With Election 2016 just months away, the Labour Party will, without doubt, provide the media with the disaster story of the election cycle - much like the hapless Lib Dems across the water. Like the Lib Dems, the real question is not whether they will suffer a meltdown, but how great that meltdown will be. What Labour also has in common with the Lib Dems is that the commentariat is grossly understating the scale of the collapse that awaits it on current polling trends. Remember that with the Lib Dems averaging 7-9% in every poll, losing between a third and nine tenths of their vote in by-elections and being reduced to one seat in the European Parliament, the psephological consensus was that they'd get between twenty and thirty five seats - they won eight. Today, the Irish media is making the assumption that Labour's "shellacking" in the upcoming election will not be sufficient to stop it from remaining a meaningful player in the game. I think that this reflects the particular groupthink that afflicts political commentators, which reflects a similar bias in favour of Labour that the British media establishment showed towards the Lib Dems. I can suggest three reasons for this bias:
- Political commentators dislike the departure of political certainties that incumbent political parties give them. We are all familiar with the media mantra: "If it bleeds, it leads." However, few understand the principle that what bleeds too much misleads. In other words, when a major political player departs the scene or its presence becomes a virtual irrelevancy, it makes the political tea leaves harder to read for perhaps years to come, increases the probability of people making predictions that are wildly wrong and undermines the ability to use previous political cycles as a guide to future ones. Worse still, in a world in which historical education has declined alarmingly, radical change may force commentators to study longer range historical trends, which takes both journalists and their consumer base of readers, listeners and viewers out of their comfort zones. I therefore submit that the commentariat is, for this reason, emotionally unwilling to contemplate a world in which Labour - a party that Sean Duignan used to describe as "the X in OXO" - might simply cease to exist.
- The typical holding pattern of Labour support over the last two decades has been circa 10%. In the last 23 years, the party only broke out of this range to the upside twice (2009 and 2011) and to the downside once (2014). However, my own experience of people who work in media and academic politics/history is that Labour support amongst them is perhaps three to four times higher than amongst the population at large. My conversations with media people over the years (especially at RTE) revealed that in 2007, they were disgusted that Enda Kenny had won the battle with Pat Rabbitte to lead the opposition and in 2011, there was a widespread disgust at the notion that Eamon Gilmore did not win the Taoiseach's prize. In general, the disproportionately humanities and postgraduate educated media class is ashamed of Ireland's political system, which has consistently failed to instal a soft left party as one of the parties of government. With left-of-centre loyalties, I submit that the media finds it distressing to contemplate a world in which Labour is gone, political power is still dominated by FG and FF and the left is led by the toxic Sinn Fein and the unelectable far left.
- Making electoral predictions involves a high degree of comfort with numbers. In my own profession, there is a saying: "Law is the last refuge of the innumerate". But it's not just law. People from humanities backgrounds - be they (we?) lawyers, commentators or historians - are simply not comfortable with statistics. Small moves in the vote generate outcomes that are intuitive and easy to predict. Big changes do not. When a party goes from 19 to 15%, it can be predicted that it will lose between a quarter and a third of its seats. However, when a party falls from 19% to 6%, the meltdown is much more severe. Nonetheless, the preferred recourse of the commentators is to assume that the vote will only fall to 8% and then take 50-60% off the 2011 seat tally. For reasons I'll explain, this is flawed.
The 2007 Baseline:
Because of the Black Swan events of the 2008 financial crisis and the meltdown of Fianna Fail, I think it is wise to treat Labour's 2011 performance as an anomaly. The 2007 election is a much more reliable indicator of how Labour will do. In that election, the party got just over 10% of the vote and twenty seats. There are two statistical patterns that help explain this result.
In 2007, I analysed the political map of Ireland to see how Labour could expect to perform in the future. The first key pattern which emerged was that there was a large geographic swathe of the country in which Labour had virtually no political presence. If one drew a line that started at around Howth Head on the North Dublin Coast, continued around the northern borders of Meath and the Southern Borders of Westmeath and Longford, looped around Galway, Sligo, Clare, West Limerick and all the way down to West Cork, one found a startling pattern. To the North/West of this imaginary line lay the following constituencies: Dublin North (4 seats), Louth (4 seats), Meath West (3 seats), Meath East (3 seats), Cavan Monaghan (5 seats); Roscommon South Leitrim (3 seats), Sligo North Leitrim (3 seats), Galway East (4 seats), Galway West (5 seats), Mayo (5 seats), Donegal North East ( 3 seats), Donegal South West (3 seats), Clare (4 seats), Limerick West (3 seats), Kerry North (3 seats), Kerry South (3 seats), Cork North West (3 seats) and Cork South West (3 seats). This takes in an aggregate of 64 seats. In this whole geographic area, Labour managed to win just one seat (Michael D. Higgins in Galway West). Stated differently, in 40% of the country, Labour won a statistically negligible 1.5% of the seats.
The flipside of this is that the vast majority of Labour's vote concentrated in the other 60% of the country. To the South and East of the imaginary line, Labour won 19 seats out of 102. This concentration served Labour well. If its vote had distributed itself evenly across the country, it would have lost many seats in the south and the east and gained very few in compensation in the north and the west. This led me to a second pattern. While Labour's regional concentration was heavy, the spread of its vote within the parts of the country where it had a presence was remarkably even, which delivered a further seat bonus. In 2007, only two Labour TDs topped the poll (Pat Rabbitte in Dublin South West and Willie Penrose in Longford Westmeath), Labour's vote exceeded 20% in just two constituencies (Dublin North West and Kildare South) and the Labour Party's vote did not reach the quota or exceed it in even one constituency. In terms of ratio of votes to seats, Labour was incredibly efficient. Its best performance in the whole country was in Dublin South West, where it won a seat with seven votes short of a quota. Meanwhile in Dublin Mid West, Labour took a seat with less than 55% of a quota.
These patterns raised severe dangers for Labour, inasmuch as:
- it was reliant on external transfers to win the vast majority of its seats, meaning that it could lose seats on the same vote if there was any kind of change in transfer patterns;
- none of its TDs had any real buffer against vote drops in their constituencies, with every Labour TD coming into some kind of danger zone on vote drops of 1-2%; and
- Labour had become disturbingly reliant on incumbent TDs to hold its vote and had an aging parliamentary party which was likely to see a disproportionate number of retirements in the near future.
Subsequent events in the economy rendered my calculations temporarily redundant. However, with the 2011 Labour bubble having been burst, economic and political cycles seem to be, in fact, amplifying the effect of political trends that I had identified in 2007.
The 10% Hypothesis:
I have tried to simulate how Labour would have done on current boundaries with the 2007 vote - naturally, this is a rather finger-in-the-air process - and I have taken local difficulties into account and assumed that Roisin Shortall and Tommy Broughan, who shrewdly walked out on Labour before it went into meltdown, will walk away in each case with what would otherwise be the Labour seat.
On this basis, I have concluded that Labour would win seats in: Dublin West; Dublin South West, Dublin Mid West, Dun Laoghaire, Dublin Bay South, Dublin South Central, Dublin Central, Longford Westmeath, Kildare North, Kildare South, Wicklow, Wexford, Waterford, Cork North Central, Cork East, Limerick City and Galway West. This assumes that Tommy Broughan would take the Labour seat at the expense of Aodhan O'Riordan and Sean Kenny in Dublin Bay North, Roisin Shortall takes Dublin North West and that Ciaran Lynch in Cork South Central loses out due to the boundary change. This puts Labour on seventeen seats. However, with lower external transference, the real number could easily be lower (perhaps 14 or 15). On the assumption that Labour were to lose 20% of their 2007 support in each constituency (which would equate to roughly 8% nationally), I have estimated that they would almost certainly lose their seats in: Dun Laoghaire, Waterford, Wexford, Dublin Central, Dublin Mid West, Limerick City, Galway West and Wicklow. This would put Labour on nine seats - a catastrophic result, but one from which it might recover in time.
However, the party's current polling numbers indicate that 6-7% is not an improbable outcome. In a pure PR system such as Israel, Labour would lose seats pro rata their percentage vote. However, our system is not truly PR - one thing I dislike about political science is its tendency to talk about PR and First-Past-the-Post as if any system that isn't pure FPTP is, by definition, a proportional representation system. The multi-seat, STV system creates anomalies. In 2011, on 17% (2 points behind Labour), Fianna Fail won 20 seats to Labour's 37 ( the former being the same as Labour got in 2007 on a mere 10% of the vote). In 2002, the now defunct Progressive Democrats won 8 seats (or roughly 5%) on just 4% of the vote, while on the same day Fine Gael won 22.5% of the vote but only 18% of the seats. Far from being more proportionate than FPTP, PR-STV is, if anything, more statistically anomalous in terms of the numbers it throws up. For the last quarter century or so, Labour has been the most consistent beneficiary of the system's anomalies. It is for this very reason that I am prepared to wager that Labour has developed in such a way that if it doesn't get 10% of the vote, it very quickly implodes.
The worst case scenario for the Labour Party is literally losing all of its seats. I consider this outcome relatively improbable. However, as of the writing hereof, the party's reduction to five seats or fewer looks not merely possible but probable. What independent or viable future would Labour have as a party with three or four seats? This may well be a question that we are asking ourselves on count day, as the electorate's verdict on Labour comes in. One thing is for certain. There will be blood on the floor. The only question is how much.