NUMBERS CAN LIE. That is the lesson we learn from the well-intentioned study conducted by the National Statistical Coordination Board, comparing the electoral fate of incumbent provincial officials vis-à-vis their province?s performance on the so-called Good Governance Index.

According to NSCB secretary-general Romulo Virola, eight of the 10 ?worst-governed? provinces (measured in 2005) re-elected their governors in 2007. This ?finding? seems unobjectionable?until we compare it with the other major ?finding?: that three of the 10 ?best-governed? provinces threw out their governors in 2007.

Virola?s conclusion: ?Our voters, essentially, do not vote for good governance.? He also told the Inquirer: ?Good performance is not sufficient for the governor to win; neither is bad performance sufficient for a governor to lose.?

Since Pampanga is one of those three ?best-governed? provinces that replaced their governor three years ago, and the Lapid dynasty was the ruling clique that was replaced, we need to ask: Just what does the NSCB mean by good governance?

The answer is determined by the GGI, which is anchored on three important sets of indicators: economic (with data from the NSCB, National Statistics Office, the Commission on Audit and the Department of Budget and Management); political (with data from the Philippine National Police and the Commission on Elections); and administrative (with data from three departments: education, health and local government). The second set, in particular, bears a special relationship to our Pampanga question: those political indicators were chosen which describe improvements in law enforcement and the administration of justice and the elimination of graft and corruption. Under the Lapids, father and son, Pampanga did not lose its reputation as a vital center of the illegal numbers game jueteng. Indeed, as the election of Gov. ?Among Ed? Panlilio showed, the forces associated with the jueteng operators in the province constituted a formidable political bloc, almost equal in strength to the traditional politicians associated with the Lapids. It was precisely because these two rival political blocs were of almost equal strength that Panlilio?s alternative politics squeaked through to victory.

If Pampanga scored high on the political component of the governance index, what does that tell us about the political indicators used? Under the Lapids, collections of the lahar quarrying fees were only a tiny fraction of the fees collected in the first year of the Panlilio administration. This strongly suggests that corruption was involved, and may in fact have been systemic. If this is the case, then the political indicators the NSCB used (provided, let us remind ourselves, by the PNP and the Comelec) were gravely deficient.

Alternatively, there is also the Maguindanao question. In 2007, the patriarch of the Ampatuan clan, Andal Ampatuan, was returned to the governorship of Maguindanao, one of the 10 ?worst-governed? provinces the NSCB identified. Before we conclude that the people of that province voted for Ampatuan regardless of his performance as governor, we need to ask: Were the elections in the province, in fact, fair and free? We have seen evidence of the Ampatuans? command vote, in the 2004 presidential election and in the 2007 senatorial contest. In the wake of the Maguindanao massacre, the authorities have unearthed unused voter?s IDs and election forms; these help confirm the common public perception that, in many places in Maguindanao, elections were not held.

Thus, if election results in Maguindanao were in fact manufactured, what does that tell us about the NSCB?s indicators? Is there a statistic that reflects, say, the level of intimidation an electorate living in fear faces?

What these questions suggest is that the NSCB must fine-tune the indicators it uses to capture the true state of a given province. It must not rely on the data provided by government agencies alone, but make room for alternative sources of information (such as social surveys conducted by reputable polling organizations). Only then will it be really possible to gauge whether ?good governance? is, in fact, a non-issue in elections.