Sacrifice accuracy for timelier elections
May 24, 2004
Realistic time-lines should be set for the next local government and general elections, according to a new study which recommends that GECOM negotiate for lowering the benchmarks for accuracy if the stakeholders want a shorter period.
The study of Compilation, Transmission and Publication of Elections Results was undertaken for GECOM by Mike James, an experienced monitor of elections here and abroad.
The report says the elections commission should prepare a timetable which includes realistic timelines for near perfect elections.
But it also suggests that should the stakeholders require a shorter period, the elections commission "should transparently negotiate for consensus on lowering benchmarks of accuracy and efficiency."
One example is lowering the accuracy level of the Official Voters' List.
It is recommended that the accuracy level of the Official Voters' List be 97% if GECOM wants to deliver near perfect elections. But James recommends that the accuracy guaranteed can be lowered to 95% given the shorter preparation time available to the commission.
The paper, which has recommendations to inform a Revised System for Declar-ation of Election Results, says an analysis of the 1992, 1997 and 2001 general elections highlights severe problems posed by deadline constraints.
Two of those elections followed postponements of election dates to give the Elections Commission the minimum time required to hold credible elections. It is noted that inadequate time to appoint sufficiently competent and trained staff contributed to the crisis surrounding the 2001 election.
In contrast, the last local government polls in 1994 are described as smooth because they were conducted within a manageable time-frame. Ten months is estimated as the minimum time needed for preparations for local government elections, provided the needed legislation for continuous registration and the local government system is passed, and agreement is reached on the use of the current database.
Both the PNCR and the WPA have raised objections in relation to the use of the database, while there has been no agreement on an electoral system for the local government elections.
Meanwhile, the EU needs assessment, completed in March 2000 for the 2001 general election, projected a minimum 10 months' preparation period for 2001 elections and an extra two months if the decision was taken to issue new ID cards.
The decision was made, and with that GECOM requested that the elections be postponed from January to March to facilitate distribution of the cards. Over 20,000 cards were not distributed by election day.
Also, extremely tight deadlines for the production of the Preliminary Voters' List, the Official Voters' List and Addendum resulted in significant repeated errors, voter omissions from the list and other problems during and after polling day.
James concludes that GECOM should take a conservative approach in projecting for local government and general elections scheduled by June 2006, given the past elections experience of slipping time-lines, unforeseen delays, and the parliamentary agenda.
He also says GECOM should calculate the maximum possible time required for it to complete processes and present them to the stakeholders for the delivery of near perfect elections.
It is also recommended that the detailed projections be shared with the public so that it is clear which deadlines GECOM is accountable for and those for which the parliamentary stakeholders are accountable.
Time limit for publication of official results
The paper recommends that the statutory deadline for announcement of official results should be fifteen days after election day.
GECOM has exclusive responsibility for the publication of election results, which is spelled out in the Representation of the People Act.
But one key area which is not specified in the law relates to a time limit for the announcement.
Trinidad and Tobago has a time limit of 12 days, while in the United States the legislation sets one month for the announcement.
James notes that the dispersed population and difficult terrain as well as the history of political polarisation at election time and serious tabulation errors at polling stations, are some of the factors which have resulted in delays in the release of results. Official results were produced 16 days after the 1997 elections. Results for the last election, however, were declared four days later.
GECOM staff and commissioners worked long hours under public and stakeholder pressure to make the declaration, which was done at four in the morning.
James suggests that a conservative deadline be set, in view of past post-election problems that GECOM has no direct control over. But, he also says that setting a conservative limit does not mean that GECOM will need all the time available.
In fact, he says that with sound planning and organisation, GECOM can declare official results within four days.
He does caution though, that the time limit must be sufficiently ample to cover unforeseen problems that have plagued past elections.
The statutory deadline is expected to be beneficial to GECOM staff and commissioners, who will be given adequate time to validate the results from all the polling stations, free from being pestered by the media and others for the results before they are ready for the Chief Elections Officer to make the official declaration.