Posted on

PETER WICKHAM: Historic St Vincent poll

Peter Wickham, [email protected]

PETER WICKHAM: Historic St Vincent poll

Social Share

The dust has just about settled on the 2015 St Vincent and the Grenadines election and those of us who monitored it can say we are now fortunate to have been witness to Caribbean electoral history.

Initially, however it is important to review the CADRES pre-election poll in light of the results which have just been presented.  Two critical polls were conducted on the mainland island of St Vincent, with one being taken before the election was called and one in the midst of the election campaign, and both effectively projecting an electoral swing of five per cent in favour of the ULP.  In the final analysis the pro-ULP swing (on the mainland) was contained to 1.1 per cent, which is within the poll’s margin of error.

The ability of the NDP to contain the ULP swing is a matter that should generate considerable interest especially as the critical factor of leadership was decisively in the favour of the ULP.  In all CADRES polls, Dr Gonsalves has maintained a significant lead over Mr Eustace, suggesting that ULP “improvement” between the last poll and election day was not related to any “push” factors like Vincentians becoming more enamoured with Eustace or his party.  Instead the improvement was either a factor of the statistical margin of error or “pull” factors such as the prevailing sentiment among critical “Uncertain Voters” that mitigates against a fourth term.  With regard to the former statistical point it should be noted that CADRES polls have previously always understated ULP support to some extent and this overstatement is therefore uncharacteristic and suggestive of a political mood related to this unique scenario.

Polls aside, the historic nature of the 2015 election needs to be put in context since it can be agreed that four straight Gonsalves victories are neither a Caribbean nor a Vincentian democratic first.  The most recent post-independence “four-peat” was achieved by Dr Denzil Douglas in St Kitts and Nevis, while the NDP’s own Sir James Mitchell won four times between 1984 and 1998.  Dr Eric Williams of Trinidad and Tobago is also in that category since he won elections there in ’66, ’71, ’76 and ’81 and promptly died thereafter.

In that category is also Sir John Compton of St Lucia who won elections there in ’82, ’87, ’87 and ’92; however, the fact that two of these elections were held in one year means that he would not have competed on the same playing field as the others.

If then we narrow the comparative analysis to Douglas/Mitchell/Williams, the historic peculiarity of the Gonsalves victory becomes clear and is facilitated in the appended chart.  In all three instances those leaders lost support and, in some instances, seats on their fourth term.  Douglas emerged with less votes and seats in 2010; while Williams gained seats (81), but ceded minus one per cent national support. Mitchell’s scenario is somewhat ironic since he ushered, in the Gonsalves era in 98 with an “unpopular victory” reflecting a loss of seats and votes.

Interesting also is the fact that Douglas, Mitchell and Williams had previously won all the seats their party contested, although in the case of Williams his victory was on account of a boycott by opposition parties.

The case of Gonsalves, therefore, is historic for two concrete reasons, with the first being that he has never been fortunate enough to gain all the seats he contested and has prevailed four times from a less favourable electoral perch.  Secondly and more importantly is the fact that his fourth-term victory has been achieved with more support than he secured in the previous election.

 The plus-one per cent swing reflected is admittedly marginal, but represents a reversal of the minus-four per cent swing he suffered in 2010 and means his ULP is more popular on its fourth term than it was on its third which is a noteworthy historic achievement.

Peter W. Wickham is a political consultant and a director of Caribbean Development Research Services (CADRES). Email [email protected]