May 31, 2007 · Posted in: 2007 Elections
(This report was written by PCIJ intern Julienne Urrea)
HALALANG Marangal’s (Halal) initial findings reveal significant discrepancies in tallies for senatorial candidates in nine provinces and one city based on data from the Commission on Elections (Comelec) and the National Citizens’ Movement for Free Elections (Namfrel).
In Negros Occidental, the top vote gainer was Genuine Opposition (GO) candidate Francis Escudero with an additional 65,000 votes in the Comelec’s provincial certificates of canvass (PCOCs), compared with provincial breakdowns released by Namfrel based on election returns (ERs).
In neighboring Negros Oriental, TU candidate Luis Chavit Singson gained 17,800 votes while independent candidate Francis Pangilinan lost 11,600 votes. The top five gainers are TU candidates while the top five losers are GO and independent candidates.
Halal noted however that Namfrel had a low completion rate of 26.6 percent in the province and that close observation would be needed at a higher completion rate. Halal said that comparisons may not be accurate due to the low completion rate of Namfrel tallies as compared to Comelec tallies.
But even in provinces like Mandaluyong City, where Halal said there is a relatively close agreement between Comelec and Namfrel tallies, discrepancies exist.
TU candidate Juan Miguel Zubiri is among the top gainers with 7,100 votes while GO candidate Aquilino Pimentel III is among the top losers with 4,100. Zubiri, who ranks 13th in the senatorial race, lags behind Pimentel by 150,017 votes.
See Halalang Marangal’s initial report.
Halal said in a statement that “the discrepancies identified in our initial reports do not automatically mean cheating by the candidates who apparently gained votes from the discrepancies. Neither do we say that no cheating occurred.”
Halal co-convenor Roberto Verzola said in a statement that there are four reasons for discrepancies between the two counts: averaging errors, ERs that were set aside by Namfrel or Comelec, clerical or encoding errors, and intentional alterations. Instead of comparing the actual number of votes in each province, Verzola compared a candidate’s percentage share of total votes (PSTV) in a province under the Comelec and Namfrel tallies.
Ideally, he said, the candidate’s PSTV under both tallies should be the same, because both tallies are based on the same ERs. Namfrel results came directly from the precincts which, according to Verzola, is “closest to the truth” whereas Comelec results went through a circuitous process — from municipal to provincial to national canvassing, where vote-padding and-shaving usually occurs.
Halal’s initial report consists of 10 spreadsheets showing the top 12 gainers and losers of votes in Bataan, Batanes, Quirino, Mandaluyong City, Catanduanes, La Union, Romblon, Negros Occidenta, Negros Oriental, and Agusan del Norte. The report aims to reconcile discrepancies between Comelec and Namfrel results and look for the reasons behind such discrepancies.
The election watchdog admits that it is too early at this point to say that there are trends in the discrepancies of the tallies. Verzola said that if such a trend exists, it would indicate “more than just random errors and random clerical mistakes.”
The Halal audit is similar to double-entry accounting, which makes it easier to spot errors.
“Without Namfrel ERs, we could not prove the balancing. That’s an official part of the law, there should be Namfrel ERs and these should be given to the citizen’s arm so that we could conduct parallel counts and compare the two results (Namfrel and Comelec results),” Verzola added.
He said that the citizen’s audit can only detect cheating in the provincial level since Namfrel has only provided Halal with a provincial breakdown. “If Namfrel releases the municipal breakdown, we can do a direct municipal to municipal comparison and identify where [cheating] occurred in the municipal level.”
Verzola said that the citizen’s audit can only identify cheating in the final stage of the whole process, which is during the canvass in the municipal, provincial and national levels. “This is not a ‘cure all’ that can solve all the cheating problems.”
Halal will not tackle other forms of cheating such as the use of government resources, overspending, ghost voters, vote-buying, or voter disenfranchising.
“We are really trying to do our part in plugging this loophole for now,” says Halal co-convenor Ma. Paz. Luna. “If we found this effective and strategic, I think many would join us. If that is the case, then we can plug more loopholes.”