More than 1 in 10 people who are registered for July’s national elections appear not to exist and 9 percent of past voters have been taken off local voting lists unfairly, according to an independent audit of the country’s latest voter registry released yesterday.
By just about every measure, the voter list is worse than it was heading into the last national election in 2008, said Laura Thornton, country director for the National Democratic Institute (NDI), one of the organizations that conducted the audit.
“The key finding is that overall, there has been a decline in all aspects of the quality of the voter list in Cambodia,” she said. “There is a decline in comprehensiveness, there’s a decline in accuracy and there’s a decline in… validity, compared to 2008.”
A joint effort by NDI, the Neutral and Impartial Committee for Free and Fair Elections in Cambodia (Nicfec) and the Center for Advanced Studies, the audit adds weight to recent warnings from another local election monitor that this year’s elections are shaping up to be the least fair in the 20 years since the U.N.-sponsored national election of 1993.
“These findings reveal flaws that may have implications on the quality and legitimacy of these elections,” Ms. Thornton said.
“I am concerned by the large number of eligible citizens who will show up on election day to discover their names are not on the list and can’t vote either because they were incorrectly deleted or think they are registered but are not. I’m equally concerned that there is a significant number of names on the list that are not attached to real people.”
Among the nearly 4,900 Cambodians that Nicfec volunteers randomly sampled across the country last month, they could not find 10.4 percent of the people whose names were on the national voter list.
In other words, Ms. Thornton said, “the names on the list are not attached to existing people.”
That compares to only 7.7 percent of registered names that did not match up with real people in 2008.
The National Election Committee (NEC) announced earlier this year that the voter list consisted of a total of 9.6 million people.
“I want to emphasize quality control here,” Ms. Thornton said. “We looked for these people. We talked to pagoda heads. We talked to neighbors. We tried to find family members. And every single one of the unknown voters was confirmed not to exist by the village chief.”
The audit also found that the data commune election offices had on voters was becoming increasingly inaccurate.
For example, the birth dates of voters on their national ID cards and the dates recorded on the national registry matched up only 63 percent of the time this year, compared to 78 percent of the time in 2008. Names matched up only 86.4 percent of the time.
Ms. Thornton said that those numbers could have major implications come election day.
“What this means is that people show up at the polling station and they have their identification documents. The polling station official looks at the data on the list, looks at the ID documents, and can likely turn away these people because the data does not match,” she said.
While election rules allow for some flexibility on names and birth dates, she added, “that puts a lot of discretion in the hands of polling station officials. If the data doesn’t match for someone, they do have the right to not allow them to vote.”
The auditors looked at the list of names taken off the registry and found that 9.4 percent had been wrongly removed, just as many as in 2008.
Again, Ms. Thornton stressed the lengths to which the researchers went to get their numbers right.
“We…checked their name in their polling station where they were deleted and checked to see if their name was still at the polling station and maybe they just deleted a duplicate name,” she said. “We also checked them in other logical locations where they might be. So the point of all this is the quality control of this operation gives us a lot of confidence in the data we found.”
Nicfec Director Hang Puthea said more than 1 in 10 of the people that the volunteers surveyed said they were expecting to vote but were not in fact registered, setting up the prospect of mass disappointment on election day.
According to Ms. Thornton, all this raises the odds of more people trying to vote under names that are not their own.
“It increases the risk because there are names on the list not attached to real people,” she said.
Mr. Puthea and Ms. Thornton would not speculate which party had the most to gain from the irregularities, as the volunteers eschewed questions about political affiliation in order to get the most honest answers out of their subjects.
But, with the commune councils in charge of voter registration, and the vast majority of those councils firmly in the hands of the ruling CPP, most election observers believe these long-standing complaints with the voter list run in the CPP’s favor.
The NEC itself is stacked with ex-CPP officials and the Interior Ministry refuses to reveal how members are selected, or to let any other party have a say in the selection process.
Just last week, the Committee for Free and Fair Elections in Cambodia said the government is actually moving further away from a democratic vote, and that this year’s vote was likely to be the least fair since 1993.
Some of the country’s Western donors have also called for the government to let self-exiled opposition leader Sam Rainsy return and stand in July’s vote.
The groups behind the latest audit made a number of recommendations.
Long term, they want the government to take voter registration duties away from the elected—and naturally partisan—commune councils, and give them to a genuinely independent body.
More immediately, they want the NEC to let observers stand behind the polling clerks as they check IDs on election day.
Contacted yesterday, NEC Secretary-General Tep Nytha was dismissive of the audit’s results.
“The report has a lot of suspicious points. Some figures are not right,” he said. “We have 900 polling stations, but they found only 144 polling stations. I don’t know how they did their job.”
Mr. Nytha appeared to be confusing the number of nationwide polling stations with the number of sites the groups said they had visited for their audit. He also said their last audit also left much to be desired.
“In 2008, they said there were people who did not have their names on the list,” he said. “They gave us 88 people to check. After we checked, more than 50 people had their names on the list.”
Even so, he said the NEC would check their figures if the groups sent them a list of names.
