BY PETER WICKHAM THE RESULT of the recent St Lucia general election has left many of us in the region stunned. Now that the dust has settled, there is an urgent need for analysis of this political outcome from several different perspectives. In this instance, the analysis will seek to illuminate some of the electoral issues and examine whatever lessons can be learnt from the result. It has gradually become a practical policy for CADRES to review the outcome of events, especially elections that we would have surveyed. However, a few important facts should be noted. The most significant of these is clearly highlighted in the chart (at right) which presents a comparative analysis of the CADRES' projections of electoral swings in the 2006 St Lucia and the 2003 Barbados polls, and it can clearly be seen that in both instances CADRES has fallen within the +/- 5 per cent margin of error that it set itself on the outset. The polls in both instances, therefore, can be said to accurately reflect public opinion, and the suggestion that either of them is flawed can be rejected. Electoral swing The CADRES poll in St Lucia was based on interviews conducted in September, and there was substantial movement in the loyalties of the electorate between then and the date of the election in December. To this extent, the CADRES poll, which made predictions based on an accurate electoral swing almost three months ahead of the St Lucia election, will surely get far more credit than the Johnson Poll which predicted a landslide based on data collected on the eve of the same election. That pollster has sought to defend himself elsewhere. However, regardless of whatever "spin" is presented, the record will demonstrate that these were two very different predictions and should be assessed on their obvious merits, or demerits. The fact that CADRES was able to correctly predict the swing in St Lucia and yet miss the outcome speaks volumes about the extent to which the two countries are fundamentally different. This is a point made by Clyde Mascoll recently, and although his expression of the differences was not the most elegant, he had a point. Based on a projection of a swing vote made on the eve of the election in Barbados, which was a mere one per cent off, CADRES was able to predict the volume of support the Barbados Labour Party received, and this prediction was out by only one per cent. In the case of St Lucia, the accurate projection of a swing vote suggested that the victorious United Workers' Party (UWP) would have received 48 per cent, while it actually received 51 per cent in the election. This demonstrated that a one-per cent swing in St Lucia accounted for three per cent of popular support which was exactly the margin of victory for the UWP. This outcome is not strange to those who appreciate the impact of size on political analysis. Almost 80 000 people voted in the St Lucia election to elect 17 MPs, while about 140 000 voted in the last election in Barbados to elect 30. Simply put, the size of the St Lucia electorate is such that the one-per cent margin of the Barbados poll had no impact on the overall prediction, while that same margin in St Lucia changed the overall course of the election. There is therefore an important lesson in the last St Lucia election for Caribbean pollsters, although this outcome has not scarred the reputation of CADRES in any way since our analysis projected a swing of within +/-5 per cent of the outcome. In each and every instance, including in the case of St Lucia, we have never exceeded this margin. Arguably, the solution would be for CADRES to lower its margin of error. However, the science of electoral prediction does not at this time allow for such precision. But CADRES does have a developmental policy which includes motivating its associates on the ground to chart trends in an election; and not rely on a single poll which clearly has a relatively short shelf life in an election situation. Surveys apart, the recent St Lucia election presents several lessons for Caribbean politics. It demonstrated how fluid political opinion is in the region: the UWP was ahead in 2005, then the SLP appeared to regain its lead, and thereafter to lose it. The outcome was not a landslide as some sections of the Press reported; instead it was one of the closest elections in Caribbean history. A mere three percentage points separated the winner from the loser and the fragility of the Compton administration is not reflected in the majority that it commands in Parliament. The average margin separating the winner and loser in each St Lucian constituency was 135 votes, while the margin in Barbados was four times that. The "closeness" of these margins suggests that the UWP's strategy methodically identified each constituency individually as a battleground and worked to secure every single vote. This was clearly not a national strategy since reports suggest that the SLP's national campaign was superior in every way and their meetings better attended. The UWP appears to have triumphed on the day of the election with superior mobilisation and organisation on the ground. Clearly the "fetes" that our political meetings have become are now less important than mobilisation on the day of the election to determining the eventual outcome. The SLP was also not helped by a lethargic support base which appears to have been influenced by the Johnson Poll released on the eve of the election, which predicted the SLP would win by a landslide. Ironically, the SLP commissioned and released this information, and, in so doing, "shot themselves in the foot". It is reasonable to assume that the release of this erroneous information had the same effect on voters, but since the SLP was only barely winning, and the poll was not a true reflection of the state of play, lethargy among its supporters was fatal. Implications for Clyde Mascoll? The other major issue that the St Lucia election will be remembered for is the fate of former Prime Minister Dr Vaughn Lewis, who crossed the floor and sought to capture the seat for the party he once opposed in Parliament. In this instance he lost. Many have suggested that Clyde Mascoll will suffer the same fate in Barbados. Although I have already expressed a similar opinion of the outcome in St Michael North West, it is not a simple comparison and warrants further explanation. The analysis advanced by CADRES in relation to that seat relies heavily on the swing analysis which we believe will determine the outcome in that and virtually every seat in both Barbados and St Lucia. Hence, we believe Mascoll will lose if the swing is against him as it presently appears. In the case of St Lucia, the data does not suggest that the personality of Lewis, or his political pedigree, had any impact on the result in that constituency. We believe that the result in St Michael North West will also be determined in a similar fashion. It is noteworthy that the national UWP swing was 14 per cent and that the UWP swing in Castries Central was exactly 14 per cent, suggesting Richard Frederick did as well as his party, and we believe that Mascoll will similarly perform as well (or as badly) as his new party does. Previously, I expressed the opinion that Sir John made an error in returning to the helm of the UWP and it could now be argued that this opinion has been proven wrong by the outcome of the election; however, this also depends on perspective. An election is a single event in the political life of a country, and it can therefore be argued that Sir John has won the battle but his UWP will ultimately lose the war. Even if he were younger, he cannot run the affairs of St Lucia forever and he has now proven implicitly that the UWP is his personal party which cannot survive without him. Sir John's success will bolster other political dinosaurs who are seeking to return to office, when they should be planning their succession and building their parties as institutions. Already Basdeo Panday has started his return. In the final analysis, both of them will have to leave politics eventually and when they do, their organisations will become easy targets for many years as they seek to locate the next Sir John Compton or Panday. Perhaps the mess that has become of the Democratic Labour Party in Barbados ought to be a lesson to all of them regarding the importance of succession planning and institution building. Perhaps then SLP leader Dr Kenny Anthony should "sit tight" since he might be called to battle once more in the not-too-distant future, and it will be so much easier next time around. * Peter W. Wickham (Wickham@sunbeach.net) is a political consultant and a director of the Caribbean Development Research Services (CADRES).
|
