Voter turnout levels in general elections in the Republic of Ireland, as in most other democracies tend to be higher than those in other types of elections, such as referendum elections, presidential elections and local elections. The same also applies in the case of Dail by-elections; as Figure 1 shows turnout levels for by-elections held during the 1990s and 2000s in the constituencies concerned are consistently lower than those for the same constituencies in the general elections held prior to those contests.

Figure 1: Voter turnout levels by constituency for all by-elections held between 1994 and 2011 and turnout levels for the general elections preceding these

Turnout levels across the nineteen by-election contests held to date during the 1990s and 2000s have averaged out at 50.5%. This compares with an average turnout rate of 65.0% for the constituencies involved for the general elections immediately preceding the relevant by-election contests, suggesting that the average drop in turnout levels between a general election and a by-election amounts to an average of 14.5%. The level of decline tends to be more notable when by-elections are held mid-term – i.e. not in the first or last year of a government’s term in office – and the average decline in turnout levels for such contests amounts to 15.7% (average turnout of 48.4% for mid-term by-elections contrasting with average turnout of 64.1% for corresponding preceding general election contests).

As the patterns shown in Figure 1 suggest, however, there is no consistent pattern to the level of turnout decline between general elections and by-elections, with limited levels of decline applying in some cases while the drop in turnout levels in other cases far exceeds the average rate of decline.

During this period, the biggest level of turnout decline associated with any of these by-elections occurred in the case of the October 1999 Dublin South-Central, which was also associated with the lowest ever turnout level recorded for a Dail electoral contest. The turnout level for this by-election was 27.9%, representing a drop of 32.1% on the 60.4% turnout level recorded in the constituency for the general election immediately preceding this, that of the June 1997 General Election. There were two other instances during this period in which the turnout level fell by more than twenty percent for the by-election contests. The turnout level of 43.0% for the 1994 Dublin South-Central by-election marked a 21.6% decline on the 64.6% turnout level recorded for that constituency in the November 1992 General Election, while there was a similar level of turnout decline recorded in Dublin West between the November 1992 General Election and the April 1996 by-election.

The lowest level of turnout decline recorded between a general election and a by-election was associated with the old Mayo West constituency, in which the 63.4% turnout level recorded for the June 1994 by-election in this constituency marked a decline of just 5.7% on the 69.1% turnout level recorded in that constituency for the November 1992 General Election. The only two other by-election contests were associated with by-election turnout declines of less than ten percent were the April 1996 Donegal North-East by-election (turnout level of 60.7%, compared with a 67.6% turnout for the 1992 General Election), the June 2009 Dublin Central by-election (turnout level of 46.4%, compared with a 55.4% turnout for the 2007 General Election) and the June 2001 Tipperary South by-election (turnout level of 58.9%, compared with a 68.6% turnout for the 1997 General Election).

In geographical terms, voter turnout levels at by-elections are lowest in Dublin, with an average turnout level of 47.5% recorded for by-elections held in the region during this period, or of 46.4% when the neighbouring commuter belt constituencies of Meath, Kildare North and Wicklow are also included in this analysis. The average turnout decline between the preceding general election and the by-election is 16.5% in the case of the Dublin constituencies or 16.6% in the case of the larger area including Dublin and its commuter hinterland. By contrast, the average by-election turnout was 59.1% for the rural constituencies, with this marking an average decline of just 8.8% on the average turnout levels recorded for those constituencies (67.9%) in the preceding general election.

The level of turnout decline may not be as marked when other election contests are being held on the same day as a by-election – this can be evidenced in the cases of the last three by-election contests in Dublin, which were either held on the same day as the 2011 presidential election (Dublin West) or the 2009 local and European elections (Dublin Central and Dublin South). While by-election turnout levels usually tend to be lowest in Dublin and especially in relation to turnout levels in the preceding general election, the level of decline was not as marked as for the other Dublin contests in these three cases.

Based on turnout averages for electoral contests held during the 1990s and 2000s, there is a notable difference between turnout trends relating to the day of the week that a by-election is held on, with Wednesday contests appearing to be the least successful in terms of pushing by-election turnout levels up.

• Two by-elections during this period were held on a Tuesday and the average turnout levels for these was 52.3%, as compared with an average turnout level of 66.2% in the preceding general elections in the two constituencies concerned, marking an average decline of 13.9%.    
• Not including the upcoming Meath East by-election, three by-elections during this period were held on a Wednesday and the average turnout levels for these was 44.3%, as compared with an average turnout level of 63.5% in the preceding general elections in the two constituencies concerned, marking an average decline of 19.2%.    
• Eight by-elections during this period were held on a Thursday and the average turnout levels for these was 54.8%, as compared with an average turnout level of 68.0% in the preceding general elections in the two constituencies concerned, marking an average decline of 13.2%.    
• Five by-elections during this period were held on a Friday and the average turnout levels for these was 46.9%, as compared with an average turnout level of 61.3% in the preceding general elections in the two constituencies concerned, marking an average decline of 14.4%.    
• Just one by-election during this period was held on a Saturday and the turnout level for this was 58.9%, as compared with an average turnout level of 68.6% in the preceding general elections in the two constituencies concerned, marking an average decline of 9.7%.

With Croatia about to become the newest member of the European Union in the coming months and with this state likely to be assigned 11/12 seats in the next European Parliament, theEuropean Union has voted to reapportion European Parliament seats between its different member states to take account of this and the Republic of Ireland is to lose one of its 12 seats in the European parliament. A new boundary review will need to be carried out to redraw the European election constituency boundaries in line with this ahead of next year’s European elections. But what shape might these new boundaries take?

First of all, it must be noted that the current constituency configuration is one of four three-seat constituencies. With eleven seats, it will no longer be possible to have four constituencies, but dividing the state into two constituencies would not be an option either (as 6-seat constituencies are currently not permitted) – the basic option for the new Commission would be to create three constituencies, with an option to have one 5-seat constituency and two 3-seat constituencies or to have two 4-seat constituencies and one 3-seat constituencies.

With the reduction of one seat, the average population per MEP in the state would increase from 382,254 to 417,114. This would actually conform nicely to the creation of three-seat constituencies based on the provincial/regional boundaries, but for the fact that Connacht-Ulster is too small to be a three-seat constituency on its own. Dublin, the Rest of Leinster and Muster all have very similar populations (all just over 1.2m) and all these populations would be almost ideal in terms of the level of population required for three-seat constituencies in a scenario where Ireland has just 11 MEPs (as illustrated in Table 1 below). The population of Connacht-Ulster would be equivalent to that required for two seats, but of course two-seat constituencies are not an option. The option here might be to unite Connacht-Ulster with either Munster or (Rest of) Leinster, which would provide an area with an almost ideal population to sustain a five-seat constituency.

Table 1: Populations of provinces in Republic of Ireland as related to mean national population per MEP level if Ireland is allocated 11 European Parliament seats.  

The problem here would relate to the size of this constituency unit. For instance, a Munster-Connacht-Ulster constituency would be one in which distances of over 600km would be involved between places located on opposite ends of such a territory, e.g. between the tip of the Inishowen peninsula in Donegal and Schull in Co. Cork.

If the 5-seat/two 3-seat constituency option was to be kept, the Commission might prefer to have the five-seater in the area with the highest population density. In this case, the five-seat constituency could comprise of Dublin and other counties located along the East Coast and within the immediate Dublin commuting hinterland, including Louth, Meath, Kildare, Wicklow and Wexford. The other counties in Leinster could then be joined with Connacht-Ulster to form a three seater, while province of Munster would form the other three-seater (See Table 2a).

Table 2a: Populations of proposed constituency areas related to mean national population per MEP level if Ireland is allocated 11 European Parliament seats.

The Connacht-Ulster-West Leinster constituency would however have the same problems as the Connacht-Ulster-Munster, being too large a territory to be feasible, stretching almost from the north-western to the south-eastern corners of the island of Ireland. The option here instead might be to include Clare along with the Midland counties in the three-seat constituency including Connacht-Ulster (and indeed Clare is currently part of the three-seat North West constituency along with Longford and Westmeath). Kilkenny could then be added to the rest of Muster to form a new three-seat South constituency while Carlow could be joined with the Dublin-East Coast five-seater (see Table 2b).

Table 2b: Populations of proposed constituency areas related to mean national population per MEP level if Ireland is allocated 11 European Parliament seats.

The last option would be to have two four-seat constituencies and one three-seat constituency, with the latter to include Connacht-Ulster. In this scenario, Dublin could form a four-seat constituency along with Meath and Kildare. Munster and the South Leinster counties (Wicklow, Wexford, Carlow, Kilkenny) could join to get to form the other four-seater, a “South” constituency. The other counties in Leinster – Louth, Longford, Westmeath, Laois and Offaly – could then join with Connacht-Ulster to form the other (three-seat) constituency (see Table 3a).

Table 3a: Populations of proposed constituency areas related to mean national population per MEP level if Ireland is allocated 11 European Parliament seats.

The Commission could also have the option of including Louth with Dublin, Meath and Kildare, instead of the North-West and Midlands constituency (as shown above in Table 4). This would not produce as proportional a result as the option of joining Louth with the North-West and Midlands however and the population for the resulting North West-Mildands three-seat constituency would be just outside the acceptable limits for such a constituency unit (with a population per MEP level of more than 10% below the national average). This concern could probably be rectified by a simple swap in which Laois would be moved into the four-seat South constituency with Clare to be joined with the North-West and Midlands constituency (see Table 3b).

Table 3b: Populations of proposed constituency areas related to mean national population per MEP level if Ireland is allocated 11 European Parliament seats.

With further seat reductions possible if further member states join the European Union, there may be a need to consider whether the rules shaping the redrawing of European election boundaries needs to be revised (to possibly allow for larger constituencies – even a national constituency) or whether there may be scope for using an alternative set of electoral rules (e.g. a List system) in the specific case of European elections.

What do Tomas MacGiolla, Dick Roche, Joe Higgins, Brian Hayes and Paul Gogarty all have in common? All of these candidates lost in by-election contests – MacGiolla (Dublin West 1982), Roche (Wicklow 1995), Higgins (Dublin West 1996), Hayes (Dublin South Central 1994) and Gogarty (Dublin West 1996). But they all went on to win seats in subsquent general election contests – MacGiolla (Dublin West November 1982), Roche (Wicklow 1997), Higgins (Dublin West 1997) – and in some cases won these in different constituencies, namely Hayes (Dublin South-West 1997) and Gogarty (Dublin Mid-West 2002). Indeed, over the past thirty years and twenty-five different by-elections, a total of 28 candidates have lost by-elections but went on win seats in subsequent general elections.

In a number of cases these seats were won at the subsequent general election and sometimes at the expense of the candidate that won the by-election, as with the cases of Aine Brady in Kildare North (2007 General Election) and Billy Kelleher in Cork North-Central (1997 General Election). In other cases, the losing by-election candidates had to wait for somewhat longer periods before eventually winning a seat at a general election – in fact two candidates – Michael Mulcahy (Dublin South Central) and Dan Boyle (Cork South Central) – lost two by-elections in their home constituencies before both finally attained success at a general election in 2002. Another candidate in this category – Tomas MacGiolla of The Workers Party – would also lose two by-election contests in his home constituency of Dublin West but his second defeat (1996) came after he had held a Dail seat for a ten-year period (winning it for the first time six months after losing the May 1982 by-election – he also had lost the 1976 Dublin South-West by-election) before losing this in 1992 and he never subsequently was to regain that seat. The losing by-election candidate who had to wait the longest before eventually being elected to the Dail over the past three decades was Michael Conaghan, who lost the 1982 Dublin West by-election as a Democratic Socialist candidate, and had to win almost three decades after this until he finally won a Dail seat as a Labour candidate in Dublin South-Central.

With the exception of Dublin West 1982 (where three losing candidates – Lemass, MacGiolla, Conaghan – subsequently went on to win seats at general elections) and Galway East 1982 (Ulick Burke), losing by-election candidates in the 1980s did not generally tend to progress to subsequent Dail careers. But since 1994 there has only been one by-election in which no losing candidate has subsequently managed to win a seat at a later general election, the 2001 Tipperary South by-election. Even in this case two of the losing candidates have succeeded in starting careers in politics at the national (Denis Landy winning a Seanad seat in 2011) or European (Phil Prendergast co-opted to replace Alan Kelly in the European parliament in 2011) levels. Apart from the exception of Tipperary South, in all of the seventeen by-election contests held since 1990 there has been at least one of the losing candidates who has subsequently gone on to win a seat in Dail Eireann at a later general election, culminating in the case of Thomas Pringle finishing 4th in the November 2010 Donegal South-West by-election but going on to win a seat in that constituency some months later in the February 2011 General Election.

Thus parties selecting candidates for by-elections need to think strategically – not alone to choose a candidate who will be a contender for the by-election seat but also to choose a candidate who can use the experience/increased profile of the by-election contest to push to win a seat in that constituency at the next general election.

Ken Carty on Twitter poses the question as to whether by-election winners pose a threat to incumbent running mates in the general election after their by-election victory.

Table 1: By-election winners/runners up who win seats at the next general election at the expense of incumbent running mates

Indeed as Table 1 details, runners-up in by-elections seem to pose the greater threat to their incumbent running mates in subsequent general election contests, with three by-election runners-up and two by-election winners taking seats at the following general election at the expense of incumbent running mates. It is interesting to note that the two by-election winners were Fine Gael candidates who were defending Fine Gael seats in these by-elections, whereas all the by-election runners-up were Fianna Fail candidates who in two of the instances (Beverly Cooper-Flynn being the exception) were contesting by-election seats that had previously been held by other parties. The Dublin West 1982 by-election is the bext example here, in which the winner (Skelly) and runner-up (Lemass) went on to both won seats at the expense of incumbent running mates in the following November 1982 general election. Indeed, one of their running mates would probably have held their seats but for the fact that another by-election running mate, MacGiolla, went on to gain a seat for the Workers Party at this election at the expense of Fine Gael.

Candidates who were runners-up in a constituency in the previous general election tend to do well if they compete in a by-election. The general election runner-up has gone on to contest 18 (including Meath East) of the 26 by-elections held since 1980. General election runners up have won the by-election on six of these occasions, with the most recent examples being Patrick Nulty in Dublin West in 2011 and Pearse Doherty in Donegal South-West in 2010. In seven other cases, the general election runner-up has also ended up as the runner-up in the by-election, with Thomas Byrne in Meath East 2013 and Paschal Donohoe in Dublin Central 2009 being the most recent examples of this. All the candidates in this category subsequently went on to win a Dail seat in a later general election, which may augur well for Thomas Byrne’s chances at the next general election. As well as Donohoe, this group includes Dick Roche, Billy Kelleher, John Dennehy, Eileen Lemass and Dinny McGinley. The worst by election performance of a general election runner-up came from Labour’s James Somers, who was runner-up in Dublin Central in the November 1982 general election but only finished in fifth place in that constituency in the November 1983 by-election and would never contest another Dail election.

228 candidates have contested the 26 different by-election contests that took place over the past three decades (between 1980 and 2011), although this number does take account of candidates (such as Jim Tallon who has to date contested nine different by-election contests) who contested in different by-elections on multiple occasions. 40 female candidates have contested these different by-election election contests, meaning that females have accounted for 17.5% of all by-election candidacies over the past three decades. In total, female candidates have won 207.776 first preference votes in these contests (22.2% of all votes cast) and won the by-election contests on eightdifferent occasions (32% of all contests).

As well as the female candidates who have won by-elections on eight different occasions, other female candidates have attained relative levels of success in terms of attaining high placings in by-election contests. A female candidate has been runner up in a by-election contest on seven occasions – five of which involve Fianna Fail candidates, some of whom actually won the most first preference votes before losing out on the seat due to vote transfers – and has finished in third position in a further four cases. The 1998 Limerick East by-election saw the first three positions being filled by female candidates; Jan O’Sullivan (Labour), Sandra Marsh (Fianna Fail) and Mary Jackman (Fine Gael). With two further female candidates in the field at this contest, females nearly accounted for half of the total number of candidates (5 of the 11 candidates) and the total number of first votes won by females amounted to more than three-quarters of all the votes cast in the Limerick East by-election (32,846, or 76.9% of the total votes). Indeed, with the exception of the 1997 Presidential Election, this by-election probably represents the best performance by female candidates in a national election. Female candidates also won more than half of the valid votes cast (56.1%) in the 2005 Kildare North by-election, in which two female candidates, Catherine Murphy and Aine Brady, finished in first and second place in that contest.

Four of these ten female candidates who finished in either second or third position went on to win a seat in a later general election (Catherine Byrne, Beverly Flynn, Aine Brady, Eileen Lemass) but a number of these candidates did not go on to contest subsequent general elections following strong by-election showings, including Sandra Marsh (Limerick East 1998) and Sinead Behan (Cork South Central 1998).

The numbers of female candidates in by-elections increased notably following the Spring Tide 1992 General Election. In the by-elections held during the 1980s females accounted for only two of the 37 candidates contesting these (5.4%), although the two females involved admittedly performed well in these contests with Eileen Lemass (1982 Dublin West) and Mary Banotti (1983 Dublin Central) both being runners up in their respective contests and alone accounting for 13.0% of the total number of votes cast in these five contests. Thirty eight female candidates have contested the eighteen by-election contests held since 1992, meaning they accounted for 20.3% of the total (187) number of candidates contesting these. These female candidates tended to poll somewhat better than their male counterparts, winning 26.6% of the total votes cast in these by-elections with each female candidate winning an average of 4,824 votes against an average of 3,396 for the male candidates who contested these elections. The higher relative success levels of female candidates in the 1992-2013 period is reflected in the fact that females woneight of these eighteen contests while accounting for other positions in the Top 3 on nine other occasions.

While females tend to be more likely to be selected as candidates and to poll well in Dublin for general and local elections, there is not a strong Dublin emphasis in term of female by-election candidacies – indeed the percentage of female candidates in the Dublin region (15.7%) is somewhat lower than the national average (17.5%), although the share of the vote won by Dublin female candidates (23.0%) is somewhat closer to the share of the vote won by females (21.7%) across all by-election contests. The strongest region for female candidacies/support levels is the other City Council areas, or rather the cities of Limerick and Cork. In the four by-elections held in these areas over the past three decades, ten females have competed in these (amounting to 24.4% of the 41 candidates contesting these elections) and these candidates have won a total of 62,910 votes, amounting to 38.1% of all votes cast in these elections. While male candidates have won 3,299 votes on average in these contests, females have accounted for a staggering average of 6,291 votes. Not surprisingly these figures are driven by the strong performances by female candidates in the 1998 Limerick East by-election, but these also include strong performances by female candidates in the Cork City constituencies, including the 1994 Lynch victory in Cork North-Central and the strong Behan showing in the 1998 Cork South-Central by-election.

Since January 1st 1980, twenty-four Dail by-elections have been held in the Republic of Ireland involving two hundred candidates who have won a combined total of 816,989 votes in these. Fianna Fail has been the most successful party in terms of winning votes in these by-elections (287,258 votes, or 31.5% of the total number of votes cast in by elections, marginally ahead of Fine Gael (255,911 votes, or 28.0% of the total number of votes cast in by-elections). The candidate who has won the biggest number of first preference votes in any by-election held during this period is George Lee (27,768 votes, or 53.4% of the vote, in Dublin South 2009) , followed closely by former Taoiseach, Brian Cowen (26,022 votes, or 54.7% of the vote, in Laois-Offaly 1984). However, the late Cathal Coughlan (Donegal South-West 1983) is the candidate to win the highest vote share in any by-election during this period.As noted above, Fianna Fail has been the most successful of the different parties/groupings in winning support levels at by-elections over the past three decades, although Fine Gael has managed to win as many (seven) of these contests as Fianna Fail has. Both parties contested each one of the twenty four by-elections held during this period. The share of the vote won by Fianna Fail in the first decade of this period (46.9%) is well in excess of that parties share of the by-election vote over the two following decades, with average support levels of 28.0% in 1990s and 22.8% in the 2000s. This can in part be explained by losses in by-election support during periods when the party was in government (and the party was in government for all bar two years during the 1990s and 2000s) but this is not the whole story here. Fianna Fail’s share of the vote in the 1982 by-elections (50.2% in Galway East, 39.7% in Dublin West) as well as the 1980 by-election in Donegal (39.0%), when the party was in government, compares well with its overall average, while the party’s share of the vote in the by-elections held during the period it was out of government in the 1990s tends to be lower than its overall average by-election support level. Fianna Fail won 37.5% of the votes cast in by-elections fought as a non-government party, against an average of 29.2% for by-elections fought as a government party. So overall loss of support is evident for periods when the party is in government, but there is also an overall decline in Fianna Fail by-election support from the 1990s onwards.

The best performance by a Fianna Fail candidate in a by-election over this period was Cathal Coughlan’s 56.5% vote share in Donegal South-West 1983, while Fianna Fail candidates have also won more than 50% (and been deemed elected on the first count) on two other occasions: 54.7% in Laois-Offaly 1984 (Cowen) and 50.2% in Galway East 1982 (Treacy). The worst performance by any Fianna Fail candidate was Maurice Ahern’s 12.3% vote share in the Dublin Central by-election in 2009.

Out of government the average Fine Gael by-election vote share (30.2%) tends to be larger than for periods when the party was in government (average of 22.2%). Like Fianna Fail, Fine Gael’s 1980s by-election vote share (34.6%) tends to dwarf that of its average vote share during the 1990s (22.0%), although their share of the vote increases again during the 2000s (30.2%). The growing strength of non-party and smaller party candidates during the 1990s and 2000s can be seen as eating into the combined Fianna Fail/Fine Gael vote shares. George Lee won the most votes, and largest vote share, of any Fine Gael candidate in a by-election during the 1980-2011 period. The next largest number of votes won by a Fine Gael by-election candidates, ironically, were won by losing candidates; Dinny McGinley won 20,022 votes when finishing in second place in the 1980 Donegal by-election and Padraig Horan won 18,173 votes in Laois-Offaly 1984 when finishing second behind Brian Cowen. The same applies to the candidate who won the second highest vote share of any Fine Gael candidate during this period, Ulick Burke who won 41.8% of the vote when finishing second behind Noel Treacy in the 1982 Galway East by-election.

The Labour Party trails some distance behind these other parties, having contested every by-election bar Donegal 1980 and Donegal South-West 1983 and accounting for 107,749 votes or 11.8% of the total by-election valid votes cast over the 1980-2011 period (with this increasing to 13.1% when the two by-elections Labour failed to contest – Donegal 1980 and Donegal South-West 1983 – during this period are not included). Being in government seems to adversely influence the Labour by-election vote to an ever greater degree than Fianna Fail. Out of government the average Labour by-election vote share (14.4%) is significantly larger than for periods when the party was in government (average of 8.3%). However, unlike Fianna Fail and Fine Gael, Labour’s by-election vote share has improved in recent decades (periods when the party was mainly out of government), with its dismally low 1980s by-election vote share (2.5%) being decidedly lower than the share of the by-election vote won by the party in the 1990s (14.6%) and 2000s (16.2%). Sean Ryan accounts for the biggest by-election vote share of any Labour candidate, winning 11,012 votes (33.3%) in the 1999 Dublin North by-election. The next highest number of votes won by Labour Party candidates were won by Jan O’Sullivan in Limerick East in 1998 (10,619) and Patrick Nulty in Dublin West in 2011 (10,264), while the second highest vote share won by a Labour candidate during this period was won by Mary Upton in the 1998 Dublin South-Central by-election (28.0%). The lowest number of votes (and share of the vote) won by any Labour Party candidate was the 703 votes (1.6%) for Brendan O’Sullivan in Dublin West 1982.

Sinn Fein candidates, who contested just over half (14) of the by-elections held since 1980 won 41,598 votes (with Pearse Doherty’s 2010 Donegal South-West vote (13,719) accounting for almost one-third of all these votes to date) and 4.6% of the total number of valid by-election votes cast during this period. The percentage Sinn Fein vote share increases to 8.4% when the analysis only focuses on the by-elections that the party contested and the Sinn Fein by-election vote share increases even further to 14.2% when focusing solely on by-elections contested by the party during the 2000s. This stresses the point that the party’s performances in recent by-election contests have been much stronger than those in earlier contests held during the 1980s and 1990s.  Pearse Doherty accounts for the biggest by-election vote share of any Sinn Fein candidate, winning 13,719 votes (39.9%) in the 2010 Donegal South-West by-election. The next highest number of votes (and percentage vote share) won by a Sinn Fein candidate in a by-election was won by Joe Reilly in the 2005 Meath by-election (6,087 first preference votes – 12.2% of the valid first preference vote).

Green Party candidates contested the same number (14) of the by-elections held since 1980 as Sinn Fein did, but won fewer votes, winning 27,807 votes or 3.0% of the total number of by-election votes cast during this period. The percentage Green Party vote share increases to 5.3% when the analysis only focuses on the by-elections that the party contested, but the Green Party by-election vote share is only 4.0% when focusing solely on by-elections contested by the party during the 2000s. Despite contests being held in constituencies (Dublin North, Dublin South) where the party held seats at the time of a by-election, the party’s two best by-election performances are both accounted for by Dan Boyles in his 1994 and 1998 Cork South-Central by-election contests, with Boyle winning 6,677 votes (15.8%) in the 1994 election. Apart from Dan Boyle, no other Green candidate has won more than two thousand first preference votes or more than 7.0% of the total votes in any by-election contest.

The Socialist Party have only contested three by-elections during this period – Dublin West 1996, Dublin North 1998 and Dublin West 2011 – and won a total of 16,977 votes across the three contests, accounting for 1.9% of all valid by-election votes cast during the 1980-2011 period but accounting for a significant 17.5% of all valid votes casts in the three by-election contests that the party contested. Ruth Coppinger’s 7,542 votes in the 2011 Dublin West contest accounts for the largest number of first preference votes won by a Socialist Party candidates across these contests, but the highest percentage vote share won by a Socialist Party candidate was won by Joe Higgins (23.7%) in the 1996 Dublin West by-election.

Ironically the most successful period for government parties in terms of by-election performances was the GUBU period of 1982. A Charles Haughey-led Fianna Fail succeeded in retaining their seat in Galway East (with Noel Treacy exceeding the quota on the first count, despite a 4.9% drop in Fianna Fail support levels in the February 1982 general election). Fianna Fail also came close to taking a seat off Fine Gael in Dublin West. In the Dublin West contest (the background to which probably warrants a number of separate posts!!!) Fianna Fail support only dropped by 2.0% relative to the February election – statistically the second best performance by a government party candidate during the 1980-2011 period – and the Fianna Fail candidate, Eileen Lemass, topped the poll some 328 votes ahead of Fine Gael’s Liam Skelly, but ultimately lost out on the seat due to transfers. One of the best electoral performances of any government party candidate during the 1980-2011 came in the very first election held during this period when Fianna Fail’s Clement Coughlan held the party’s seat in the Donegal constituency and his 39.0% share of the vote marked a 2.6% improvement on Fianna Fail’s share of the vote in that constituency in the preceding 1977 General Election. This was the only time during this period that a government party actually increased its share of the vote in a by-election contest, relative to that in a preceding general election in the same constituency.

The largest vote share won by government party candidates was in the 1982 Galway East by-election where Fianna Fail’s Noel Treacy won the seat while winning 50.2% of the first preference votes. This is the only instance since 1980 in which government parties candidates won more than half of the votes cast in a by-election contest also. The next best electoral performances for government party candidates in terms of the percentage share of the vote won by these were associated with the 1984 Laois-Offaly by-election in which the Fine Gael and Labour candidates combined won 41.8% of the first preference votes and the 1994 Mayo West by-election in which the Fianna Fail and Labour candidates combined won 41.4% of the first preference votes. But no one government party candidate won more than 40% of the vote in either of these contests – Fine Gael’s Padraig Horan won 38.2% in Laois-Offaly in 1984 and Beverly Cooper-Flynn won 37.7% in Mayo-West in 1994.

The least successful government in terms of by-election performances was the Brian Cowen led Fianna Fail-Green Party-Progressive Democrat* government performances in the three by-elections held during the lifetime of that government – the Dublin South and Dublin Central by-elections in 2009 and the Donegal South-West by-election in 2010. In all of these contests, government party support levels fell by more than thirty percent relative to their results in the preceding election, dropping by 30.7% in Donegal South-West 2010, 35.6% in Dublin Central 2009 and 36.5% in Dublin South 2009. (Dublin South represents the worst ever performance by government parties in a by-election over the past three decades and to make matters worse the support levels for government party candidates fell by a further 5.1% at the subsequent general election contest in that constituency.) Naturally, government party candidates were not in contention for seats in any of these contests (and indeed no government party candidate succeeded in winning a seat in any of these constituencies in the February 2011 General Election) although Brian O Domhnaill did succeed in finishing second on first preference votes in Donegal South-West before being overtaken by Fine Gael’s Barry O’Neill on the basis of transfers.

In terms of percentage vote share, the worst by-election for government party candidates during the 1980-2011 period was the 2009 Dublin Central contest where the combined vote share won by the Fianna Fail and Green Party candidates was 15.1%, with the combined performance of the government party candidates (Fine Gael and Labour) in the 1996 Dublin West contest being only marginally better than this at 16.8%.

Table 1: Average support levels for government parties in by-elections

On average, government party candidates tend to poll best in by-election contests held during the first year of their government’s terms in office, when they presumably may be enjoying a honeymoon period. The average level of declining support at by-elections in these cases was just 10.0%, contrasting with the average level of decline of 14.5% for all by-election contests held during the 1980-2011 period. The Galway East by-election win by Fianna Fail in July 1982 fell within the first few months of the short lived Fianna Fail government’s term in office, as indeed was the case with Dublin West by-election win by Labour in October 2011. As Table 1 above shows government parties are likely to perform best in by-elections if these are held in the first year, or during the honeymoon phase, of that government’s term in office. The worst time for government parties to face by-elections is during the middle of that government’s term in office, with average support levels being much lower than what these parties may expect to win in the first or last year of that government’s term in office.

At a regional level, government parties tend to perform worse in by-elections in Dublin than in any other region, as Table 1 above shows, with the best by-election performances by government parties being associated with the Leinster region.

So why do government parties fare so poorly at by-election contests? Anger or disappointment at government party performances may be one factor. This ties in with the second-order election model, which argues that at election contests (such as by-elections, European elections etc.) which voters perceive to be less important than first-order elections (e.g. general elections) voters will be more likely (a) not to vote, (b) to vote against government party candidates, even if they are government supporters, to ‘safely’ send a message to the government. Another factor explaining a drop in government party support (and applying in the cases of Fine Gael and especially Fianna Fail) relates to number of candidates – if that party has run three or four candidates in a preceding general election it is probably not too surprising that their support level will drop at a by-election where only one party candidate is contesting. Moreover, while Fine Gael or Fianna Fail will expect to poll very well in the areas where their by-election candidates are based, the party’s vote share would probably decline in other areas of the constituency, which might have had a strong local candidate from those parties competing in the general election. Finally, if the local electorate may feel that a government win will leave too many government party TDs in the one constituency they may opt to vote against that party in the by-election as a result; this it could be argued could have been a factor in the 2005 Meath by-election where a win for Fianna Fail’s Shane Cassells would have left Fianna Fail with four of that constituency’s five TDs.

Figure 1: Voter turnout for the 2012 Childrens Referendum by electoral division for the Dublin City constituencies

This post can be viewed as a follow on from the previous two posts, which studied the geography of voter turnout for the 2012 Fiscal Stability Treaty referendum at a sub-constituency level for the Dublin City constituencies, as well as the national turnout geography (at a constituency level) for the more recent 2012 Childrens Referendum.

As noted in the previous post, one of my main research interests (and the one I focused on for my PhD researches) focuses on the geography of voter turnout and especially on voter turnout levels at the micro-level involving studies of turnout data for very small areas/at a sub-constituency level where these are available. Constituency level turnout analysis/maps do point towards interesting trends but the usefulness of such figuress is limited by small numbers problems arising with these (forty-three general, or Dail, election constituencies at presenting, falling to forty following the next general election). The most striking turnout geographies also occur usually at the sub-constituency level and significant variations in turnout levels often exist within a single constituency. Studies of such geographies/trends can offer a deeper understanding of what factors are shaping/influencing turnout propensity and can add to/complement individual level studies of turnout behaviour engaged in by those in the field of political science.

In this post I am going to look at turnout in the six Dublin City constituencies at the electoral division level for the November 2012 Children’s Referendum. This study is based on my own geographical analysis of the excellent turnout by polling box figures that the Dublin City Returning Officer has made available from their own website, involving the calculation of electoral division level turnout figures based on these and the mapping of these figures (as was illustrated in Figure 1 above).   

The map of voter turnout (Figure 1) for this election shows a similar geography to that of other previous electoral contests, as can be evidenced with reference to the recent post on Dublin City voter turnout levels at the electoral contests held during 2011. and indeed the previous post which focused on turnout levels for the Fiscal Stability Treaty referendum. The area with the most notable concentration of low turnout electoral divisions is once again the Dublin Inner City, as was the case for the Fiscal Stability Treaty referendum and was the case with all the electoral contests held during the Celtic Tiger and post-Celtic Tiger eras, namely the late 1990s and 2000s. The Dublin Inner City has proven to be THE low voter turnout area nationally, as well as within the Dublin region, at all electoral contests held over the last decade and a half. This is because high levels of the different factors associated with low turnout propensity tend to be found in the inner city area, such as rented housing, residential mobility, social deprivation, single people and younger mobile populations. The level of non-voting behaviour in the inner city suggested by this map is inflated somewhat by the fact that it was not possible to exclude those on the register who could not vote in this election (i.e. all those who are not Irish citizens) from the polling box electorate figures. But previous analyses in which this has been possible to do so (i.e. exclude ineligible voters) shows that while the recalculated turnout levels would be some percentage points higher, there would still be strong and consistent evidence of a sink of low voter turnout levels focusing around the inner city areas. As part of my PhD researches, I did focus particularly on low inner city turnout levels and part of this work (done to support the political mobilisation efforts of community groups in the South West Inner City and other similar areas) was published as Unequal Participation, Unequal Influence by the South West Inner City Network in 2012.  

The other low turnout areas highlighted in Figure 1, as with other previous studies of turnout in the Dublin City area, include the working class areas of Ballymun and Cherry Orchard, as well as Darndale and the Northern Fringe area in the north-eastern region of the map and the Rathmines flatland areas. (The Phoenix Park area (Phoenix Park electoral division) also emerges as a low turnout area.) By contrast, the higher turnout areas tend to be mainly middle class and settled (i.e. high levels of residential stability) areas, including Raheny, Sutton, Clontarf, Drumcondra and Glasnevin areas in the northern part of this area, as well as the areas around Terenure and Templeogue in the southern part of the map.

Figure 2: Social areas within the Dublin City constituencies based on a factor analysis of Census 2011 variables

The linking between relative levels of affluence and turnout propensity is underpinned by a comparison between the turnout map and the above map showing a division of Dublin into different social areas, based on the calculation of relative levels of affluence drawn from a factor analysis by Adrian Kavanagh of affluence/deprivation related variables drawn from the 2011 Census.  This is a rather different/simpler way of analysing/representing different levels of affluence/deprivation within an area to the 2011 Pobal HP Deprivation Index, which is discussed in more detail in this post on the Ireland After NAMA website. Figure 2 above divides the Dublin City constituencies into five different social areas. Rank 1 areas prove to be the most affluent 20% of electoral divisions within the Dublin City constituencies and the Rank 2 areas are the next 20% most affluent electoral divisions, with the Rank 5 areas including the 20% least affluent (or most deprived) electoral divisions arising from this analysis. A quick comparison of the two maps (Figures 1 and 2) suggests a strong link between social well-being and turnout propensity for this electoral contest (and the same can be applied to previous electoral contests) with Rank 1 areas tending to be associated with higher turnout propensity and Rank 5 areas with lower turnout levels. This linkage is made even clearer in Figure 3, which shows average turnout levels by social area for three of the most recent referendum election contests, namely the 2009 Lisbon Treaty, 2012 Fiscal Stability Treaty and 2012 Childrens referendum votes, as seen below.

Figure 3: Average turnouts by social area for the Dublin City constituencies area for the three most recent referendum votes.

Figure 3 shows that the more affluent areas constistently tend to have the higher average turnout levels for each of the different referendum contests (as well as pointing towards the signficantly lower turnout levels in the Childrens Referendum as compared with the two previous referendum contests), with turnout propensity decreasing in line with reduced affluence levels. On average, the analysis suggests that voter turnout levels in the most affluent parts of the City (the “Rank 1″ areas) tended to be constistently ten percentage points higher than turnouts in the least affluent (or more socially deprived) parts of the study area (the “Rank 5″ electoral divisions). It is worth noting also that this study is focused on averages and the gap in turnout levels between individual higher turnout middle class areas and lower turnout working class areas can prove to be even higher, as is indeed illustrated by the turnout map (Figure 1) above.  

While this finding is specific to referendum contests, the previous post and indeed a number of the other voter turnout themed posts have all pointed towards the existence of class-based turnout patterns in the Dublin area and indeed have stressed the serious problem of very low turnouts in the Dublin Inner City area (and indeed in the inner cities of the other Irish cities). Given the strong focus on political reform in recent years, the issue of area-specific low voter turnout levels should be a major focus of such political reform efforts as lower participation levels amongst certain social groups and in certain areas does diminish the representativeness of our political system.  But this is an issue that the political elite and the political chattering classes have tended to ignore, as well as the media given the limited coverage of such concerns. Given the marginalised nature of the areas and groups concerned in relation to the problem of low turnout levels, it is perhaps not hard to understand why. The strong likelihood is that such striking geographical differences in voter turnout levels will remain part and parcel of Irish politics over the coming years, and with such unequal levels of participation so too will unequal levels of influence remain a characteristic of the Irish political system.

As someone who has been interested in the area of voter turnout since commencing PhD researches on this topic back in…gulp…1999 and as a geographer, I am always interested in attaining turnout data for very small areas/at a sub-constituency level. While constituency level turnout analysis/maps can uncover interesting trends and point towards certain factors impacting on these, there are however only a small number of these (forty-three general, or Dail, election constituencies at presenting, falling to forty following the next general election) meaning that any analyses of turnout at this level faces issues to do with small-numbers problems. It is also the case that the really interesting turnout geographies occur usually at the sub-constituency level and there can be significant variations in turnout levels within a single constituency – analyses of such geographies/trends can offer a deeper understanding of what factors are shaping/influencing turnout propensity and can add to/complement individual level studies of turnout behaviour engaged in by those in the field of political science.

In this post I am going to look at turnout in the six Dublin City constituencies at the electoral division level for the last referendum prior to last weekend’s Children’s Referendum. This study is based on my own geographical analysis of the excellent turnout by polling box figures that the Dublin City returning officer makes available from their own website, involving the calculation of electoral division level turnout figures based on these and the mapping of these figures (as in Figure 1 below).   

Figure 1: Voter turnout by electoral division (%) in the Dublin City constituencies for the 2012 Fiscal Stability Treaty referendum

The map of voter turnout (Figure 1) for this election shows a similar geography to that of other previous electoral contests, as can be evidenced with reference to a recent previous post on Dublin City voter turnout levels at the electoral contests held during 2011. The areas with the most notable concentration of low turnout electoral divisions is again the Dublin Inner City, as with the electoral contests held during 2011 and indeed as with all electoral contests held during the Celtic Tiger and post-Celtic Tiger eras, namely the late 1990s and 2000s. The Dublin Inner City has proven to be THE low voter turnout area nationally, as well as within the Dublin region, at all electoral contests held over the last decade. This is because high levels of the different factors associated with low turnout propensity tend to be found in the inner city area, such as rented housing, residential mobility, social deprivation, single people and younger mobile populations. The level of non-voting behaviour in the inner city suggested by this map is inflated somewhat by the fact that it was not possible to exclude those on the register who could not vote in this election (i.e. all those who are not Irish citizens) from the polling box electorate figures. But previous analyses in which this has been possible to do all consistently also point towards a sink of low voter turnout levels focusing around the inner city areas.  

The other low turnout areas highlighted in Figure 1, as with other previous studies of turnout in the Dublin City area, include the working class areas of Ballymun and Cherry Orchard, as well as the Northern Fringe area in the north-east of the map and the Rathmines flatland areas. (The Phoinex Park area (Phoinex Park electoral division) also emerges as a low turnout area.) By contrast, the higher turnout areas tend to be mainly middle class and settled (i.e. high levels of residential stability) areas, including Portmarnock, Clontarf and Glasnevin areas in the northern part of this area, as well as the areas around Terenure and Templeogue in the southern part of the map.

Figure 2: Social areas within the Dublin City constituencies based on a factor analysis of Census 2011 variables

The linking between relative levels of affluence and turnout propensity is underpinned by a comparison between the turnout map and the above map showing a division of Dublin into different social areas, based on the calculation of relative levels of affluence drawn from a factor analysis by Adrian Kavanagh of affluence/deprivation related variables drawn from the 2011 Census.  This is a rather different/simpler way of analysing/representing different levels of affluence/deprivation within an area to the 2011 Pobal HP Deprivation Index, which is discussed in more detail in this post on the Ireland After NAMA website. Figure 2 above divides the Dublin City constituencies into five different social areas. Rank 1 areas prove to be the most affluent 20% of electoral divisions within the Dublin City constituencies and the Rank 2 areas are the next 20% most affluent electoral divisions, with the Rank 5 areas including the 20% least affluent (or most deprived) electoral divisions arising from this analysis. A quick comparison of the two maps (Figures 1 and 2) suggests a strong link between social well-being and turnout propensity for this electoral contest (and the same can be applied to previous electoral contests) with Rank 1 areas tending to be associated with higher turnout propensity and Rank 5 areas with lower turnout levels. This linkage is made even clearer in Figure 3, which shows average turnout levels by social area for the recent general and presidential election contests, as well as for the Fiscal Stability Treaty referendum vote, as seen below.

Figure 3: Average turnout levels by Dublin City social areas for electoral contests held in 2011 and 2012

Figure 3 shows that the more affluent areas constistently tend to have the higher average turnout levels for each of the different contests, with turnout propensity decreasing in line with reduced affluence levels. What is interesting to note here is that turnout levels for the presidential election in 2011 tended to be more sensitive to class differences than those in the general election held during the same year (as was discussed in a previous post on this topic) but the same acceleration of turnout decline for the lower ranked areas is not as evident in the case of the Fiscal Stability Treaty referendum. This is an interesting finding, given that the previous findings of my researches on turnout differences between turnouts in general election and referendum elections have tended to prove that class-related turnout differences tend to be more evident in referendum contests than they are for general elections (and also local elections). This is further illustrated by Figure 4 below, which shows a relatively similar trend in terms of turnout decline between the general election and the referendum vote across the Dublin City constituencies area, with no notable spike in turnout decline in the more working class areas being evident.

Figure 4: Turnout decline (%) by electoral division between the 2011 General Election and 2012 Fiscal Stability Treaty Referendum in the Dublin City constituencies area

Does this mark a trend towards relatively higher turnout propensity amongst the more working class areas for referendum elections, or could this be a “one-off” relating specifically to the nature of the referendum issue? It is hard to say, but a similar analysis of turnout levels in last weekend’s Children’s Referendum may go some way towards answering that question. I hope to engage in such an analysis as soon as I can after the voter turnout by polling box figures for that election are made available from the Dublin City returning officer’s website.

Figure 6: Voter turnout (%) by electoral division in the Dublin City constituencies area for the 2011 General Election

The Children’s Referendum vote, which took place on Satuday (10th November 2012),  was passed on a lower than expected (but rather comfortable) margin of 615,731 votes (58.00%) to 445,863 (42.00%). This referendum vote had the third lowest ever turnout level of any referendum vote held since the foundation of the state (see Figure 1), with an overall national turnout level of 33.49% and turnouts levels as low as 23.81% and 24.47% being recorded in the constituencies of Donegal South-West and Donegal North-East, respectively.

Figure 1: Turnout levels in all referendum election contests held since the foundation of the state (1937-2012)

Following this contest, the average turnout level across all referendum contests held since the foundation of the state now stands at 52.11%.

Was the low turnout down to holding this referendum on a Saturday? I would argue that was not the case, except maybe for certain places with high levels of residentially mobile populations who may tend to be elsewhere at the weekends  (i.e. “home”, visiting family) other than their place of residence, and believe that the turnout level would have been even lower has this vote taken place on a Wednesday or a Thursday. Indeed it is worth noting that Ireland’s limited experience of Saturday voting (which, to date, has never been employed yet for a general or local election contest) has been one in which the Saturday voting brought about a notable improvement in turnout propensity. The clearest example of this relates to the two Nice Treaty referendum contests in the early 2000s. The first of these was held on a Thursday in 2001 and was associated with one of the lowest turnouts for a referendum vote in the history of the state (34.79%, the third lowest referendum turnout in the history of the state prior to last weekend’s vote). The second Nice Treaty vote took place little more than a year after the first referendum; this was held on a Saturday and the turnout for this (49.47%) was almost fifteen percent higher than that for the first referendum. The two Tipperary South by-elections held during this same period also adds further proof of a higher turnout effect for Saturday voting, albeit to a less dramatic degree. The first by-election was held on Thursday 22nd June 2000 and the turnout for this was 57.97%, while the second by-election, held on Saturday 30th June 2001, had a somewhat higher turnout level (58.50%) despite this probably having been a less competitive/lower profile contest than the first by-election, the level of voter fatigue that would have been expected and the fact that a general election was due to be held in less than a year. Evidence from political science literature and from other countries also supports the positive impact weekend voting has on turnout levels. Ultimately blaming Saturday voting for the low turnout at the weekend is akin to blaming the lifeboats for the sinking of the Titanic - in my opinion we could have been looking at a national turnout level in the mid-to-high 20s (in a similar vein to the 1979 Adoption/Seanad voting rights and 1996 Bail referenda) had the contest been held on a weekday as is usually the norm.   

Figure 2: Voter turnout (%) levels by Dail constituency for the 2012 Children’s Referendum

In geographical terms, the voting trends observed for this referendum contest are quite similar to those observed for recent referendum election contests, despite the fact that the issues invovled in this vote were very different to those of the recent, mainly EU focused, referendum contests (Figure 2). Turnouts again proved to be highest in the more middle class urban constituencies, with the highest turnout level at a constituency level being recorded in Dublin North-Central (42.03%), followed by Dun Laoghare (41.54%) and Dublin South (40.92%). Outside of Dublin the higher turnout levels were found in the immediate Dublin commuter hinterland (the Wicklow and Kildare North constituency) while the highest rural turnout levels were associated with the counties of Tipperary and Cork. In keeping with the pattern for other referendum votes, the lowest turnout levels were again found in the Donegal constituencies (as noted above), with low turnout levels also being record in other border and western counties, such as Kerry, Galway and the Cavan-Monaghan constituency.

Figure 3: Yes vote (%) by Dail constituency for the 2012 Children’s Referendum

In general, constituencies with higher turnout levels, such as the Dublin middle class constituencies, also tended to record the highest percentage Yes votes. The Yes vote exceeded seventy percent in three constituencies in this election, namely Dublin South (73.03%), Dublin South-East (71.87%) and Dun Laoghaire (71.57%). Overall, there was a highest support level for the Yes side in Dublin (62.40%) - as opposed to the rest of the state and especially the Ulster region, with support levels for the Yes side standing at 57.21% in Leinster, 58.87% in Munster and 53.87% in Connacht-Ulster (and 47.81% in Ulster) – but Dublin North-West (49.61%) stood amongst the small number of constituencies opting to vote against the proposed change to the constitution. The only other two constituencies to vote against the proposed change to the constitution were the two Donegal constituencies, with the highest No votes (in percentage terms) in the state being recorded in Donegal North-East (with 59.66% voting No) followed by Donegal South-West (56.47% voting No). This is in keeping with previous referendum votes, in which the highest levels of opposition to referendum proposals were recorded in the Donegal constituencies as well as the more working class Dublin constituencies. As such, even though Sinn Fein supported the proposed change to the constitution, there is an evident pattern here (as with the earlier EU Fiscal Stability Treaty referendum) in which the constituencies with the highest level of opposition to the referendum tends to be those also in which Sinn Fein currently polls the strongest in.

Referendum elections have taken place on twenty seven different occasions in the history of the Irish state, with a significant increase in the incidence of these in more recent decades with 55.6% of these (15) having taken place over the past two decades. Referendum elections took place on nine different occasions (including cases where a number of referendum votes took place on the same day) between the founding of the State and 1984, but have also taken place on nine different occasions in the 2000s.

This post will look at voter turnout levels in these referendum elections, with a view to the holding of the Children’s Referendum on 10th November and the low turnout level recorded for that electoral contest.

As the graph below shows (Figure 1), voter turnout levels have varied quite significantly in Irish referendum election contests, ranging from a high level of 75.8% for the very first such referendum on the draft constitution in 1937 to a low of 28.6% for the 1979 referenda on adoption rights and university representation in the Seanad.  The mean turnout level across these different referendum polling days is 53.02% (falling to 52.11% following the holding of the Children’s referendum on 10th November). It is worth noting that, by default, turnouts in a referendum will always be some percentage points lower than those for general and especially local election as a not insignificant number of people (all registered voters with exception of Irish citizens) on the register do not have the right to vote in referendum contests (although in more recent times these numbers have tended to be excluded from the valid poll figures for referendum contests).

Turnout levels vary depending on the time period involved, with higher turnouts generally being recorded for the earlier contests held and with a notable difference in turnouts between the contests held in the low turnout era of the late 1990s and early 2000s against those held in the mid-to-late 2000s when turnout levels, on average, showed a marked improvement on those for the earlier years of the new century. Turnout propensity in referendum elections may also be shaped based on whether other types of elections are being held on the same day as the referendum contest – the most notable example here being the high turnout for the second set of referendum elections held in 1992, which were held on the same day as a general election.  The fact that the 1999 and 2004 referendum contests were held on the same day as local and European elections probably was also a factor in terms of increasing turnout propensity for those contests, especially in the more rural areas where turnout levels for local elections will generally tend to be highest.

The level of turnout can also be seen to reflect the issues involved in the contest/these contests, with high profile issues often resulting in higher turnout, as evident in the high turnout recorded for the 1971 referendum on joining the EEC. With the exception of the contests in the late 1990s and early 2000s, referendum elections on European Union issues or moral issues (divorce, right to life/choice) are usually higher than those for other contests, especially those in which the issues appear to be relatively non-contentious, as with the adoption rights/university representation in the Seanad contests in 1979 and the 1996 bail referendum, which are associated with the two lowest turnouts for referendum contests to date in the history of the state.

Figure 1: Voter turnout levels (%) in Irish referendum contests, 1937-2012

In keeping with the second order election model, which argues that turnout levels in second order election contests will tend to be lower than those in first order elections (general elections, in the Irish context), referendum turnouts usually tend to be somewhat lower than those recorded in general elections. The relatively high turnouts in the 1950s, 1960s and early 1970s pale in comparison with the average turnout levels in general election contests during that era, which would consistently have been in the high 70s. The difference in turnout levels between general election and referendum contests tends to be sharpest in rural and urban working class areas, but not as significant in urban middle class areas.

The day of the week an election is held on is believed to impact on turnout levels, with the general consensus being that weekend voting is the best in terms of facilitating a higher turnout level. Ironically the Children’s referendum will only be the second time a referendum election has been held on a Saturday. The only other occasion has been the second referendum on the Nice Treaty, held in October 2002, in which the holding of the election on a Saturday is believed to have played a role in pushing up the turnout level by c.15% relative to the first Nice Treaty referendum, which had been held on a Thursday, almost a year and half beforehand. Thursday has proven to be the most popular day on which to hold a referendum election (or a series of referendum elections), with Thursday being used as a polling day on 13 occasions to day (exactly 50.0% of all cases, if tomorrow’s contest is also factored in). Wednesday and Friday are the next most popular days to hold polling on, with these days each being used on five different occasions (19% of all cases) to date. Wednesday, however, proved to be a more popular day in the case of the earlier referendum contests while Friday was used as a polling day for a referendum contest for only the first time in 1995 (Divorce referendum). The contests held on a Friday (1995, 1998, 1999, 2004, 2009) and Saturday (second referendum vote in 2002) tend to compare favourably in terms of how turnout levels in these compare with those for otheer referendums held in the same period (1990s and 2000s) and it could be argued that such days are chosen in cases where government believes a high turnout to be necessary for the passing of a referendum. There has never been a referendum election on a Monday or a Sunday to date, while polling has taken place on a Tuesday on just one occasion (1987) to date.

Holding a number of referendum votes on the same day does not mean that turnouts will be higher; indeed some of the lowest turnout referendum days were ones that involved a number of contests being held on the same day (1979, 2001).

Figure 2: Turnout levels by Dail consituency for the 2012 Fiscal Stability Treaty referendum vote

One final point to note is that geographical studies of turnout patterns point to fairly consistent trends running across different referendum contests. As opposed to the trend for local and general elections in which the highest turnout levels tend to be recorded in the more rural areas, the lowest turnouts in referendum elections tend to be found in the more western constituencies (with turnout levels usually lowest in Donegal) and average turnouts for urban areas can be as high, or higher, than the average for rural areas (as shown by Figure 2). Referendum turnout levels tend to be highest in the more middle class urban constituencies, with the highest levels usually being recorded in the constituencies of Dun Laoghaire, Dublin North-Central (which will form part of the new Dublin Bay North constituency following the next general election) and Dublin South (whose boundaries are being significantly amended and which is being renamed as Dublin Rathdown).  Looking at a sub-constituency level, it can be seen that class trends for referendum contests tend however to reflect those for other electoral contests, although class differentials in turnout rates tend to be heightened for referendum contests, relative to those for general and especially local election contests. A more detailed discussion of referendum turnouts (in this case relating to EU treaties) in the Dublin City area has been covered in an earlier post. Indeed this and all other voter turnout related posts from this website can be accessed using the following link.

So what does this mean for tomorrow’s contest? Holding the contest on a Saturday will have a positive impact on turnout propensity. On the other hand, the issue involved in tomorrow’s vote appears to be similar in tone to those of the 1979 and 1996 contests, which may infer a low turnout propensity given that these contests recorded record low turnout levels. It can be more clearly suggested, however, that turnouts will probably be again highest in urban middle class constituencies (and will probably be highest in Dun Laoghaire, Dublin South or Dublin North-Central, although some rural constituencies, such as Tipperary North, will also figure amongst the higher turnout constituencies). Turnouts will again be lowest in the more working class urban constituencies and western rural constituencies, with the lowest levels again likely to be recorded in the Donegal North-East and Donegal South-West constituencies. It will also be interesting to see what impact holding the contest on a Saturday has in terms of resulting in especially low turnout levels in flatland areas especially if a number of registered voters in these areas have “gone home” for the weekend.

Note: The official government publication, containing detailed (by constituency or county) referendum results for all contests held up to the present date, is well worth checking out.