Gloria Macapagal-Arroyo

President of the Republic of the Philippines 2001 - 2010

Automated Elections

Elections, referenda, and polls are all critical processes for the fitting operation of a modern democracy, not only providing the means of transfer of power from citizens to their representatives, but also supporting citizen trust and confidence in government and democracy. This critical role requires these processes to function as designed. The integrity of the election process is fundamental to the integrity of democracy itself. The election system must sufficiently withstand a variety of fraudulent behaviors, and must be transparent and comprehensible enough for voters and candidates to readily accept the results of the process.

The design of a "good" voting system, whether electronic or one utilizing traditional paper ballots or mechanical devices, must satisfy a number of sometimes competing criteria. These include the preservation of the anonymity of a voter's ballot, primarily to protect the voter from any adverse consequences of his voting choices, and tamper-resistance of the voting system, in order to thwart a wide range of possible attacks such as ballot stuffing and incorrect tallying by insiders.

Today, information technologies offer governments and citizens unprecedented access to tools and processes that can serve to strengthen the exercise of democracy, if built upon a robust foundation for implementation, monitoring and security. The information technology revolution has affected election management in a number of ways: electoral authorities use computer systems to make their internal management and communications more effective, to systematize voter registration records, and to communicate with voters, among other tasks. In recent years, computerized voting has become prevalent worldwide, expanding rapidly in the United States since the 2000 elections. India, the world's largest democracy with 660 million registered voters, moved to full electronic voting in 2004.

Automated voting systems have increasingly been implemented globally in order to make vote counting and results tabulation faster and more accurate, because the use of hand-counted paper ballots has proven to be time-consuming, costly, error and manipulation-prone. Moreover, voter confusion can lead to effective disenfranchisement, especially of vulnerable voters (such as illiterate or elderly voters). Electronic systems for voting promise to reduce such figures by making spoiled ballots impossible. It is via this line of reasoning that technology can be seen to minimize "lost" votes in a variety of ways.

The Historic Automated Elections of May 10

The conduct of elections in the Philippines for the past four decades has remained largely unchanged. Elections have relied heavily on manual tallying and canvassing of votes, making them vulnerable to control and manipulation by politicians and others with vested interests. The cost of winning an elective post is highly expensive, and the absence of mechanisms to check and limit sources of campaign funds has proven to be fertile ground for corruption and divisiveness, shaking the substratum of national harmony. Electronic voting technologies, on the other hand, possess a greater capacity to combat common electoral fraud problems, such as capturing polling places, stealing ballot boxes, and erasure of votes, to name just a few troubles we have all too often experienced.

Popular perception of these problems inevitably led to a consensus on the need for substantive change in the electoral system in the Philippines. But right from the start, various controversies immediately swirled around the automated system being introduced into the country.

Such controversies were unavoidable, not being limited to within this nation's boundaries. In many other countries, automated election systems have been and remain under intense scrutiny by policy makers, social scientists, computer and network engineers, and activist groups--and rightly so. Issues on reliability, user-friendliness, and essential legal and constitutional requirements remain pervasive, and the discussions have not reached an end. Any computer program can have undetected, unintentional error and can also be manipulated in a malicious manner. True, various measures can help to reduce an electronic system's vulnerability, including computer security, physical security, testing and analysis of systems and coding, and good electoral procedures. But none of these steps--and no combination of these steps--can completely eradicate the irreducible, immutable vulnerability inherent in computer systems.

The critical argument in favor of election automation starts with the premise that great knowledge and expertise is required in order to maneuver and threaten the security and reliability of such systems. By comparison, manual electoral procedures and processes require lesser skills and are thus more vulnerable to control and fraud. Convinced by this argument, Government went ahead and staunchly pursued the automation specifically of the counting and canvassing processes of the election system.

Up until mere days before the recently concluded automated elections of May 2010, many issues remained unresolved and the actual implementation of the momentous exercise was still hanging in the balance. Protests abounded regarding the actual cost of automation and the apprehension that the voting public would be overwhelmed by the new system of casting ballots, due to the absence of a long-running, comprehensive public service educational program to prepare the citizens. The security and reliability of the PCOS machines and the transmission and tallying process itself was constantly questioned. Critics continued to point out issues that had been percolating for months, among them:

How secure is the automated electoral process? How susceptible is automated vote-counting to tampering and rigging? What are the contingency plans for extraordinary events, such as extended power interruptions or critical malfunctions of the PCOS machines? Finally, given these concerns, how credible would the final results of the first-ever automated elections be?

That the PCOS machines did not perform well during mock polls and tests only served to aggravate the naysayers' fears and pessimistic predictions of chaos and complete and utter failure. With less than a week left before election day, Comelec and Smartmatic discovered the reason behind the PCOS machines' vote-counting errors – flawed machine instructions which did not match the printed layout of the ballots – and had to remedy the situation by replacing all 76,000 memory cards. Though it was a relief that no malicious software or actual tampering was found to be the cause, the solution to the problem involved a massive effort that was conducted and accomplished with barely enough time to spare. The 11th hour testing and problem-solving was predictably met with harsh criticism, scrutiny and skepticism.

On the day of the elections, three major problems directly related to the modernization of the electoral process cropped up time and again: the PCOS machines' varied problems and malfunctions; the lack of voters' awareness of the actual ballot casting process; and the long lines and delays caused by the clustering of precincts due to the limited number of PCOS machines available. Other developments on election day which affected the polling process included mishaps and misunderstandings in voter registration and precinct finding, with many citizens finding out only on the day itself that they had been disenfranchised, deactivated, or that they had gone to the wrong precinct. Civil unrest and loss of power plagued scattered parts of the Philippines, which caused some areas to ask for a declaration of a failure of elections.

All three problems with the automation process, as well as other problems outside of the modernization taking place, were comprehensively taken into account in the formulation of various contingency plans.

First of all, more than 5,500 backup PCOS machines were made available in case machines already operating in precincts could not be repaired and needed to be replaced. More than 50,000 PCOS technicians were fielded under more than 6,500 PCOS supervisors, all of them trained to handle mechanical problems relating to machine breakdowns, power interruptions, system interruptions, telecommunications failures, and weather disturbances. And if for some reason a PCOS machine should malfunction beyond repair and not be replaced, manual voting was still possible. The actual physical ballots could still be collected and secured until they were brought to a working PCOS machine.

As for the delays due to precinct clustering, the Comelec released a statement less than five hours into the polling period that an extension of one hour would be granted, in effect closing the polls at 7:00 in the evening instead of 6:00. Voters' awareness about the ballot casting process and precinct finding was heightened through media advertisements, public service announcements, helpful websites and hotlines. All of these contributed to a veritable barrage of information in the last few days before May 10, culminating on the day itself. The media played a big part in the education of the masses, as ballot casting no-no's (such as smudging the ink, or handling the ballot with dirty fingers) were repeatedly discussed throughout the live coverage on several TV networks and radio stations. Although Comelec's website went down due to the surge in online traffic on Election Day, hotlines were made public and the information was disseminated through the media again. Lastly, any other type of election contingencies were met with an increased amount of security from the AFP and PNP.

A Pivotal Moment in Philippine History

What are the key lessons that were confirmed by the qualified success of the Philippines' first automated elections?

The most important insight is that citizen participation is the driving force behind the integrity and success of any democratic system. Electronic, automated voting is simply a means to make voting more convenient to the average citizen, thus inducing increased participation. The most critical element of the successful adoption of any electoral reform system is broad support from the public and from political actors. Voting technologies must be a reaction to a widely perceived need, one that has long been recognized in this country.

On the technical side, electronic systems must meet detailed, unambiguous and superior standards with regards to security, privacy, etc. Voting technology must be able to respond to whatever range of electoral procedures and systems of representation are presented; they must be robust to the physical environment in which they will operate; and they must be user-friendly to the intended voters. In addition, they must be rigorously tested and certified. These requirements can be more difficult than they appear.

On balance, however, and despite some problematic PCOS machines – which acted up in varying degrees, from suffering simple paper jams, to overheating, to flat-out refusing to accept ballots – making headlines every hour during election day, the automated elections have come to be widely regarded as successful. The doomsday predictions of massive failure did not come to pass, and contingency plans did work to keep mishaps and failures to a minimum.

Yes, there were incidents of spoiled ballots. Yes, there were areas where a failure of elections might have been declared. The automated system was not perfect. But nobody is seriously denying that the recently concluded elections were the most peaceful and orderly one in years – a fact that is backed up by AFP and PNP officials who stood watch and kept Filipino voters safe during the whole of May 10. The election returns were also easier, quicker, and safer to count, making it possible to declare winners in a more expedient manner than ever before. The Government sacrificed much and suffered intense criticism in order to bring this innovation to fruition, and the results did not disappoint. Truly, the future of digital democracy in the country has already begun.