MANILA, Philippines—After last Monday’s Philippine Daily Inquirer editorial, I heard that an Indonesian diplomat told his Filipino counterpart, “Blame the global economy — with the price of oil so favorable to us now, what you paid us last year to do worse in the Corruption Index isn’t enough anymore.” (Joke only.)

The editorial, “Growing in corruption,” was about the Philippines, at 2.3, scoring below Indonesia (2.6) for the first time on the Corruption Perceptions Index of Transparency International. Incidentally, Singapore was at 9.2, Malaysia at 5.1, and Thailand at 3.5.

But what is the significance of a global index for a country’s objective of reducing corruption? Geographical data only have geographical implications, i.e., try to be like Singapore, and not like Indonesia or the Philippines.

Actually, global indicators are mainly for audiences like international investors, multilateral banks and foreign aid agencies, whose mandates cut across the globe. Doing better on the global corruption contest can lead to a higher Philippine share of the funds of these institutions.

Yet, reducing corruption in the Philippines is good in itself, regardless of what happens to Philippine standing relative to Indonesia on the global score, and regardless of our success in attracting foreign funds. If we (and I don’t mean only the government) are serious in wanting to reduce it, then we should apply science, and hence measurement, to the task. We must adopt means for measuring progress (or regress) in reducing corruption.

In the first place, we must reject the notion that it is impossible to measure corruption. Centuries ago, people thought that heat was not measurable, being only “a feeling.” Though things illegal or immoral tend to be hidden, and are more difficult to measure, yet such challenges to the research profession are not impossible to meet. (For example, we have been able to survey such matters as using prohibited drugs, participating in theft, forcing another person to have sex, and being forced into having sex.)

Obviously, surveys also measure personal, or up-close, experiences of corruption, aside from perceptions, or impressions about it, from afar.

National surveys on corruption are being done in many countries, quite apart from those used for the highly publicized global comparisons. In 2005, Ghana’s “Voice of the People” survey got ordinary people’s perceptions of corruption and bribery, to provide data for policy reforms. In 2006, the Kenyan government did a survey of public perceptions and experiences with key governance and legal institutions, such as the anti-corruption commission and the police. In 2005, in Uganda, corruption was part of the agenda of an urban Cost of Doing Business Survey done by the Ugandan Manufacturers’ Association. In the Philippines, Social Weather Stations (SWS) included the subject of corruption in seven surveys of enterprises for a Transparent Accountable Governance project in major metropolitan areas in 2000-2007; the eighth round is underway this year.

Subjective indicators are not inferior to objective ones, if the scoring system and manner of selecting the respondents or evaluators are valid. Sometimes the entire electorate is the best judge — see the quarterly SWS ratings of public satisfaction with the performance of the national administration in fighting corruption. In the global measurement systems with only a few judges, I think the judges, or at least their nationalities, should be identified, just as the judges in a boxing match are named. (I have been the Global Integrity scorer for the Philippines for the past three years.)

Since the government has the primary duty of curbing corruption, it has equal responsibility for applying the scientific method to it. It should not complain about “too much” data. If it finds research of others inapplicable, it should commission research to its own specifications. It should fund respectable academic institutions and thus buy into the measurement process itself.

But if the government is too slow, then private research institutions should start anyway. My suggestion is to focus on only a few individual data ingredients at first, as long as they are well conceived, understandable, and defensible. Indexation, which means a summarization or combination of the ingredients, is secondary.

What matters most is to measure corruption repeatedly over time, so that the effects of action programs can be constantly analyzed. Program instruments used at one point should be related to outcomes at a later point. The measurement should be valid in concept yet simple in design so as to be affordable to replicate at least once a year.

One should be aware of foreign methods of measuring corruption, but not slavish to any of them. It is the Filipinos who are the stakeholders in the battle against corruption, and whose concerns should be met, rather than those of the global institutions. Don’t worry, for instance, about foreign insistence that corruption measurement be gender-sensitive, unless the Filipino people insist on this feature in anti-corruption programs.

Do corruption measurement locally even if your system isn’t being done nationally. A city shouldn’t be afraid of being the first to do it.

Do measurement nationally even if your system isn’t being done internationally. Why be afraid of being the first country to do it?

* * *

Reference: UNDP and Global Integrity, “A Users’ Guide To Measuring Corruption,” September 2008, www.undp.org/oslocentre. Contact UNDP Oslo Governance Centre, oslo.governance.centre@undp.org.

* * *

Contact SWS: www.sws.org.ph or mahar.mangahas@sws.org.ph