Wake Up, Philippines!

Automated election can be used to cheat

Posted in comelec, Election, Modernization by Erineus on March 16, 2009

By Manuel A. Alcuaz Jr.
Philippine Daily Inquirer
First Posted 23:17:00 03/15/2009

Filed Under: Elections, Graft & Corruption

IT IS UNFORTUNATE THAT many people, including senators and congressman, are under the impression that election automation would automatically lead to clean elections.

In an automated election, good old Garcillano might not be able to employ his old tricks. But there are countless computer whiz kids who could modify programs and alter voting results electronically.

Many ways to cheat

If we look at the election process, there are many ways to cheat.

1. People in power or with lots of money could buy support from local leaders or directly from voters.

No automated system can prevent this.

2. In the old days of guns and goons, voters were either coerced to vote for certain candidates or scared away and their ballots used.

We thought we had progressed beyond this. Unfortunately, cheating prevailed in 2007 in Maguindanao and other areas.

3. Ballot box stuffing or ballot substitution.

With the proposed Comelec Automated Election System, ballots would have to be substituted before these are fed into the counting machine.

This is a little more difficult, but the actual production of the fake marked ballots is a lot easier. And it is harder to identify ballots marked by one person.

4. Misreading/mistallying of votes during precinct count.

Normally, OMR counting machines can be very accurate. But who can tell if the machine has been programmed for automated and undetected dagdag bawas? Comelec has not done enough to assure the public that this will not happen.

5. Substitution of election returns (ER).

This may have happened in the 2004 elections in ARMM.

We suspect that in Pampanga, Cebu, Iloilo and Bohol, Namfrel and the other parties may have been given fake ERs.

An honest, automated system would prevent the substitution of ERs with previously prepared faked ERs. But we can never tell if manipulation is done inside the OMR counting machine.

6. Substitution of ballot box and ER on the way to the municipality.

In the automated system, electronic ERs would be sent to the municipal canvassing center through the communication system.

How can we be sure that the results transmitted are not changed at the source or at the receiving end?

7. Fraud in the computation of the municipal COC.

This is hard to detect if the precinct results are not visible to watchers.

In the automated system, we will not see how computations are done in the canvassing server. There is no independent means to cross check what the server generates.

I think that contrary to the common belief that delays create opportunities for cheating, some delays are needed for checking and auditing.

In an automated election, moving too fast without checks and audits could result in massive cheating.

8. Substitution of Municipal COCs on the way to the province.

This could have happened in Muslim Mindanao in 2004.

Proponents of automated systems suggest that this would be prevented with secure electronic transmission. There still is the possibility of manipulation within the system.

9. Fraud in the computation of the provincial COC.

This could have happened in 2004 and could happen again within the provincial canvassing server.

10. Substitution of COCs on the way to Congress and Comelec.

This could have happened in 2004. And even with an automated system, this could still happen.

11. Errors in computation of national total.

Counting machines

The P9.5 billion the Comelec intends to spend on the rental of 80,000 OMR reading machines will not hasten the completion of national election counting. But the use of reading machines could lessen retail cheating in peaceful areas.

However, OMR voting is not a deterrent. For cheaters, OMR voting facilitates the production of ballots.

Hazards and safeguards

Comelec would like us to assume that automation will prevent cheating.

That is not true. Let us make sure that safeguards and audits are instituted.

The OMR system is similar to the classic, paper-based election system, except that:

1. Voters mark candidate of choice instead of writing the name.

2. The OMR ballots are machine-counted instead of being read and tallied.

For those who think that cheating can only take place when human hands are involved, this would look like a fraud-free system.

Comelec’s new procedure calls for each voter to physically feed his ballot into the machine.

A picture of the ballot is then taken.

As we pointed out earlier, the voter in some areas may be influenced or forced to feed another ballot into the machine.

Programmed to cheat

Let us pretend we are in a precinct where law and order prevails, and you are the voter feeding in your ballot.

How can you be sure that the machine will not change one or more of your votes?

How can you be sure that the total votes in the printed ER are truly what the voters in the cluster voted for?

The law provides for testing of the machines prior to Election Day.

If the machines are not stand-alone, how can you be sure that a modified program was not downloaded on Election Day to add votes for certain candidates and subtract from others (electronic dagdag bawas)?

At the end of counting, the original program could be restored.

The Election Law should call for stand-alone machines.

To verify that the OMR machines are counting properly, the two parties and the Citizens Arm should be allowed to run their test ballots before the start of counting and at the end of counting.

If discrepancies are detected, these should be noted and could be the basis for reverting to a manual count or a protest.


The Comelec proposes to automatically transmit election returns from the 80,000 OMR counting machines to the municipal servers.

While this is the fastest way to do it, it does not guarantee honest elections and does not provide transparency of the election counting process.

If the OMR counting machines can send electronic ERs to the municipal servers through the communication system, someone who knows the system well could change the programs on the machines from a remote and undetected location.

The best way to detect fraud is to create and provide at least seven printed and electronic copies of the ER.

The OMR Counting Machines should not be equipped with any communication capability.

There should be a separate stand-alone PC from where the ERs can be sent to the municipal canvassing/consolidation server as well as to the seven organizations entitled to receive the seven copies of the ER.

The Comelec AES does not provide for visible canvassing or parallel transmission and canvassing.

This will raise concerns about the honesty of the count and would certainly result in a loss of credibility of the results.

The Comelec should provide PCs for the major parties in each municipal tabulation center.

There should also be at least three projectors in each canvassing center.

The projectors would show the statement of vote for the municipality.

Watchers would be able to compare the projected totals on the three computers (Comelec, majority and opposition).

The COC should not be finalized until the discrepancies are resolved.

There are 1,631 cities and municipalities, 80 provinces, 13 regions and two national canvassing centers for a total of 1,736 sets.

Let’s provide 10-percent backup sets. That would be 1,910, let’s say 2,000.

The total cost would only be P360 million.

Cost reduction

One could easily reduce the cost of the OMR Counting Machines by increasing the cluster size per OMR machine to 10 and allowing feeding of ballots into the machines by the BEI after the end of voting.

That would mean savings of at least P4.5 billion, which is more than enough to pay for a transparent and more credible transmission and canvassing system.

Hopefully, wholesale cheating could be lessened.

But let us not expect canvassing for national candidates to be done in three to four days.

(The article reflects the personal opinion of the author and does not reflect the official stand of the Management Association of the Philippines. The author is president of Systems Sciences Consult Inc. Feedback at map@globelines.com.ph. For previous articles, please visit .)