Review of Dail Constituencies (Part 5)

In this final part, (Part 1, Part 2, Part 3, Part 4) we come to the jewel in the crown, the crème de la crème, the pièce de résistance, Dublin. Ok, so it ain't all that, but for the amateur constituency maker it provides the greatest opportunity for creativity while still remaining within the guidelines laid down by the legislation. In keeping with the rest of my carve up, we will be maximizing the number of large constituencies. So lets take a look at how the 44 seats allocated to Dublin can be best divided to create eight 5-seaters and a single 4-seater.

To paraphrase Caesar, Dublinium est omnis divisum in partes duo. The Liffey forms the boundary between the Northside and the Southside and so we will follow suit and maintain this boundary. North of the Liffey there are just under 580,000 residents while just over 690,000 people live South of the river. Dividing the allocated seats proportionately we get 20 TDs for the Northside and 24 TDs for the Southside. This means we will create four 5-seaters on the Northside with four 5-seaters and a single 4-seater on the Southside. The map here shows my proposed boundaries.

North Dublin

Currently there are 6 constituencies with 21 seats in North Dublin. In the infamous Tullymander, the Dublin area was divided into many 3-seaters. Then for the 1981 election the boundaries were redrawn to create 4-seaters in urban areas. However, due to dropping population levels compared to the rest of the country, over the last 30 years each of Dublin North-East, Dublin North-Central and Dublin North-West have been reduced to 3-seaters, losing territory to neighboring constituencies to minimize the population variance. I propose that the current boundaries are not fit for purpose and suggest that it is now time for a major redrawing, ignoring the requirement for continuity as it is the least important requirement in the legislation.

Starting at the top, the major issue in Dublin in the last redraw was the division of Swords between Dublin North and Dublin West. I propose this be reversed and that Swords-Forrest rejoin Dublin North. The addition of Portmarnock, Balgriffin and the Airport will bring Dublin North up to the required level of population for a 5-seater. What remains of the existing Dublin North-East should then be merged with almost all of the existing Dublin North-Central to form another 5-seater which I am going to call Dublin North-East. The only transfer required here is the DED of Drumcondra South A out of the new constituency.

On the 2011 census figures, the existing Dublin Central already has just about sufficient population to warrant remaining a 4-seater in our reduced 160 seat Dáil. However, we are aiming for 5-seaters so we need to transfer in around a total of about 30,000 people which we will take from the existing Dublin North-West constituency along with Drumcondra South A which we have already ditched from the new Dublin North-East. By absorbing most of the Ballygall and Whitehall DEDs along with Ballymun C and Beaumont A we get a total population of just under 147,000 which is fine for a 5-seater.

This leaves the remains of old Dublin North-West (Finglas and Ballymun mainly) and old Dublin West (Blanchardstown and Castleknock). These areas combined have a population of almost 146,000 which again is just right for a 5-seater. I am calling this area Dublin West, but it could just as easily be called Dublin North-West. This completes the division of Dublin north of the Liffey.

South Dublin

Moving to the Southside, we need to create four 5-seaters and a 4-seater from the existing mix of two 5-seaters and four 4-seaters. This represents a loss of two seats and one constituency. There are several ways to carry out this division but I feel that the cleanest is to disband the existing Dublin South-Central constituency and divide its territory into its neighbours, bringing each up to the required population level.

Starting in Dún Laoghaire, we are short about 0.3 TDs worth of population to maintain the constituency as a 4-seater. I propose returning some of the DEDs on the western side of the N11 back from Dublin South that were transferred during the last boundary review, namely Foxrock-Carrickmines, Cabinteely-Loughlinstown and Shankill-Rathmichael. This provides an increase of about 15,000 in population which brings the constituency in line with the requirements for a 5-seater.

Having lost 15,000 to Dún Laoghaire, Dublin South finds itself needing a similar increase in population to keep its 5 seats. This can be achieved by adding Firhouse-Village. This allows us to retain the Dodder as the northern boundary of Dublin South for most of its length. This border also means that for Dublin South-East to move from a 4 to a 5-seater it must expand westwards into the old Dublin South-Central. This requires adding the DEDs of Merchant's Quay and Ushers as well as the areas of Crumlin C, Kimmage C and Terenures A and B. This serves to unite the south city centre into a single constituency, much as Dublin Central serves the entire north city centre, while also retaining some balance between high and lower density areas by uniting Harold's Cross and most of Terenure into a single constituency.

Dublin Mid-West then expands eastwards into Kilmainham, Inchicore and Crumlin as well as taking all of Clondalkin-Monastery. To keep the population within limits we need to transfer the semi-rural DEDs of Newcastle, Rathcoole and Saggart to Dublin South-West. This leaves Dublin Mid-West at 147,000 which is acceptable for a 5-seater. The transferred DEDs along with the remainder of the old Dublin South-Central (mainly Walkinstown and Templeogue) when added to Dublin South-West give us a constituency with just over 139,000 which, although slightly lower than average, is still acceptable as a 5-seater. It should also be noted that some of the DEDs in the new Dublin South-West had population increases of between 22% and 51% since the previous census and so it is expected that the variance will decrease over time.

Dublin Summary

The table shows the population and variance of each of the new Dublin constituencies. As we expect most of the constituencies have a slightly positive variance. This is due to the region being under-represented by 0.4 TDs which was felt a reasonable thing to do to increase slightly the representation in Connacht/Ulster earlier on.

Local Tweaks

One thing that needs to be checked is the precise alignment of the boundaries of constituencies with the boundaries of the DEDs. This is because, in most cases, the DEDs pre-date a lot of development, especially in suburban areas in Dublin. For example, in the map here, we have a close up of the boundaries between my Dublin South-East, Dublin Mid-West and Dublin South-West constituencies. The logical divider should be the Crumlin Road from the Canal to the Children's Hospital and then the Drimnagh Road and Long Mile Road as far as the Naas Road.

However, the marked area to the north of the Long Mile Road is actually part of the Walkinstown A and Clondalkin-Ballymount DEDs which straddle the main road. In this case it would be sensible to split the DEDs and put the parts on the north side of the road into Dublin Mid-West, leaving the parts on the south of the road in Dublin South-West. Due to the land use (mainly commercial and industrial) not much population would change with this modification but it would make the boundary between the Dáil constituencies much clearer.

Similarly the section of Crumlin Road near the canal actually bisects the Crumlin C DED. Again it may make more sense for the constituency boundary to follow Crumlin Road rather than turn left onto Herberton Road. However, this would result in some population shift between Dublin South-East and Dublin Mid-West so care would have to be taken not to put Dublin Mid-West over the 30,000 people per TD limit. This would be possible with access to the Small Areas database which is more fine-grained than the DEDs.

Conclusion

In this series I have shown that the Constituency Commission can tackle the problems associated with 3-seater constituencies without breaking too many of the guidelines set out in their terms of reference. The table shows that I have managed to reduce the number of 3-seaters from 17 down to 2. The quantity of 4-seaters has also been reduced from 15 to 11 while the number of 5-seaters has doubled from 11 to 22. These posts will be amalgamated into somewhat coherent English and submitted to the Commission in the coming days. I hope that they will follow my lead and try to consign the 3-seater to the dustbin of history where it belongs.