The Tabulation Process
Tabulating results represents a critical part of the electoral process, requiring extensive efforts to ensure accuracy, minimize any opportunities for fraud, and guarantee that election results represent the free expression of the will of the voters.[4] To ensure public confidence, tabulation procedures should follow five fundamental principles: transparency, security, accuracy, timeliness, and accountability.

The Carter Center congratulates HNEC staff for their diligence in completing the tabulation process in a timely manner. Despite some barriers to transparency and a lack of strong measures to identify anomalous results, the tabulation process for the July 7, 2012, General National Congress elections was conducted in a credible manner and was adequate to determine the results of the election accurately.

The tabulation procedures were released by the HNEC on June 30, 2012. The procedures outline the manner in which copies of the results forms should be posted at the polling station and the originals transferred in tamper-evident envelopes to the National Tally Center. They include provisions for a double-blind data entry system to ensure accuracy and an audit committee to review results forms containing clerical or other errors. In the case that the audit committee cannot resolve an error, results forms are referred to the HNEC board of commissioners to authorize a recount. The regulations do not reveal the process for electronic aggregation of the results.

Carter Center observers noted that HNEC officials at the tally center sought to be transparent and were very welcoming to observers and agents of political entities and candidates. However, the computerized aggregation of results and the layout of the tally center did not permit comprehensive observation of the tally process. In future elections, observation of the aggregation process should be facilitated, for instance by posting results forms as they are entered in the database or by projecting results forms on a screen that is easily visible to observers and political entities and candidates' agents.

Recounting is an inevitable process in most elections and requires clear procedures to ensure accuracy and transparency. Carter Center observers monitored recounts, or the re-opening of ballot boxes to extract results forms that had been mistakenly sealed in the ballot boxes, in five cities: Al Bayda, Benghazi, Misrata, Tobruk, and Tripoli. In all but one case, recounts were triggered by missing or incomplete forms or auditors being unable to make sense of the figures on the results forms.[5]

In most cases, HNEC officials actively requested the presence of observers in order to ensure the transparency of recounts. In all observed cases the recounts were conducted in a professional manner with training or polling staff showing a strong commitment to accuracy. Carter Center observers did not report any cases of intentional manipulation of results.

The Carter Center notes three issues that could be addressed in future elections:

First, according to the election law and regulations, both counting and recounting should take place in the district where voting took place. This ensures the transparency of the process by allowing polling staff, observers, and political entity and candidates' agents to be present. While recognizing serious security concerns, the decision to count and recount ballots from Kufra in Tripoli is inconsistent with the law and violates the principle of transparency.

Second, if a mistake is made by a member of the polling staff, he or she should be present to append their signature in agreement with the correction. Failure to observe this procedure jeopardizes the principle of accountability. Nevertheless, on several occasions auditors at the tally center appeared to correct calculations or typographical mistakes without the knowledge of polling staff. Means for polling staff to be included in the process should be examined, including decentralization of the tabulation process.

Third, triggers for quarantining questionable results were extremely limited. According to HNEC officials, the database was programmed to quarantine results forms in which the data was inconsistent, the number of votes received by candidates exceeded the number of votes cast, or turnout was greater than 100 percent of registered voters. In the future, the HNEC should consider employing additional and more stringent quarantine triggers, for instance, to detect over 95 percent votes for a one candidate; extremely high turnout in a particular polling station; or discrepancies in the reconciliation of used, unused, spoiled, and cancelled ballots with final vote totals.

The HNEC should be praised for the way it handled delays in the tabulation process by reassuring candidates and voters, holding frequent press conferences, and announcing partial results. These measures helped to increase transparency, reduce potential distrust of the tabulation process, and reassure voters that the process was not subject to undue delays.