A closer analysis of last weekend's election results suggests that first impressions were not that reliable, write Richard Sinnott and James McBride.
A certain conventional wisdom is beginning to develop around the recent local and European elections.
It has a number of components. Firstly, Fianna Fáil suffered unprecedented losses. Secondly, the Sinn Féin surge was of dramatic proportions. Thirdly, Sinn Féin's new-found supporters are assumed to consist mainly of Fianna Fáil defectors. Fourthly, these defectors are thought to be likely to stay with Sinn Féin, making Fianna Fáil's task in the next election extremely difficult and portending victory in that election for a Fine Gael-led coalition.
These propositions are not altogether consistent with the fifth and final component of the received wisdom, namely that the striking increase in turnout was due mainly to the mobilisation of previous non-voters by Sinn Féin.
The evidence from the elections offers more support for the last of these five propositions than for the previous four. The turnout story is indeed interesting. First of all the rise in turnout was especially noticeable in urban areas. As the graph above shows, whereas turnout in recent local elections in Dublin had lagged behind turnout in the rest of the country by about 15 percentage points, this gap was reduced to six points in 2004. Similarly substantial increases in turnout were was also manifest in other urban areas.
More importantly for the Sinn Féin mobilisation interpretation, there was a statistically significant correlation (0.33 with a significance level of 0.058 for the statistically minded) between the increase in turnout and the growth in support for Sinn Féin; statistical tests for other possible political correlates of the increase in turnout proved entirely fruitless.
Because it is based on data relating to aggregates rather than individuals, correlational analysis of this sort can never be fully conclusive; such correlations can, however, be highly suggestive and in this case the correlation in question supports the idea that Sinn Féin has the capacity to mobilise previous non-voters.
But what about the other elements of the growing conventional wisdom? Let's start with the unprecedented nature of Fianna Fáil's losses. As pointed out on this page two days before the election, Fianna Fáil suffered losses of broadly similar magnitude in five of the seven previous local elections and in four of the five previous European elections.
Thus, except in one respect, last week's outcome for Fianna Fáil was par for the course. The exceptional aspect is that the fall was from the relatively low base (about 39 per cent) that has characterised the Fianna Fáil vote since 1992. Thus Fianna Fáil's problem is not the size of its losses in 2004, but the fact that making good those losses may not be enough to see it back in Government Buildings after the next election.
And what about the Sinn Féin surge? In one sense, Sinn Féin went from three per cent to 11 per cent. But this is not comparing like with like. What in fact happened is that, as the graph above shows, Sinn Féin gained approximately 5 percentage points in each type of election - from 3 to 8 per cent in the local elections and from 6 to 11 per cent in the European elections.
And even this 5 percentage-point gain slightly exaggerates SF's march forward because it does not allow for the fact that the party was contesting a number of local electoral areas (LEAs) for the first time. Including these in the comparison tends to inflate the estimate of the increase in support for the party because one end of the comparison is, by definition, zero.
Comparing LEAs in which Sinn Féin had candidates in both 1999 and 2004 shows that, in directly comparable situations, the party's vote grew by 3.9 percentage points between 1999 and 2004. Impressive, yes; dramatic, no.
Nor is there much evidence that defections from Fianna Fáil to Sinn Féin were the factor that determined the outcome of the election. While, as pointed out above, aggregate-level correlations cannot provide definitive proof, the fact that the correlation between Fianna Fáil losses and Sinn Féin gains is not remotely statistically significant strongly suggests that this is not the story of the election.
Of course, if Fianna Fáil defections to Sinn Féin were not as extensive as has been assumed, the question of wooing them back is less pointed. However, from a Fianna Fáil perspective, the defections have to be won back from somewhere.
Two factors influence our assessment of how difficult this will be. The first is that Fianna Fáil has usually bounced back from defeats in elections of this kind before. The second lies in the nature of local and European elections. They tend to be bad guides to what will happen in the next election because, despite their portrayal in the media, they are not primarily referendums on the performance of the Government and they are substantially more local and substantially more European than is often assumed.
Accordingly, with local issues in abeyance, with European issues and personalities safely consigned to limbo in Brussels/Strasburg and with TDs returning to centre stage as they struggle to deal with the consequences of the abolition of the dual mandate, all parties (winners and losers) can expect at least a partial reversion to the status quo ante in the distribution of first-preference votes in the run up to the next election.
In addition to those first-preference votes, the transfer of subsequent preferences will be crucial in determining the outcome of the election.
Evidence from previous elections indicates that voters are highly responsive to the signals that party leaders give them regarding the preferred destination of transferred votes. Coming to an agreement on such signals is one of the biggest problems facing the putative coalition of Fine Gael, Labour and the Greens.
Evidence on transfer patterns in the 2004 local elections will take some time to sift. In the meantime, while European elections include many factors that are likely to produce unusual transfer patterns, a few pointers can be gleaned from how the voters distributed their later preferences last Friday week.
On the whole, the pointers are mixed. Thus Fianna Fáil, Fine Gael and Labour all showed some good instances of party solidarity (of around 60 per cent) in the east of the country (Dublin and the East constituency) but Fine Gael and Fianna Fáil showed decidedly lower levels of internal party solidarity (40 to 45 per cent) when faced with substantial independent candidates in the South and North West constituencies.
The evidence of coalition overtures at the level of the voters is difficult to assess because the number of relevant transfer situations is very small. Nonetheless it is worth noting, on the positive side, the 62 per cent FG to Labour/Greens transfer in Dublin and, on the negative side, the less than enthusiastic coalitional response of the Greens in the East constituency.
Finally, what parties receive in transfers is as important as what they give and, in this respect, the limited ability of Sinn Féin to garner transfers from other parties may hamper its efforts to translate growing first-preference support into seats in Dáil Éireann.
Again the European elections only provide limited evidence. With this reservation in mind, it is notable that Sinn Féin won only 11.6 per cent of transfers from the Greens and 11.9 from Independents. Less surprising is the fact that it gained only 4.6 per cent from the terminal Fine Gael transfer in Dublin.
The evidence of previous local and European elections indicates that their immediate effect on the party politics and on politics in the Dáil is greater than their longer-term effects on the balance of electoral support.
It is not that the considerable changes witnessed last weekend are without implications. It is rather that the implications are complex and uncertain. Ultimately, and this is the good thing about electoral politics, it all depends on how political elites respond.
Richard Sinnott is director of the Public Opinion and Political Behaviour Research Programme at ISSC (Institute for the Study of Social Change), UCD. James McBride is director of the Irish Social Science Data Archive, also at ISSC