Good example of using Big Data

In this age of the internet and powerful use of technology to crunch numbers, the concept of Big Data has huge opportunities to help in better decision making. In this particular case, a study is being made on elected officials spending habits (from budgets provided for their elected positions) and how they affect voting results is a good example. It will be interesting to see the results of this study. More importantly, I would like to know if there is any direct co-relation. I initially suspect most elected officials spend public money without much strategic planning in mind. Let’s see.

From BusinessWorld Philippines

Governors, mayors, dynasts, reelectionists, and last-termers: Local elections and competitive challenge
CROSSROADS (Toward Philippine Economic and Social Progress) By Gerardo P. Sicat (The Philippine Star) Updated October 10, 2012


The Philippine electoral season has just started. All candidates are trooping to the Commission on Elections to file their candidacies for office. It is not only a national election for mid-term senators of the nation. All the positions for local office, from that of governor and mayors to local council members, are up for electoral choice.

“Political incumbents in the face of elections.” When governors and mayors assume their electoral posts and become incumbents, they take decisions that influence the course of local history in their community. Local candidates come in different states of preparedness for office: educational attainment, experience, quality of background, affiliations, and so on. They also have different reasons in running for office.

What motivates candidates when they assume office as incumbents? Politicians have their own selfish goals. But they also assume leadership roles as local leaders challenged by circumstances that surround them. In the course of their public service they would be faced with situations in which they weigh when to advance their selfish goals and when to move their community (or their constituents) forward.

Do they have a metric against which they measure their performance? If not, how can they be motivated to benchmark against performance criteria ? Does it make a difference if the incumbent is a first termer or one who could seek future reelection or one with no more possibility of immediate re-election because of term limits? (The Constitution does not permit reelection for local elective officials after three consecutive terms.) Does it matter if the incumbent is a member of a political dynasty whether or not he can seek reelection?

It is possible to study how incumbents behave in office in relation to the electoral process. Like all executives, local leaders in elective position achieve their goals in relation to the amount of resources that they can control. Their budgets come from fiscal transfers received from the national government and from the amount of local taxes that they can raise.

They make decision choices when they spend these resources for the community‘s needs. Will they spend for development and social services which would benefit the constituents more directly or would they allocate more of the budget for projects that build their prestige visibly like enhancing public buildings and community overheads and other directly self-glorifying activities?

“A study of local government incumbents and their spending decisions.” One such study – “Perks and provisions: Effects of yardstick competition on local government fiscal behavior in the Philippines” – seeks to find answers to these intricate questions. The authors of the study belong to a younger team of economists at the UP School of Economics — Joseph Capuno, Stella Quimbo, Aleli Kraft, Carlos Antonio Tan Jr. and Vigile Marie Fabella.

The study is not one for the layman as yet. It is written for specialists because it uses complex statistical methods and intricate elucidation of economic and political behavior. But this study has serious relevance for public policy in the understanding of the behavior of political incumbents. For this reason, the ordinary citizen will find the findings of great value.

The study asserts that mayors of cities and municipalities look at the average spending and revenue raising activities of immediately neighboring jurisdictions when they make decisions on where, how much and what to spend the locally available resources. They propose these as metrics for performance and competition. Such spending patterns help determine their electability as voters make their choices during the local elections.

The study finds that local incumbents who are eligible for re-election are sensitive to yardstick competition from neighboring local jurisdictions. In short, they learn from the performance of neighboring communities and exert efforts to excel them. However, in the case of incumbents who face a definite term limit – that is, they are prohibited from reelection after the third term – yardstick competition has no influence.

In the case of local incumbents who are members of political dynasties, there is also no significant relationship between their performance and yardstick competition. In theory, a political dynasty takes a long term view of the electoral process and would find that performance standards from surrounding communities could serve as an incentive for even surpassing the performance benchmark. In practice, however, the concentration of political power might make that yardstick competition of little importance.

“Statistical basis of the study.” The study was taken from data of 1,513 municipalities and cities in the Philippines during three local electoral periods – 2001, 2004, and 2007. The three-year data set was pooled to include further information about the classes of local governments as to income and about the elected local officials together with demographics such as family relations. To these were associated the local expenditures, national tax transfers to the community and local revenue generation.

The figures on local fiscal budgets and central government tax transfers were taken from the Bureau of Local Government and Finance and the Commission on Audit. Price and economic data were from the National Statistical Coordination Board and from the National Statistics Office.

Family lineage of political candidates was searched from the Philippine Health Insurance Corporation while election and other political background were taken from the Commission on Elections. The resulting panel data constituted a huge data set, permitting the use of an intricate statistical and economic modeling.

The study covered mayors of municipalities and cities nationwide. Although governors were not included, as the highest local political officers of the province, perhaps their typical behavior has direct association with that of lower local officials. This explains why the title of this essay took liberties to include them.

“Why are economists studying political behavior?” It is fair to ask why economists are studying political behavior. One reason is that politicians make decisions which have economic outcomes. Another reason is that economic studies have embraced the use of statistical and other hard analytical tools of study ahead of some traditional social sciences.

In the example of the study just reported, the essential nature of the problem of leadership performance is associated with fiscal developments at the community level. The problem is a representative economic problem: the actions of leaders involve the choice of types of public expenditures balanced against the limited revenues that are available from the central government or from local taxes.

It would in fact be possible to change the performance benchmarks – perhaps specific types of public spending against others – and perhaps discover that the findings are altered. Such an effort has to be validated against the massive data basis of the study. That is something that traditional studies of political behavior often lack.

My email is: Visit this site for more information, feedback and commentary:


Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: