ANALYSIS:Close examination of the first Irish Timesopinion poll of the election campaign does not make for happy reading for Fianna Fáil
The IPSAS-MRBI Irish Timespoll last week putting Fianna Fáil at 17 per cent, Fine Gael at 33 per cent and Labour at 24 per cent confirmed the picture of electoral preferences painted by many other polls over the last few months. But Irish elections are not fought in one national constituency (although they are in the Netherlands, and in Israel) but in 43 constituencies, and even if we take these results as a reliable national estimate, we can speculate about what this might mean for the election on the ground.
Many people will argue that it impossible to say anything worthwhile about constituency outcomes because of the presence of “local factors”. However, we must assume survey respondents have taken such factors into account in giving us their vote intention. Moreover, unless these local factors cancel out one another the national poll estimates would typically be much more inaccurate guides to vote share outcomes than they in fact prove to be.
Local factors do account for considerable variations in the swings between parties. If we look at 2002-2007 for instance, the Fianna Fáil vote remained almost unchanged at national level at 41 per cent. At constituency level though, we saw the Fianna Fáil increase in many places and decline in others, with a change of almost 4 per cent on average. FG increased its vote by 5 per cent, and there was considerable variation around that national swing, with a few even showing no change, or change of 10 per cent or more.
When we consider that the sort of swings implied by the IPSAS -MRBI Irish Timespoll are more like 25 per cent (in the case of Fianna Fáil), the possibility of pretty large exceptions is a very real one. One way to mitigate this somewhat is to use the regional estimates from the poll rather than the national ones. These are less reliable than the national poll figures because they are based on a subsample of only 250 respondents or so rather than 1,000 in the national polls. However, they do provide an opportunity to get at possible regional variations.
The swings implied at regional level in this Irish Timespoll can be compared with those in other recent polls and in general they coincide very closely, so we can have some confidence in using them as a basis. There are some marked variations between the regional swings suggested. Fine Gael, for instance, appears to be gaining more votes in Dublin and the rest of Leinster than in Munster and Connacht-Ulster, while Labour's gains are clearest in Dublin. SF's gains are greatest in Munster, as are those of Others.
The most striking aspect of projecting these changes on to the 2007 results is that Fianna Fáil is expected to average only 0.7 quotas per constituency. Labour is expected to win 1.3 and FG 1.6. A vote amounting to 0.7 quotas will translate into a seat more often than not, so we might guess that Fianna Fáil could win in excess of 30 seats with its 17 per cent of the vote, but this is almost certainly an optimistic figure if the poll estimate translates into votes on February 25th. A single candidate, helped by transfers, will often make it, but where a party spreads its vote over two candidates, 0.7 of a quota is an unlikely basis for a seat. This is because a party’s voters cannot be relied on to transfer support to a running mate. Moreover, in 11 constituencies Fianna Fáil manages no more than half a quota or less: Dublin South-East , Kerry North ,Tipperary North , Cork North-Central, Dublin Mid-West, Dublin South-Central, Kildare North, Wicklow, Kerry South, Cork East and Dún Laoghaire. Given that Fianna Fáil is running two candidates in Kildare North, Wicklow, Cork East and Dún Laoghaire, the chances of success in these are very poor on the basis of these figures.
Even where Fianna Fáil's expected quota is higher, that vote will be divided more often than not. In recent elections Fianna Fáil has often been applauded for its vote management. Under Bertie Ahern the party won more seats per vote than it had done in the past. This was seen in part to be because the party became more transfer friendly, but it was also a tribute to the way voters were organised on the ground to try to maximise the seat share. There seems little likelihood of such management this time, not least because it could prove counterproductive. In many cases the party may hope one of its candidates outpolls the other by a big margin. This could then keep them in the race long enough to get transfers from their running mates and other eliminated candidates. Unfortunately for Fianna Fáil, rather than being transfer friendly this time, it appears the party may prove transfer toxic. The Irish Timespolls suggested very few voters indicated they would give a second preference to Fianna Fáil. In marked contrast to 2002 and 2007, only 10 per cent of independent voters, 9 per cent of FG voters and 7per cent of Labour voters indicate subsidiary support for Fianna Fáil. While we may be sceptical at this point about the exact numbers, it is clear the party is relatively unpopular and will not do well in the transfer market.
The over-nomination is also likely to damage the party even where it does win a quota or more, as it is expected to do in places such as Kildare S, Wexford, and Galway E. Cork SC, Donegal SW and Carlow Kilkenny (where it is running three candidates). Two candidates in Limerick city is of less concern since nobody expects Willie O’Dea to share votes with his running mate, while despite her place on the party’s new front bench, Mary Fitzpatrick may be perceived to be Independent Fianna Fáil in Dublin Central.
If Fianna Fáil has been slow to adjust downwards, Labour may have adjusted upwards a little too quickly. The intention of giving most candidates running mates to capitalise on the meteorological forecast of a Gilmore Gale perhaps gave undue weight to apparent leader popularity and a few opinion polls. The expected Labour vote of 1.3 quotas per constituency hardly justifies two candidates in many, or indeed in most seats, but Labour is running multiple candidates more often than not with two candidates in 21 constituencies and three in two more.
My figures suggest an expected quota of 0.9 where there is one candidate and 1.5 where there are two. In Galway East and Louth the party is running two candidates with less than one quota expected, and this may cause difficulties for both candidates in each case. Elsewhere the party looks a little safer with 1.3 quotas or more. However, Red C consistently place Labour a little lower than does IPSOS-MRBI and a small slippage in the Labour vote could cause difficulties for it in a number of other seats such as Dublin North, Dublin North-East and Dublin Mid-West unless the leading Labour candidate gets the lion’s share of the vote. In only eight constituencies – most of them in Dublin – is Labour in a position to hope for a couple of seats, but in only one of them, Dublin SC, is it expected in this analysis to exceed two quotas. In declining order of possibility, Cork E, Dublin SW, Wicklow, Dublin Fianna Fáil, Dublin S, Dún Laoghaire and Dublin SE could see two Labour TDs, but good vote management may be necessary to realise many of these.
What will help is that Labour is likely to pick up more transfers than any other party. The IPSOS MRBI Irish Timespolls indicated that 20 per cent would give Labour a lower preference, as opposed to only 12 per cent for FG and 7 per cent for Fianna Fáil. Again, it is the relativities that are important. Fine Gael and Labour voters are most likely to transfer to each other so that in the battle for the last seat in many constituencies it will be important for each to stay ahead of the other so as to benefit from that potential boost.
Fine Gael has fewer such problems. As perhaps befits a party with an outside chance of an overall majority, it is running two candidates almost everywhere. However, there are five constituencies where Fine Gael gets a quota or less, and in three of those – Dublin NW, Tipperary S and Dublin Mid-West – it is running two candidates, as it is in three of the five constituencies where it could expect only 1.1 quota on these figures, Donegal NE, Dublin N and Kerry S. Over-nomination seemed to work well for the party in Clare in 2007, when four candidates still produced two seats. Fine Gael will hope for the same result this time, but will hope for a few more votes than suggested by this poll if it is to capitalise properly on its candidate strategy. While the broad parameters of the election results may be set, the distribution of seats will depend on who gets the benefit of the margin of error, as well as the small bonus, or loss from the campaign.
Lots done, much to do – for all of them.
Michael Marsh is pro vice-provost and professor of comparative political behaviour at TCD. He is a co-author of
The Irish Voter
(Manchester University Press 2008)