BSP to start big data use next year

Published June 9, 2022, 3:45 PM

by Lee C. Chipongian

BSP to start big data use next year


Outgoing Bangko Sentral ng Pilipinas (BSP) Governor Benjamin E. Diokno said the central bank will begin using big data sources and applications in 2023 starting with the News Sentiment Index (NSI).

Currently, BSP is modernizing its information technology infrastructure in preparation for big data use.

BSP Governor Benjamin E. Diokno

“At the moment, the BSP is still in the exploratory stage in using big data. Although there are several promising used cases for big data for economic research, financial supervision and monetary policy, a well-designed IT infrastructure is necessary to maximize and utilize the value of big data,” said Diokno on Thursday, June 9, in a virtual press chat.

“So until we have the necessary big data tools in place, we can’t fully operationalize our big data applications and integrate this into our operations,” he added.

Diokno, in one of his last weekly press chats as BSP chief before assuming his new post as finance secretary on July 1, discussed the BSP’s development of its latest information index, the NSI.

The NSI is based on big data, and it is a tool not meant to replace BSP’s traditional surveys.

“The news sentiment index will complement and supplement the confidence index data from our quarterly consumer and business expectations surveys,” Diokno told reporters. “(Its) purpose is to narrow information gaps by providing us with sentiment data more frequently. This information will allow us to intervene and make timely policy decisions based on the prevailing economic environment,” he added.

The NSI is based on news from recognized and reputable media outlets in the Philippines only, said Diokno. The new BSP took will collect only financial, economic, and business news posted by media outlets.

“Of course, we must differentiate between the release of inaccurate information versus fake news. If a news outlet inadvertently publishes wrong data or information, this is not fake news. It is an error that will eventually be corrected by an erratum or updated version article by the concerned media outlet,” said Diokno.

“BSP ensures that it uses only truthful and fact-checked data in its decisions and policies,” he added.

The BSP is currently procuring the necessary systems and tools for the NSI. “This takes time,” according to BSP Senior Director, Redentor Paulo Alegre Jr. of the Department of Economic Statistics.

Alegre said the NSI “can be real time” in terms of frequency. “Depends on how accurate the sentiment indices are to our survey-based indices as the benchmark,” he said during the press chat. The NSI is however a cost-effective and efficient data gathering solution useful for monitoring real-time economic developments, and which can serve as an important input for enhancing evidence-based policy formulation.

Alegre also said the NSI will not replace BSP’s quarterly consumer and business sentiment surveys because “nothing beats first hand information.”

The NSI basically will capture relevant views on key macroeconomic events that may affect the current and emerging economic and financial environment. It will leverage on big data, machine learning and artificial intelligence.

The project involves the development of software that will automatically gather information on topics of interest, such as consumer and business sentiment, from online news sources. The BSP explained that the information gathered will, in turn, be processed using algorithms to derive the general sentiment such as positive, neutral, and negative on the economic and financial front.

“This novel way of gathering and processing information will result in a more efficient method of generating sentiment data on a real-time basis,” said Diokno.

Based on studies of other central banks and researchers, the results of the NSI’s of other countries closely reflect those of the respondent-based sentiment surveys, such as the business and consumer expectations surveys, said the BSP.

The BSP also noted studies that NSIs are “strong predictors of leading economic indicators, such as inflation, interest rates, and gross domestic product.”

The BSP’s use of big data was set up with the help of the University of the Philippines’ School of Economics.

Big data, as defined by the BSP, is characterized by high volume, velocity or variety of data that cannot be processed using conventional tools and software, and that require specific technology and analytical algorithms for its transformation into value for mission critical processes.