On December 20, 1998, THE NATION carried an article entitled Safety Zones In Barbados which reflected my initial thoughts on the challenge of identifying “safe seats” in Barbados in anticipation of the 1999 election.
At the time, I was commencing my career in political consulting and was therefore considerably less experienced and more timid.
As such, that article stayed well within the proverbial “crease” and essentially reviewed the applicable data post-Independence. It concluded that a safe seat was one in which the incumbent maintained a relatively high level of support and one which had never changed “hands” since being created.
Based on this objective criterion, that article implicitly concluded that Christ Church West and St John were the only two seats on either side of the political divide which could be considered “safe”.
It is always useful to review one’s analysis with the benefit of hindsight and the experience gained from more than a decade of work in the field of political consultancy across the region.
This experience has been more practical than the academic instruction I had received up to that point (1998). Such a review would immediately reveal the strengths and weaknesses of that approach which will ultimately “sharpen” the predictive tool explored in the first edition of this article (1998).
In the wake of its publication, Dr George Belle commented privately (and I agreed) that the analysis appeared somewhat “conservative”, since there were clearly more than two safe seats in Barbados at that time.
As fate would have it, this approach was largely vindicated since the Democratic Labour Party (DLP) was only able to save two seats and one of these was highlighted in the analysis, while the other (St Lucy) behaved in a way that was inconsistent with the national swing.
Certainly, if the constituency swing in St Lucy was at or above -3 per cent, which was the DLP’s national average in 1999, that too would have been lost, and the fact it was saved would suggest that the Member of Parliament there performed “well”, which is to his credit.
The 1999 election was a useful laboratory within which an analysis of this nature could be tested since it was easily the most peculiar of elections in Barbados. For the first time in our history, there was a swing against the DLP, although that party was in Opposition. Ironically, this swing was a mere -3 per cent against the Dems.
However, as is the case with this 2013 election, the slightest swing will cause severe electoral devastation. Notwithstanding the challenge that 1999 election presented, the analysis established which DLP seats were likely to be saved and to this extent the conservative statistical approach properly told half of that story.
Upon reflection, however, a more rigorous analysis could easily have identified other relevant factors that could possibly have “flagged” St Lucy as a safe seat as well.
There were two major (and related) weaknesses of that approach to identifying safe seats. The first was entirely historic and as such paid no attention to contemporary factors that would for example have allowed me to identify St Peter and St Michael North-East as safe seats as well.
In both instances these seats were lost in the past and by the same candidates that defend them now; but there is little possibility that they would be lost again in the near future.
The other major weakness was that the analysis presumed constituencies would behave more independently than is the case. Since that time, my analysis has moved steadily in the direction of highlighting the extent to which common party-political threads link the outcome in several constituencies.
As such, the prevailing party-political environment would have made several Barbados Labour Party seats safe that would otherwise have been unsafe.
It would now appear there are at least six factors that impact on the stability of a seat and, if properly analysed, could help with the determination of whether or not a seat is safe. Some of these factors are quantitative, while others qualitative and, consequently, more difficult to measure.
In next week’s article, CADRES will present this list of ideas and later, on one occasion prior to the 2013 election, an assessment of these factors relative to each seat as part of its predictive activities.
• Peter W. Wickham is a political consultant and a director of Caribbean Development Research Services (CADRES).
