How should Ireland’s MEPs be elected amidst Brexit uncertainty?

How many seats will Ireland have in the new European Parliament due to be elected in late May? If the UK has left the EU by then, Ireland will have 13 seats, but if not then only 11 will be elected with 2 others waiting on the subs’ bench until the first stage of the Brexit process is completed. The 13 MEPs are due to be elected from three constituencies very similar to those used in 2014: an expanded South with 5 MEPs, a slightly reduced Midlands North-West with 4, and Dublin with 4. If only 11 are able to take their seats at once, the plan is that only 4 of the 5 South MEPs, and 3 of the 4 Dublin MEPs, will do so, with one MEP from each of those two constituencies having to wait until the UK MEPs cease to retain seats in the European Parliament.

 

All of this may be the least bad way to deal with an unprecedented and difficult conundrum, but the question arises: how will it be decided which two MEPs are handed the short straw and told to cool their heels while we wait for the UK to complete its exit from the EU? According to reports, the government will propose that the last elected MEP is to be deemed the weakest and hence will be the one who has to wait. Quoting from TheJournal.ie:

“The last two Irish European election candidates elected in the South constituency and the Dublin constituency will not be able to take up their seats in the European Parliament until the UK has left the EU.”

 

However, this approach shows a serious misunderstanding of the rationale of PR-STV, which is rather surprising given that it has been used at elections in this country for over a century (the first such election, for local government in Sligo, took place in December 1918). Under this electoral system, the sequence in which the MEPs are elected is irrelevant; none can claim any superiority over any other. Even if it sometimes seems that one candidate has ‘powered their way’ to election with a large surplus on the first count while another ‘scrapes into the last seat’, the fact is that once the election is over they all have equal status, and in fact the last elected candidate may in some sense have more support than the first elected.

 

Which candidate is elected first, second, third, fourth or last reflects not just their own innate support, as the government proposal seems to assume, but also several other variables including the number of running mates they have and the order of elimination of other candidates. A candidate with a running mate may well win fewer first preferences, and hence be elected later in the count, than a candidate without a running mate, even if the former is from a better supported party.

 

The point can be illustrated by an example. At the 2016 general election, in the 4-seat Waterford constituency, the breakdown of votes was

     Fine Gael 29%

     Fianna Fáil 21%

     Sinn Féin 19%

     Independent John Halligan 16%

     Labour 4%

     Green 4%

     AAA–PBP 3%

     Renua 2%

     Others 2%

 

The seats were divided 1–1–1–1 between FG, FF, SF and Halligan. The last candidate elected was John Deasy of Fine Gael, who finished 567 votes ahead of the runner-up, his running mate Paudie Coffey, on the ninth and final count. However, it is obvious that in no sense was Deasy the worst-supported of the 4 elected TDs. Quite the contrary: he was a candidate of the best-supported party, and that party would have been certain of winning a seat had the constituency been returning only 3 seats, in which case the combined votes of the two FG candidates would still have exceeded the quota (which would then have been 25%). If Fine Gael had run only one of those two candidates, he would have headed the poll with a sizeable first-count surplus. In those circumstances, allowing the other three elected TDs to take their seats and requiring Deasy to step aside for an indefinite period would have been positively perverse.

 

And this is not an exceptional case. There are other examples from the 2016 general election where the last candidate elected would undoubtedly have been elected had the constituency been returning one TD fewer while one of the candidates elected earlier in the count would not have been. That is certainly true of Cork South-Central, Cork South-West, Galway East and Laois, and probably of several others as well, such as Dublin Bay North, Dublin Bay South, Limerick City and Tipperary.

 

Deeming the last candidate elected to be the least-supported favours those with higher first preference support but perhaps less broad support over those who have less first preference support (possibly because their party’s votes are spread among two or more candidates) but broader support. It is not difficult to imagine that, if this provision actually has to be applied after the elections in May, the last candidate elected may well contemplate legal action on behalf of him/herself and their voters if denied election on such flimsy grounds.

 

The remedy is undeniably time-consuming. Put simply, the votes in the huge Euro-constituencies need to be counted twice: in the case of the South constituency, once to fill 5 seats (based on a quota of 16.67 per cent) and then again to fill 4 seats (with a quota of 20 per cent), and similarly for Dublin, a count based on 4 seats and a count based on 3 seats.

 

It may be that the Brexit process will have worked itself out by the second half of May and that no contingency plans will be required. But given the possibility that such plans might be required, the government needs to come up with a better idea than the current proposals.

6 thoughts on “How should Ireland’s MEPs be elected amidst Brexit uncertainty?

  1. Michael makes a convincing case.
    There is an interesting piece by Farrell & Katz (Representation 2014, I think) that makes the point that calculating vote shares on the first count does NOT give the same result as a calculation based on later (the last) count.

  2. Pingback: How should Ireland’s MEPs be elected amidst Brexit uncertainty? – Politics Article – Politics Blog

  3. Thanks for this interesting post. Two points:

    (1) If your suggestion is taken up, with (e.g.) in the South counts (a) to fill 5 seats and (b) to fill 4 seats, it is possible that at least one of the 4 elected in (b) might not be elected in (a). Philosophically, that is reasonable in that the best set of 5 to represent the electors’ views does not need to include the best set of 4 (think of 9 candidates of roughly equal popularity, with views differing mainly on a 1-D scale; then the odd-numbered candidates might make the best set of 5, the even-numbered ones the best set of 4). In practice, it’s not likely, but quite possible, and presumably would attract a lot of opposition if not legal challenge. One way round it is a nested election, as is done by the Liberal Democrats in selecting party candidates for list elections; for the present case, you would first elect 5, and then recount for an election of 4 excluding all candidates except the 5.

    (2) The work involved would be negligible if the votes were computer counted, as is done for local elections in Scotland. I know there was an experiment in Ireland in 3 constituencies in 2002, but not why it wasn’t followed up.

  4. Thanks for this interesting post. Two points:

    (1) If your suggestion is taken up, with (e.g.) in the South counts (a) to fill 5 seats and (b) to fill 4 seats, it is possible that at least one of the 4 elected in (b) might not be elected in (a). Philosophically, that is reasonable in that the best set of 5 to represent the electors’ views does not need to include the best set of 4 (think of 9 candidates of roughly equal popularity, with views differing mainly on a 1-D scale; then the odd-numbered candidates might make the best set of 5, the even-numbered ones the best set of 4). In practice, it’s not likely, but quite possible, and presumably would attract a lot of opposition if not legal challenge. One way round it is a nested election, as is done by the Liberal Democrats in selecting party candidates for list elections; for the present case, you would first elect 5, and then recount for an election of 4 excluding all candidates except the 5.

    (2) The extra work involved would be negligible if the votes were computer counted, as is done for local elections in Scotland. I know there was an experiment in Ireland in 3 constituencies in 2002, but not why it wasn’t followed up.

    [corrected]

  5. It’s a nice idea to solve the problem, certainly, but it wouldn’t work or at least is highly risky. Here’s the problem.

    First, look at the ‘no-problem’ scenario. The results of the two elections are the same, except for one, an extra person is elected in the count for four seats. All is well. When or if the contingent MEP is required in the Parliament, we wish them well, give them a kiss and send them on their way to Brussels or Strasbourg to join their three constituency colleagues.

    However, what if there are more differences between the two counts. So that a person who was elected on the count-for-three-seats is not also elected on the count-for-four seats. This is something that is reasonably likely to happen.

    So what do we do now, on the day the contingent MEP is required to come to parliament? We cannot then deselect one of the already-elected MEPs or somehow call them back to Dublin to be replaced with the new MEP. An MEP, once sent to Brussels, is there for the term of the parliament, unless he or she dies or quits.

    So now, you have an illegitimate MEP sitting in Parliament, and a legitimate MEP-elect who cannot take his or her seat. This is disastrous, and it is a very difficult situation to resolve legally and democratically.

  6. Do two counts with appropriate quotas?
    Do a count in the South for both 4 and 5
    Do a count in Dublin for both 3 and 4
    Extra cost I know – and as someone who is mainly British I am deeply deeply ashamed that British Selfishness & Incompetence is causing so much trouble in so many places that an STV election looks minor!

    But watch Carole Cadwalladr at Ted Talks on what went on during the referendum – https://www.ted.com/talks/carole_cadwalladr_facebook_s_role_in_brexit_and_the_threat_to_democracy#t-903263
    Watch it, check it
    Spread it, Protest it
    Think it; could it happen where you live?
    The British were screwed by forces bigger than them,

Leave a Reply to Ken Carty Cancel 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 )

Google photo

You are commenting using your Google 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