of any even number magnitudes in Australian STV systems, although some may exist for local councils. The ACT electoral legislation is partly entrenched, a result of the referendums in the early 90s that abolished the notorious modified d’Hondt system, and one of the entrenched provisions is that all magnitudes be odd numbers.


      • Tom, I think that Alan is referring to bodies tasked with forming governments. In the Senate, it makes some sense to have even numbered district magnitude in order to prevent a single party majority, but for the ACT and Tasmanian Legislative Assemblies (elected by ‘pure’ STV), odd numbers work best, as it means that (generally) majority of votes=majority of seats, which is good for government formation.


      • And I should not hit the button so quickly.

        A majority of legislative councils, NSW, Victoria and SA use odd numbers.

        I actually know some of the people responsible for preparing the legislation increasing the senate to 12 from each state. They were aware pf the odd number issue but an increase to 14 was thought politically impossible and there would still be an even number in the event of a double dissolution when it would most count.

        I raised a point Tom made some time ago about electing 5 senators at one election and 7 at the next. Apparently the solicitor-general was consulted about the constitutionality of a 5/7 scheme and became, to quote my informant, ‘mildly hysterical’.

        I do not know if the issue was canvassed in WA.


    • I don’t know why this theory became common in Australia, but no such practice is found in Parliaments or local governments in Ireland or Northern Ireland, or local governments in Scotland.


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.