Alarm bells for May 10 elections?

Sharp spike in Maguindanao,
ARMM population a big riddle

A RECENT conference of the Philippine Population Association had none of the media frenzy that usually attends the ongoing bail hearing for ex-Datu Unsay Mayor Andal Ampatuan Jr., the prime suspect in the massacre of 57 people, including 32 media workers, in Maguindanao last November 23.

But what population experts said about demographic questions and mysteries in the Autonomous Region in Muslim Mindanao (ARMM) – topics previously discussed only in the confines of the classroom or office – could prove just as important as the gory details coming out of the criminal trial.

From the plenary discussions and interviews on the sidelines of the conference held late last week, the emerging picture on how the 2007 population census conducted in the ARMM is not pretty. Outright fraud, political pressure, and security threats marred the 2007 population census in the region, according to government statistics officials and demographers.

On top of these, some local leaders also engaged in a “balik-ARMM” campaign to encourage migrants to visit their hometowns, artificially boosting population numbers during the census, officials and experts said.

All these and more could help explain the big riddle that is the autonomous region’s excessively high annual population growth rate of 5.46 percent between 2000 and 2007, almost triple the national average of only 2.04 percent.

Overall population growth is slowing but ARMM proved to be a big exception. It expanded faster in the 2000-2007 period compared to the average of 3.75 percent between 1995 and 2000.

All other regions posted lower population growth since 2000, except Metro Manila because of high levels of in-migration from other regions.

Far from being just a matter for statisticians, soaring population numbers underscored heftier allocation of internal revenue funds for the ARMM’s regional, provincial and municipal governments.

Government investigators are now checking if some of the funds were used to build the arsenal and palatial mansions of the powerful former Maguindanao governor Andal Ampatuan and his sons, two of whom are now in jail for their alleged role in the November 2009 massacre, by far the most gruesome and most deadly case of election-related violence in years. It was also the biggest single-day murder of media workers in world history.

EXCEPTIONAL. The population growth rate of the Autonomous Region in Muslim Mindanao in 2000-2007 was higher than in 1995-2000. Except for Metro Manila, the nation’s center of commerce, education and government, all other regions posted slower average population growth rate (APGR).


More voters, too?

High population growth is also related to the rapid rise in the number of registered voters in the region, where President Gloria Macapagal Arroyo and her allies won some of the ruling party’s biggest election margins in the 2004 and 2007 polls. Coincidentally, ARMM was tagged as allegedly being the center of dagdag-bawas (add-subtract) operations of then elections commissioner Virgilio Garcillano of the ‘Hello, Garci’ scandal during the 2004 elections.

Many demographers and social scientists had been entertaining doubts and questions about the ARMM’s unusually high population growth rate between 2000 and 2007. In last week’s Philippine Population Association conference, they finally had a chance to discuss these openly and in the presence of top officials of the National Statistics Office (NSO), which is responsible for carrying out the census.

One of the speakers, Dalisay Maligalig, senior statistician at the Asian Development Bank (ADB), put forth the question that was probably foremost in each participant’s mind. “Is this statistically possible?” she asked after taking note of the wide disparity between population growths in the ARMM and the national average.

ARMM also stands out as an outlier even when compared to other areas in Mindanao, none of which had a population growth of no more than 2.5 percent a year since 2000, said the ADB statistician.

She pointed out that Region IX (Zamboanga Peninsula), which has the same population as ARMM, grew by only 1.83 percent, or just a third posted by the autonomous region. Nearby provinces also had much lower population growth rates.

The differences are starker when some ARMM provinces are compared with their neighbors. Maguindanao’s population grew by 6.99 percent a year, way above 1.5 percent in South Cotabato, 2.1 percent in Sarangani, and 2.2 percent in North Cotabato.

Much worse

In a poster presentation on using barangay-level data to assess the quality of the 2000 census, Florio Arguillas of Cornell University and Joy Arguillas of the University of the Philippines observed that one barangay in Shariff Aguak town, Maguindanao’s capital and traditional stronghold of the Ampatuan clan, grew by an average of 13.6 percent per year between 2000 and 2007. Perhaps because of the big jump in population, what was one barangay in 2000, aptly named Labu-Labu, became three barangays in 2007.

“You have to question the growth rates in ARMM. It is so high, so unlikely given the population, the economic condition of ARMM,” Florio Arguillas told the PCIJ in an interview before the conference.

NSO chief Carmelita Ericta, and one of conference’s plenary speakers, admitted there were numerous problems in doing the ARMM census in 2007.

“Actually the ARMM results have undergone internal evaluation,” she said during the open forum following her presentation. “The initial results were much worse but we had no way of going back and verifying for ourselves because of security concerns.”

Talking to the PCIJ after the conference, she said the statistics agency had to weed out questionable data as part of a careful review of initial results that suggested population in the region grew by about 10 percent each year since 2000.

TEEMING WITH NEW VOTERS. The age-sex distribution patterns in the entire Philippines do not compare with that of Maguindanao province and its capital, Shariff Aguak town, based on the 2000 Population Census. There are more 18-year-olds (see red arrow), the voting age for Filipinos, than any younger age-group in Maguindanao (except 10 year olds) and in Shariff Aguak.


Fraud, pressure, threat

Census-taking is not easy, said Ericta. As it is, affluent households suspect that census data will be used for tax investigations while informal settlers believe it is a prelude to eventual demolition of their communities.

But, Ericta said, these difficulties are compounded by instances of outright fraud, political pressure, and even threats against the security of census-takers in the ARMM.

She said the easiest to deal with were the cases where the same names – usually of famous movie stars – were written down as those of members of unusually big households. “We just usually cross-out the data,” she said. She added that big jumps in average household size compared with previous census data were another indicator of possible attempts to manipulate the census results.

A little bit harder to tackle were cases of pressure coming from powerful local leaders who are keen to see populations soar in order to seek bigger revenue allotments, the creation of new towns or congressional districts, and more votes to command during elections. Mayors chair the local census coordination board, which is supposed to sign on to the results of the census in the towns and cities.

The NSO administrator said the problem was “more apparent in the ARMM than in other areas.”

“Some government officials think this is an exchange. They provided t-shirts or snacks or equipment so census-takers should accommodate what they want. There is a thin line between coordination and influence,” Ericta said.

The NSO, she explained, tries to address local government concerns that the population was being undercounted in their territory by holding saturation drives to recount the number of people through sample survey. She said these kinds of complaints from local leaders are common.

Local execs’ demand

“Even at my level, there are attempts,” to press for higher population numbers, Ericta said. “Mayors, governors, senators (demand it) because it affects legislation, especially if they want to create (new territories).”

Though no census-taker was killed for refusing to accommodate local leaders’ requests, some were subjected to threats, she said. “In some cases, they were refused entry,” into a community, she said. “In some cases, they (local leaders) want to influence the results. That is part of the risk we take in census-taking.”

The NSO has also heard about cases where local officials in the ARMM actively encouraged former residents to come home for a visit to temporarily boost population numbers during census-taking, Ericta said.

“It’s not the only place where they do that,” she said. But she noted that officials in ARMM “have a tendency to call their people and ask them to be there for the census.”

Back to Caesar?

Mercedes Concepcion, one of the country’s foremost demographers and founder of the UP Population Institute, said she also heard about the “balik-ARMM” campaign of local officials before the census. She commented: “This meant we are going back to the first census ever taken, and that was the time of the first Caesar Augustus when Joseph and Mary went back to be enumerated.”

The NSO can undertake post-enumeration surveys (PES) to check the accuracy of the census but statistics officials said conducting such a survey in ARMM was nearly impossible because of security problems. None was undertaken in spite of the mounting questions and doubts about the quality of census data in the region.

Socorro Abejo, the NSO’s chief demographer, told the conference: “In the 1990 post-enumeration survey, we were able to penetrate all areas. I went to the ARMM region, I was able to visit all the barangays sampled. The difference now is that in 2007 there was this security concern. There was difficulty in penetrating barangays which were away from Cotabato City.”

She appealed for understanding about the limits of what the NSO can do amid possible dangers to the lives of the census-takers. “The indirect evaluation techniques that we can do in our office would tell us there is something wrong but we should also be aware that if we will venture in conducting post enumeration survey there is this concern, the fear that our procedure will not work in this region because of this security concern.”

‘Remove faulty data’

In an interview at the sidelines of the conference, Concepcion said, “I have always maintained that the ARMM (census) data are so faulty they should be removed entirely from the population totals.”

But NSO officials said they are not allowed by the law to exclude census data for a locality even if they themselves still have some questions about accuracy. “After data cleaning and evaluation procedures, we had to stop at some time,” and release the figures, Ericta said. In the case of ARMM, she said, “by the time we stopped, it was down to just five percent,” instead of about 10 percent initially.

She said the NSO had wanted to take and use aerial photographs to estimate population in the ARMM but “we couldn’t do that either because of security considerations.”

Biggest IRA growth

The real-world implications of all these quibbling over numbers is serious. Indeed, when the 2007 population census data were first applied to compute local governments’ internal revenue allotments (IRA) from the national government in 2008, the allocation for ARMM soared by 27.3 percent to P8.7 billion from P6.8 billion the previous year, according to the Department of Budget and Management.

The autonomous region posted the biggest percentage jump in IRA. The revenue allocation for the rest of the regions rose by an average of only 13.8 percent.

This is hardly surprising because population accounts for half the weights of the criteria used to calculate a local government’s IRA. Land area has a weight of only 25 percent.

Similarly, ARMM posted the biggest increase in the number of registered voters between 2002 and 2007, rising by 26.2 percent compared with the national average growth of only 18.8 percent, according to data from the Commission on Elections.

In Maguindanao, there were more 18-year-olds, the age at which one can vote in the Philippines, than any other age group below 18, except 10, according to the poster presentation by Florio and Joy Arguillas, citing results from the 2000 census. The study suggests that data quality problems beset the census not just in 2007 but also the one in 2000.

“There is one barangay (where) you can’t find children under two years (old). So two years before the census, they stopped producing children?” asked Florio Arguillas.

For good or bad, census and elections – perhaps because both involve counting people – seem to be so intrinsically linked in the Philippines.

In fact, on May 17, 2010, just a week after the May 10 elections, the government is scheduled to roll out the next population census across the nation. – PCIJ, February 2010