Hocus PCOS: the specter haunting the 2013 elections

I wrote this article for Philippine Online Chronicles.

A specter is haunting the recently concluded 2013 mid-term elections, the specter of alleged massive automated fraud to benefit the ruling party of President Noynoy Aquino.

Like every other election in the Philippines, this year’s race was not spared from massive vote buying, streets and airwaves clogged by political ads and posters, and election-related violence.

But while the trapo “3G technology” of guns, goons, and golds seems to have proven decisive in clinching the votes, what is special with the 2013 polls are the massive glitches, glaring delays in counting the results, and serious allegations of automated fraud.

This is the second implementation, after the 2010 presidential elections, of the Automated Election System (AES) which was designed and controlled by the Venezuelan company Smartmatic.

Technological Disaster

The Commission on Elections (Comelec) said polling was generally smooth with only 200 Precinct Count Optical Scan (PCOS) machines suffering from glitches.

But Comelec Chief Sixto Brillantes would later admit that around 18,000 PCOS machines or a quarter of the total 78,000 machines malfunctioned.

On the eve of the polling day, the Comelec boasted of being 99.99999 percent ready, saying most winners may be proclaimed within 2 or 3 days.

However, because of various problems with the AES, all winning senators were announced a week after this pre-election day pronouncement. It would take 15 days before Comelec can finally announce the results of the party-list race.

All these problems have led independent election watchdogs to declare the recently concluded polls as “technological and political disaster.”

The AES Watch said that PCOS defects – from the machine failing to initialize, getting bogged down, rejecting ballots, failing to transmit election returns to the Comelec servers and up to to defective Compact Flash (CF) cards – affected if not potentially disenfranchised 8.6 million voters.

Brillantes tried to pin the blame on the failure of data transmission to the lack of cellular signal or network congestion. Telecom companies quickly responded that their combined signal cover pretty much all of the country and that they delivered whatever data was there to be transmitted.

60-30-10” Pattern

But the various PCOS glitches and defects were not the only problems hounding the much-vaunted AES. Graver still are the suspicions of automated fraud and pre-programming of results stemming from a statistically improbable “60-30-10” pattern for the votes obtained by senatorial candidates.

Sorting through the tallies of the senatorial race from the first to the 16th canvass reports, Ateneo de Manila University professor Lex Muga observed that the administration coalition consistently got 60 percent, the opposition United Nationalist Alliance 30 percent, and the independents 10 percent.

The same disturbing pattern was noted by the blogger Conrad Gozalo of Radar Sweep using the tallies in all 11 intervals released from the news website Rappler.com, the official media partner of Comelec.

Brillantes countered by accusing critics of having “dirty minds” that are “full of doubts and speculations.” He decried independent poll watch groups like AES Watch for conspiring to tarnish the credibility of the electoral system.

But which side is engaging in a conspiracy? Prof. Rene Azurin did not mince words. He warned that the uniformity of the senatorial race results pointed to the pre-programming of the automated poll system in order to ensure an astounding 9-3 victory for the ruling party.

A Conspiracy?

The Comelec’s rushed proclamation of the top six senatorial bets even as only 72 of 301 Certificates of Canvass were canvassed only aggravated these suspicions.

Former Comelec IT director Ernie del Rosario adds that the tallies seem to “follow some sort of deterministic linear equation” that points to “a pre-designed results reporting mechanism that fits the 9-3 survey instead of a tally of the actual votes.”

But Dr. Michael Purugganan of the New York University said that nothing’s fishy: if you look at local precincts “each will have a different number that is further away from the 60-30-10 pattern.”

Purugganan explains that while every canvass released by Comelec was nearly identical in percentage allocation for each party, it does not mean that votes from the candidates’ bailiwicks were disregarded by pre-packaged results.

“Each canvass released by Comelec does not reflect any particular region or province – they put together all the results from around the country that they have COCs from, and release the result as a canvass,” said Purugganan.

Same Pattern in the Grassroots

Center for People Empowerment in Governance IT consultant Pablo Manalastas points out that “the more correct indicator of a conspiracy is if we get the same 60-30-10 figures in a precinct-by-precinct comparison,” said Manalastas.

Some new reports say that the averages of the voting tallies in the country’s key cities and regions do not conform to the 60-30-10 formula.

But while the actual percentages vary from this figure, a glimpse at the results in the regions by social media users show that they still uniformly come close to the disturbing pattern.

The share of the administration party ranges from 56 percent (in the Autonomous Region of Muslim Mindanao) to 63 percent (in Bicol).

The share of UNA ranges from 28 percent (in Bicol) to 35 percent (in Cagayan Valley) while the independents range from 7 percent (in Zamboanga) to 15 percent (for the Overseas Absentee Voters).

Netizens also saw the same pattern in many provinces, towns and cities. Seeing this in all 18 towns of Bohol , Rick Bahague of the Computer Professionals Union quipped that voters may have been genetically modified: “They voted the same. The rank of the senators is almost the same in all the towns.”

No Transparency

Despite snide threats and name calling directed by Brillantes to his critics, various groups are continuing the calls for accountability for the innumerable poll irregularities and a genuine investigation of “hocus PCOS” allegations.

The AES Watch is calling for the creation of an independent fact-finding body that will probe the questionable conduct of the 2013 mid-term elections that go against the Poll Automation Law.

Fears of electronic fraud has gained currency because of the very absence of transparency in the AES, including the disregard of security safeguards like the review of the PCOS source code, prescribed digital signatures, voter verification, and the use of WORM (write-once-read many) CF cards.

Moreover, the Random Manual Audit (RMA) of selected PCOS machines, by checking only one clustered precinct per legislative district than a reliable statistical sample, seems to be conducted by the Comelec for mere convenience.

One striking incident that exemplifies the danger of this absence of checks and balances came just an hour after voting ended last election day when partial results were released by the Parish Pastoral Council for Responsible Voting (PPCRV).

With just over 1,418 precincts canvassed, senatorial race top-notcher Grace Poe already amassed more than 10 million votes: “The error is significant, as a maximum of 1,000 voters per precinct should only yield a total of around 1,418,000 maximum votes,” said Kontra Daya.

The official explanation said that it was a simple case of double counting and as a solution Smartmatic immediately had the computers reformatted to display the correct results.

But more than the wrong count, what this episode reveals is our powerlessness in independently verifying the authenticity of the election results: “That Smartmatic can change the script of the source code during the canvassing shows serious problems with the entire automated system.”

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s