Election 2020 has resulted in seismic shifts in the Irish political landscape. The outgoing minority government has seen significant losses, and Fianna Fáil and Fine Gael have achieved their lowest combined share of votes and seats ever. Sinn Fein has surged and is now the most popular party in the state, while the Greens have scored their best result ever, and the Social Democrats have advanced. At the same time, Independents, so long a ubiquitous feature of the Irish system, remain a potent force in the 33rd Dáil. While the focus of most moves to figuring out how a new government can be put together, academics and political anoraks are parsing the results to understand the underlying dynamics of this momentous election. The objective of this contribution is to decipher how pivotal lower preferences (transfers) were in shaping the 2020 election result.
How to measure the decisiveness of transfers? In a 2017 contribution with my colleague Hannah Schwarz, we conceptualize when lower preferences have a pivotal impact at the constituency level. In short, their decisiveness depends on candidates who do not occupy a winning position (defined as occupying one of the n-slots on count 1 in a district with n-seats) winning election after the distribution of lower preferences. In a forthcoming paper at the International Political Science Review, we provide the most comprehensive investigation of the decisiveness of lower preference votes on election outcomes by exploring 62 elections in Ireland, Northern Ireland, Malta, and Scotland, all polities that use STV to elect a significant legislative body by popular vote. We found that on average transfers are pivotal in the election of about 1 in 10 candidates – so yes, they clearly matter, but perhaps their critical importance is not what some might have assumed. In Ireland, there was remarkable uniformity of the pattern across time, and we discovered little difference between the decisiveness of lower preferences in general elections compared with local elections. More innovatively, our paper also investigated in what circumstances transfers are decisive at the district level. In the Irish case, we found that Fianna Fáil benefitted the least from lower preferences. Instead, smaller parties like the Greens and non-party candidates were much more likely to owe their victory to lower preferences.
And so to the impact of transfers in 2020. At face, we might expect the decisiveness of transfers to be more significant given the seismic changes observed in this election, especially as Sinn Féin had a plethora of substantial surplus votes to transfer and, in most instances, no party running mate. An analysis of the 39 constituencies reveals that lower preferences decisively elected 21 candidates in the 2020 general election – or 13.1 per cent, the highest level since 1933. It is higher than the 1 in 10 candidates that have previously owed their election to lower preferences, giving some evidence that in 2020 transfers were more critical than in previous contests. Nonetheless, while they were more important, their effects should not be overstated– about 5-6 more candidates in this election owe their victory to transfers than would have been expected.
Table 1 drills down into the data and details the constituencies where transfers had a decisive impact on the outcome, the candidates who lost out, and who won due to lower preferences, their party affiliations, and their position on Count 1. In a majority of cases (57 per cent), candidates who owe their election to transfers were just outside the winning slot on count 1 – for example, David Stanton of Fine Gael in Cork East came fifth on Count 1 in this four-seater but ended up elected after the distribution of lower preferences. Some bucked this trend, most notably Cormac Devlin of Fianna Fail in Dun Laoghaire, who was seventh on Count 1, three slots from the winning position, and who ultimately achieved election despite only winning 0.46 of a quota. Marc O’Cathasaigh of the Greens in Waterford (sixth on Count 1 with 0.37 of a quota) and Joan Collins (Ind) in Dublin South Central (also sixth on Count 1 and scoring only 0.32 of a quota) were also notable in this respect. Speaking of Dublin South Central – it was the only district where lower preferences had a determining influence on more than one seat. Catherine Byrne (FG) and Catherine Ardagh (FF) both lost out (to Patrick Costello (Greens) and Joan Collins (Ind)) despite occupying a winning position on Count 1.
There is some evidence that Sinn Fėin surplus votes helped put some candidates back into contention, who eventually emerged winners. For example, Mick Barry (PBP) in Cork North Central, Joan Collins (Ind) in Dublin South Central, and Thomas Pringle (Ind) in Donegal all came into contention because of significant vote transfers from Sinn Fėin.
The above analysis also reveals that one party above all lost out because of lower preferences – Fianna Fáil. Of the 21 instances where transfers were pivotal at the district level, Fianna Fáil candidates lost out on twelve occasions (and gained two: Anne Rabbitte in Galway East and Cormac Devlin in Dun Laoghaire), yielding a net loss of 10 seats due to lower preferences. It concurs with our historical analysis, which showed that Fianna Fáil was the party most likely to suffer when transfers became pivotal.
Until now, we have focused on lower preferences’ decisiveness at the district level. Of course, they play a role in determining the final number of seats in parliament for each party and thus can also be said to impact government formation. Historically, they have had an influence. For example, in 1987, lower preferences denied Charles Haughey a majority government costing Fianna Fail eight seats. Moreover, despite the prevailing narrative that Bertie Ahern was a transfer talisman for Fianna Fail, lower preferences also cost Mr. Ahern a majority government in 2002, with Fianna Fail winning eight fewer seats after lower preferences were counted. So what about 2020? Figure 1 shows the actual result (inner speedometer) and contrasts it with what the result would have been had transfers not been distributed (outer speedometer). As we can see, Fianna Fáil is the big loser – they would have won 48 seats had lower preferences not been distributed, compared with the 38 they ended up with. Sinn Fein would also have won a couple of extra seats and come home with 39 instead of 37. Meanwhile, the Greens, Independents, and the Social Democrat representation have been enhanced due to lower preferences while Fine Gael held even (gaining six but losing six).
Three points emerge from this analysis. The first is that the results Quinlan and Schwarz observed in Irish elections largely hold true even in 2020 – transfers matter but their pivotal impact remains the exception and not the norm, and when they do matter, they are to the disadvantage of Fianna Fáil. Nonetheless, lower preferences were somewhat more pivotal in 2020, with 21 candidates owing their election to transfers, when we might have expected between 15-16 candidates to win on this basis. Second, lower preferences scuppered the chance of a majority Fianna Fáil and Fine Gael coalition government emerging. Without the distribution of transfers, Fianna Fáil and Fine Gael would have commanded 83 seats between them. With transfers, both parties combined have only 73 seats. Third, lower preferences ensured that leftist parties (Sinn Fėin, PBP, Labor, and the Social Democrats, admittedly all of a different leftist ilk) achieved a net five seats through lower preferences. Consequently, perhaps the most significant influence of lower preferences this time out is their impact on government formation options. Like 1987 and 2002, Fianna Fáil has fared the worst with transfers, and this has weakened their bargaining power significantly. We can say that lower preferences have strengthened the left, as broad a church as it is, and ensured that any link-up between Fianna Fáil and Fine Gael is more challenging to realize.
 It is not assumed that the outcome would have been the same under a system where only the first preferences are taken into account. Rather, the focus here is on what happens to transfers.
Dr. Stephen Quinlan is Senior Researcher at the GESIS – Leibniz Institute for Social Sciences in Mannheim. He is the Project Manager of the Comparative Study of Electoral Systems (CSES, http://www.cses.org) project, which explores electoral behavior in over forty states worldwide. His research has been published in journals such as Electoral Studies, Journal of Elections, Public Opinion and Parties, Economic and Social Review, Information Communication & Society, Irish Political Studies, and European Political Science Review. The author thanks Conor McGarry and Deirdre Tinney for constructive suggestions.