Counting Some of the Votes in Haiti

by Brian Concannon

IJDH — Haiti's elections on February 7 went well enough that the post-election vote counting should have been uncontroversial. The turnout was huge, there was almost no violence, and the people's choice was so clear that the second place finisher received less than 12% of the vote. But incredibly, a week later the final results have not been declared, and the Electoral Council is in disarray. The voters have taken to the streets to protect their vote, and the clear winner is alleging fraud.

The battle lines have been drawn around the 50% of the total vote that former President Rene Preval needs to avoid a runoff election against his distant nearest challenger. Initial official results and unofficial reports had Mr. Preval comfortably above that bar, but his official numbers crept steadily downward over the last week. As of Tuesday morning, with 90% of the votes counted, Mr. Preval was stuck at 48.7%, 22,586 votes shy of outright victory.

What's At Stake

In a better world, Mr. Preval would be happy to go into a runoff with a 48.7% share, assured that he could attract 1.3% of the voters more easily than his opponent, Leslie Manigat, could attract 38%. Mr. Manigat might even save his country time and money by conceding an obviously futile contest. But this is Haiti, where electoral support does not always translate into political power. Mr. Preval and his supporters know that the vote only came close to 50% because the votes of Haiti's poor- who overwhelmingly voted for Mr. Preval- had been systematically suppressed through a series of irregularities, from the voter registration last summer through election day. They draw a line from this vote suppression through questionable tabulation practices, and see it pointing towards a second round somehow stolen from them.

Mr. Manigat may have Haiti's history on his side, if not Haiti's voters. He knows from experience that there are many routes to Haiti's Presidency, not all of them requiring electoral support. He ran in the first elections under Haiti's current Constitution, in November 1987, and was projected to run a distant third at best. But the army and paramilitaries stopped the voting by firing at voting centers, killing at least 34 people. Two months later the army ran new elections. The candidates with democratic convictions called a boycott of the charade, which the voters supported. But Mr. Manigat, Hubert de Ronceray (who won less than 1% this year) and one other candidate threw their hats in the ring, and the army declared Manigat its President.

Last week's election was Haiti's fourth Presidential election since 1990. The previous three- 1990, 1995 and 2000- were all conducted without serious violence. Each time, the voters delivered a landslide to the candidate of the Lavalas political movement-no runner-up ever topped 16% of the vote. But each time a minority in Haiti, usually with outside support, successfully limited this mandate. President Jean-Bertrand Aristide, the victor in the first and third of those elections, suffered two successful coup d'etats, and spent half of his two terms in exile. President Preval managed to spend his whole term in office and pass power to an elected successor (the first Haitian President to do so), but a manufactured political crisis and perpetual squabbling about the extent of the Lavalas landslides prevented the seating of a legislature. More important, the crisis successfully diverted President Preval's energies and attention from the economic and social development policies he was elected to implement.

Mr. Preval did not run this year under the banner of the Fanmi Lavalas party, but with a brand-new party, Lespwa (Hope). Fanmi Lavalas boycotted the elections because the Interim Government of Haiti (IGH) refused to stop its persecution of the party, which included jailing dozens of political opponents, attacking anti-IGH protests and mounting murderous police raids in the poor neighborhoods that were the party's strongholds.

But Preval's victory was nonetheless delivered by the Lavalas base. Voters said as much to anyone who would listen as they waited to vote, afterwards, and in this week's demonstrations. More tellingly, Preval won his landslide with almost no institutional support or even campaigning. The Espwa party is brand new, fielding candidates in barely half of the senatorial races. Preval received almost no formal endorsements, and did not even speak publicly until the last weeks of the campaign. He planned very few rallies, and many of these were cancelled after two events were violently attacked. But despite these handicaps, he won a landslide because the Lavalas base voted overwhelmingly for him (candidate Marc Bazin claimed the Lavalas mantle, but had the support of neither the party's top leadership nor its base, and won less than 1% of the vote).

Pre-Election Vote Suppression

The IGH engaged in a comprehensive program to suppress Lavalas political activities in the ten months before the elections. Several prominent politicians were not able to participate, as candidates or activists, because they were kept in jail illegally, including Haiti's last constitutional Prime Minister, a former member of the House of Deputies, the former Minister of the Interior, and dozens of other local officials and grassroots activists. When Haiti's most prominent dissident, Rev. Gerard Jean-Juste, was diagnosed with leukemia, it took a massive campaign, including intervention of top U.S. Republicans, just to obtain his provisional release for desperately needed treatment.

Making Registration Difficult

The voting registration process systematically discouraged poor rural and urban voters from signing up. Where Haiti's democratic government provided over 10,000 voter registration centers for elections in 2000, the IGH installed less than 500. The offices would have been too few and far between for many voters even if they had been evenly distributed. But placement was heavily weighted in favor of areas likely to support the IGH and its allies. Halfway through the registration period, for example, there were three offices in the upscale suburb of Petionville, and the same number in the large and largely roadless Central Plateau Department. In cities, the poor neighborhoods were the last to get registration centers, and Cite Soleil, the largest poor neighborhood of all, never got one. Complaints and protests forced the IGH to extend the registration period three times and open additional registration facilities. Eventually over 3.5 million voters registered, about _ of the estimated eligible voters. But we will never know how many voters could not get to a registration center, or gave up after losing too many precious work days in the effort. We do know that the registration difficulties disproportionately impacted the rural and urban poor, who voted overwhelmingly for Preval.

Making Campaigning Difficult

Neither Lavalas nor the Preval campaign was able to effectively engage in pre-election campaigning. Police repeatedly fired guns at peaceful pro-Lavalas demonstrations throughout the two years of the IGH's reign. In January, a pro-government gang destroyed structures erected for a Preval campaign speech in the town of St. Marc, cancelling the event. No arrests were made. Violence and threats of violence forced the cancellation of subsequent events, even the campaign's grand finale the week before the election.

Election Day Vote Suppression

The IGH had limited the voting centers to 807, which would have been inadequate even if the elections had run smoothly (Los Angeles County, with a slightly larger population but only 37% of Haiti's land area and infinitely better private and public transportation had about 4,400 polling places in November 2005). But by 1 PM on election day, Reuters' headline read: "Chaos, fraud claims mar Haiti election." Most election offices opened late and lacked ballots or other materials; many did not become fully functional until mid-afternoon. Voters arrived at the designated centers to find the center had been moved at the last minute. Many who found the center identified on their voting card waited in line for hours only to be told they could not vote because their names were not on the list. At some centers, tens of thousands of voters were crammed into a single building, creating confusion, and in one case a deadly stampede.

As with the registration deficiencies, the poor bore the lion's share of the election day problems. The two voting centers for Cite Soleil, both located well outside the neighborhood, saw the worst. One of the two, the Carrefour Aviation site, was transferred at the last minute to a single building where 32,000 voters had to find the right line to wait in without posted instructions, lists of names or an information center. Throughout the day, journalists and observers noted over and over that centers in Petionville and other wealthy areas were better organized and equipped.

As with registration, many voters persevered despite the obstacles. After frustrated would-be voters took to the streets in spontaneous protests, the IGH made concessions, such as keeping the polls open later and allowing people with voting cards whose names were not on the local list to vote in some places. By the end of the day, most voting centers were operating at a minimal level, and over 60% of registered voters did vote. But we will never know how many people gave up, because they were sick or frustrated or needed to get back to their families.

Counting Some of the Votes

After the problems with registration and voting, Mr. Preval's supporters were pleasantly surprised that the CEP gave him a large lead in initial reports. On Thursday, the CEP announced that with 22% of the votes counted, Preval had a commanding lead with 62% of the vote. Mr. Manigat trailed at 11%, and Charles Henri Baker, in third place, had 6%. Unofficial reports of the local results from international and Haitian observers and journalists consistently had Preval far over 50%. But by Saturday night the Preval's official vote had decreased to 49.61%, and by Monday it was at 48.7%.

The IGH claims that Preval's decrease was the result of more information coming in and better calculations. But many questions about the tabulation process, combined with the efforts to suppress the Lavalas vote before and during election day, raise doubts about those claims. On Tuesday afternoon, Mr. Preval claimed that he had proof that he won 54% of the vote, and that the Electoral Council fraudulently reduced his number.

Who's In Charge?

The Electoral Council is supposed to be running the counting, but it is not. Jacques Bernard was appointed "Executive Director" of the Council- a position not previously recognized in Haitian law- by the Prime Minister late last year. He is running the show and has kept regular Council members out of the counting room. Councilor Pierre Richard Duchemin charges "manipulation," and "an effort to stop people from asking questions." Another Councilor, Patrick Fequiere, claims that Mr. Bernard is working without the Council and not telling them where his information is coming from. The UN Peacekeeping mission was forced to remove the doors to the tabulation center to prevent Mr. Bernard and his advisors from acting secretly.

Uncounted Votes

A large number of tally sheets from polling centers are not being counted. 254 sheets were destroyed, reportedly by gangs from political parties opposed to Preval. 504 tally sheets reportedly lack the codes needed to enter them officially. The missing tally sheets probably represent about 190,000 votes- over 9% of the total votes cast- and according to the UN, disproportionately affect poor areas that support Preval.

Null and Blank Votes

Electoral officials have also discarded 147,765 votes, over 7% of the total, as "null." Article 185 of the Electoral Code allows officials to nullify ballots if they "cannot recognize the intention or political will of the elector." The Presidential ballots were complicated- 33 candidates, each with a photo, an emblem and the names of the candidate and the party. Some Haitian voters, unused to filling out forms or writing, undoubtedly made mistakes-like marking two boxes- that made determining their choice impossible. But 147,765 voided votes is a lot, especially when that decision was made by local officials handpicked by an Electoral Council that had no representation from Lavalas or Lespwa. Overly strict criterion (such as requiring an "x" to be completely within a candidate's box), even if neutrally applied, would have a disproportionate impact on Preval voters, who are more unused to filling out forms than their better-heeled compatriots, and therefore more likely to make mistakes.

Another group of votes, 85,290, or 4.6%, are classified as blank ballots. These votes are actually counted against Preval, because they are included in the total number of valid votes that provides the baseline for the 50% threshold. This is a potentially reasonable system, just unreasonably applied to Haiti. It allows voters to show their displeasure with all the candidates by voting for no one. It makes sense in wealthy countries, but it is absurd to think that 85,000 people would leave their babies, their fields and other work and spend hours walking or waiting in the tropical heat just to say they did not like any of the 33 candidates. A more likely explanation is that illiterate voters got confused by the complicated ballots and marked nothing. Again, this problem would disproportionately affect poor voters likely to vote for Preval. But even if it did not- if the blank votes were allocated to candidates based on their percentage of other votes- Preval would clear 50%.

The blank and null ballots combined exceeded Mr. Manigat's vote by 17,000. The rules for blank and null votes are consistent with previous Haitian elections, so it is hard to call the rules themselves fraudulent. But the scale of the distortion of the vote caused by these rules was both foreseeable and preventable. The same problem has arisen at every election since 1990, most of which were observed by the UN and the Organization of American States, which were active in preparing the elections this time around. The distortion could be sharply reduced with a simple voter education campaign: going into poor neighborhoods, showing how to mark ballots and giving voters an opportunity to practice on sample ballots. There was money available for such a program- the election cost over $70 million dollars, most of it coming from abroad, more than $30 for every vote cast. The political parties, many of which represented a fraction of one percent of the electorate, received generous subsidies. But no concerted effort was made to help the much larger share of the voters who had demonstrated difficulty with filling out the ballots.

Taking the Streets

Haiti's voters may be inexperienced in filling out forms, but they have seen enough stolen elections to qualify as world-class experts in the field. They can trace the pattern from registration through election day to the current calculations, and they can see their votes discounted at every step. They know that they did enough to win according to the rules of the game, which they believe in. But they know that voting, in Haiti, is not enough, so they are now out in the streets by the thousands, erecting barricades, protesting, even occupying the pool at the luxurious Montana Hotel, where the votes are counted and the journalists and other expatriots are lodged.

The IGH and the US government have responded by calling on Preval to call off the protests. He implored his supporters not to damage people or property, but also recommended that they keep demonstrating until the IGH stops trying to steal the election. Haiti's voters will undoubtedly take this recommendation. They have done their job in marking their ballots, but know that they need to make sure that the IGH counts them.

Brian Concannon Jr., Esquire, directs the Institute for Justice & Democracy in Haiti, www.ijdh.org, and observed several elections in Haiti for the Organization of American States.