U.N. Data Show Discrepancies in Afghan Vote

By Colum Lynch and Joshua Partlow
Washington Post Staff Writers
Wednesday, October 7, 2009

UNITED NATIONS, Oct. 6 -- Voter turnout data kept confidential by the United Nations' chief envoy in Kabul after Afghanistan's disputed August presidential election show that in some provinces the official vote count exceeded the estimated number of voters by 100,000 or more, providing further indication that the contest was marred by fraud.

In southern Helmand province -- where 134,804 votes were recorded, 112,873 of them for President Hamid Karzai -- the United Nations estimated that just 38,000 people voted, and possibly as few as 5,000, according to a U.N. spreadsheet obtained by The Washington Post.

The disclosure of the data seems likely to worsen a credibility crisis for the U.N. special envoy, Kai Eide, who is already facing allegations that he sided with Karzai. In the past week, two U.N. political officers in Kabul have resigned because of a lack of confidence in Eide's leadership, according to U.N. officials who spoke on the condition of anonymity to discuss personnel issues.

The departures were triggered by U.N. Secretary General Ban Ki-moon's decision last week to fire Eide's American deputy, Peter W. Galbraith, after he accused his boss of failing to provide Afghan and international officials with evidence of fraud, primarily by Karzai's supporters.

Galbraith pressed Eide to turn over to international monitors the United Nations' estimated turnout data, which indicated that many fewer voters cast ballots in certain provinces than the number of votes recorded by election officials. Galbraith said Eide refused to share this data with the internationally led Electoral Complaints Commission once it became clear that the information reflected poorly on Karzai.

In an interview last week, Eide acknowledged withholding the data, saying that the information could not be verified and that he required a formal request in order to share it. He said he was confronted by a "confusing situation" in which "a lot of information was coming from sources that had their own agenda. We can't just hand over a bunch of information if we haven't made a solid assessment of it."

Eide added that he "really feels offended" by allegations that he favored Karzai, saying he had taken a balanced approach that enjoyed the "unanimous" support of the international community.

The U.N. spreadsheet shows widespread discrepancies between turnout and results, particularly in the volatile southern and eastern provinces where Karzai won with large margins. There are also allegations of fraud by followers of Abdullah Abdullah, Karzai's main competitor, but on a lesser scale.

Diplomats in Kabul have previously referred to such discrepancies, but the U.N. data have not been publicly disclosed until now.

In Paktika province, for example, where Afghanistan's Independent Election Commission has reported that 212,405 valid votes were cast, including 193,541 for Karzai, the United Nations estimated that 35,000 voters turned out. In Kandahar province, which recorded 252,866 votes, including 221,436 for Karzai, the United Nations estimated that 100,000 people voted.

In several provinces won by Abdullah, the United Nations estimated a larger turnout than election officials recorded. In Balkh province, for example, the organization estimated that 450,000 people voted, while the results showed 297,557 votes, 46 percent of them for Abdullah.

Although the estimates in some cases include a broad range of possible turnout, Galbraith said it was important information to share with Afghan officials and international monitors. "I favored turning it over to the Electoral Complaints Commission," he said. "I think we did an excellent job at collecting data. . . . We collected it with the idea of assisting the Afghan legal party that was investigating fraud, but Kai opposed turning it over."

CONTINUED     1        >

© 2009 The Washington Post Company