In Search of Credible Elections and Parties: The Philippine Paradox 53 government militiaman and paramilitary unit commander in the 1970s, Andal Ampatuan, Sr., the clan patriarch, rose to power initially through local positions as municipal officer-in-charge of his hometown during the Corazon Aquino administration and later as elected mayor. Having accumulated economic power “through the forcible and violent acquisition of land” and with his political network and the military's support, Ampatuan won the governorship of the province in 2001, getting reelected, unopposed, in 2004 and 2007. In the war against the Moro Islamic Liberation Front (MILF) whose main support base is in Maguindanao, the government treated the Ampatuans as an ally, further expanding the clan's private armed groups (Arguillas 2011: 17-41). Demonstrating its ability to defy laws with impunity, the Ampatuans orchestrated the country's worst cases of electoral fraud and manipulation in the 2004 and 2007 elections, benefitting the ruling party and Mrs. Arroyo who ran for the presidency in 2004. The pervasiveness of private armed groups in the country is documented by the findings of The Independent Commission Against Private Armies, a special government fact-finding body headed by Justice Monina Arevalo-Zeñarosa, retired associate justice of the Court of Appeals. In its 2010 report to President Arroyo, the Zeñarosa commission identified the PAGs active in each region and province and also provided estimates of loose, unlicensed firearms in each region. (See Table 2). However, the official data also underreports the number of PAGs since many of the official civilian groups armed by the military for counterinsurgency purposes often end up effectively under the control of powerful local politicians and can be activated for their private ends. Moreover, in some municipalities and provinces even the local police forces may actually function as PAGs or activated as such by powerful politicians who can buy their support or acquiescence in illegal operations through patronage networks. Only two regions (NCR and Region 11) were reported not to have PAGs but the National Capital Region had the highest concentration of loose firearms (315,127) while Region 11 also had a significant amount at 49,178. The official data on loose firearms also tend to be on the low side since it cannot take full account of firearms controlled by armed groups and movements operating outside the pale of law. One significant information that emerges from these data is that the NCR rivals the ARMM in terms of the distribution of loose firearms relative to the population, with the former accounting for one firearm per 37 persons while the latter has one firearm per 36 persons. Not surprisingly, these deadly combination of PAGs and huge numbers of loose firearms are ready-made ingredients not only for election violence but also for ordinary criminality in the face of a weak state.