The Center for Economic and Policy Research has posted the full OAS election report on its website [PDF], and after analyzing the report, released the following press release. Also see Robert Naiman's analysis on the parallels between what the OAS is advocating for in Haiti and the Bush vs. Gore recount in 2000.
Washington, D.C.- The Center for Economic and Policy Research (CEPR) has analyzed the Expert Verification Mission's Final Report from the Organization of American States (OAS) on Haiti’s presidential elections, which has not been released to the public but is now available on the CEPR website here [PDF]. CEPR’s analysis found that the OAS report cannot help determine the outcome of the first round of Haiti’s election.
“This report can’t salvage an election that was illegitimate, where nearly three-quarters of the electorate didn’t vote, and where the vote count of the minority that did vote was severely compromised,” said Mark Weisbrot, CEPR Co-Director and co-author of the report, “Haiti’s Fatally Flawed Election.”
CEPR has been unable to find a presidential election in the Western Hemisphere, including Haiti, with such a low turnout, going back to 1947. Haiti’s parliamentary election of 2009, in which the country’s most popular political party was also banned, had a turnout of less than 10 percent.
The OAS report does confirm some of the most important conclusions from CEPR’s analysis of the elections, which was published on Sunday. For example, the OAS finds that 12 percent of the tally sheets were either not received by the Provisional Electoral Council or were quarantined – a much larger number of lost votes than the OAS or the Provisional Electoral Council (CEP) had previously publicly acknowledged.
Yet the OAS report concludes that the election should not be re-run, but rather that the results should be changed so that Michel Martelly, rather than the government candidate Jude Celestin, would finish second and therefore proceed to a run-off election.
But it is easy to see that there is no sound basis for such a conclusion. For example, the missing tally sheets came from areas that had a different distribution of votes than the average for the country as a whole – one that favored Celestin. CEPR’s analysis found that if the missing tally sheets had a distribution that was the same as the other tally sheets that were received from the same areas, then Celestin would have finished second, rather than third.
The OAS analysis was also methodologically and statistically flawed in numerous other ways. Unlike the CEPR report, which examined all of the 11,181 tally sheets, and subjected each of the vote totals of the top three candidates to a statistical test to look for irregularities, the OAS team focused on tally sheets that had unusually high voter participation levels. They then subjected this set of tally sheets to the following criteria:
“In accordance with this provision of the law, the Expert Mission set four specific criteria to determine if a PV [tally sheet] should be included: 1) the inclusion or absence of the required signatures of the polling officials on the Procès-Verbal [tally sheet]; 2) the inclusion or absence of the list of registered voters; 3) the presence and accuracy of the CIN [voter national identity] numbers to identify those voters who cast their ballots at that particular polling station; 4) if a Procès-Verbal [tally sheet] had been obviously altered to change the results of the elections, for instance adding a digit to a number to increase a vote total by a hundred or more, that PV [tally sheet] was also excluded.”
On this basis, the OAS team threw out 234 tally sheets, and with the remaining tally sheets calculated the results.
“This methodology really tells us nothing about who really finished second in the first round of the election, even among the small minority of voters that actually voted and had their votes counted,” said Weisbrot.
Weisbrot also noted that the small margin of difference between Martelly and Celestin in the OAS’s recount – 0.3 percent – was too small to statistically distinguish between the two, given the sample size and variance.
“This appears to be a political, and not a professional, decision,” Weisbrot added.