UPDATE: AxiomSL Drives Industry Collaboration On Open Taxonomy for MAS 610

Editor’s Note: This article has been updated to include reflections from BR-AG CEO Michał Piechocki and AxiomSL’s Abraham Teo about the role of collaborating banks and why the taxonomy will incorporate XBRL. 

In order to establish the taxonomy, the regulatory reporting and risk management solutions provider is partnering with two domestic systemically important institutions (D-Sibs), seven international banks, including four global systemically important institutions (G-Sibs), advisory firm PwC Singapore, and data consulting firm BR-AG.

“The taxonomy will describe what the MAS is asking for and is independent of how banks meet MAS requirements,” says Peter Tierney, CEO of Asia-Pacific at AxiomSL. “It’s not dictated by the reporting platform.”

The group aims to define the taxonomy for the revised MAS 610, which regulates the submission of statistics and returns by banks, by the middle of this year. The joint initiative seeks to reduce reporting complexities and the compliance burden financial institutions currently face. In addition to standardizing interpretations of the MAS requirements, the taxonomy will allow for increased automation of testing and change management, as well as driving higher-quality data and data governance.

Tierney says that once it is established, the taxonomy will be made available to other banks in Singapore for a nominal fee, which will go towards maintaining the taxonomy. There will be no restrictions on who can use it, he adds.

Data-point Deluge

The revised MAS 610 includes a new set of reporting requirements covering 340,000 data points in more than 200 pages of forms—a substantial increase from the 4,000 or so data points required of banks today. This presents an onerous task for banks of any size, Tierney says.

“We got together with a group of forward-thinking banks to collectively do some smart analysis on the data that MAS was asking for. We quickly realized that we can come up with a set of relationships—a taxonomy—between those data points, which simplifies the reporting task for the banks,” he says.

Not all of the data points the MAS is asking for are unique—in some cases, they could be a permutation of a data set. “For example, if MAS is asking for the total of outstanding home loans extended to non-Singapore permanent residents for public housing, if you break that down, you can organize your data [according] to type of client, nationality of client, type of loan product, and so on,” Tierney says. “We are painstakingly going through all 340,000 data points to allow banks to generate these data points by slicing and dicing the underlying data. It’s a simple concept, but it’s hard for a single bank to analyze it themselves.”

The new MAS 610 mandate also requires the breakdown of data by counterparty type multiple times. Using the taxonomy, banks would be able to map this counterparty dimension once and re-use it, so when there is a change to the mapping, the banks would just need to update it in one place and all the returns would be updated.

Abraham Teo, global head of tax products and head of product management for Asia-Pacific at AxiomSL, says the difference between tier 1, 2 and 3 banks will be the volume of data and the complexity of processing required. “For example, a big G-Sib might require consolidation of data from 40 or 50 systems, whereas a smaller player might only rely on two or three. However, on the flip side, all banks face the same mammoth task of having to collect 340,000 data points for the new MAS 610,” he says.

Tierney adds that by collaborating, the founding banks will reduce the effort of compliance across all banks in the industry. PwC will provide advisory services and act as the project management office for the initiative. BR-AG, which has previously worked on a similar model in Europe, will draft the data model and XBRL taxonomy.

Teo says the banks will mainly be involved in reviewing and validating the taxonomy that is produced. “AxiomSL and PwC will work with BR-AG to produce a draft of the taxonomy, which the banks can then review and provide feedback. We hope to leverage the rich knowledge, expertise, as well as synergies, between the nine founding banks to refine the taxonomy,” he says.

XBRL’s Flexibility

Michał Piechocki, CEO at BR-AG, says that the data modeling and taxonomy-based approach, specifically using XBRL, has proven to be effective, particularly in the context of changing financial environments and new reporting mandates. “We have seen its merits firsthand in similar projects developed by BR-AG with the data modeling methodology (Data Point Model) at, for example the Bank of England (BoE), the European Central Bank (ECB), European Insurance and Occupational Pensions Authority (Eiopa) and the European Banking Authority (EBA),” he adds.

Piechocki says that since 2005, financial regulators such as the Australian Prudential Regulatory Authority, Bank Negara Malaysia, the Japanese Financial Services Authority, and the European Banking Authority have been among several that have implemented the open international XBRL standard. Following that, many of them have also adopted the corresponding open international Data Point Model methodology.

“The aim is to build understanding of what is required to be reported and then to take full advantage of improved data quality based on common electronic format [XBRL],” he says.

In its projects with the EBA, Eiopa, the ECB and the BoE, BR-AG applied the Data Point Model methodology, which is the same used for the MAS forms in Singapore. The Data Point Model is an open data-centric methodology that focuses on the consistent and explicit description of every piece of information required.

The description is based on a single dictionary of business concepts that is used across time and reporting frameworks. The result is a data model built on respective regulations, forms and market practices, and supportive of data exchange and usage.

The first step in these projects is to establish a clear understanding of all underlying regulations as well as the internal and external purposes of the data model and related XBRL taxonomy. The challenge is then keeping up with evolving project requirements and aligning the taxonomy accordingly.

Piechocki says BR-AG has, over the years, worked out ways to address all these key issues. This includes using the Data Point Model’s capacity to add and incorporate new frameworks.

SaaS Benefits

In addition to delivering an on-premise deployment, AxiomSL can offer the MAS 610 regulatory data management solution through a Software-as-a-Service (SaaS) delivery option.

Tierney says the SaaS method is a cloud-based approach aimed at the middle of the market, which will help smaller banks with modest local finance and IT capabilities that may be currently generating reports in a spreadsheet. “When you have more than 300,000 data fields to fill in monthly and quarterly, and you have to be ready for all the supervisory questions that MAS may come back with afterwards, the spreadsheets just don’t cut it,” he adds.

Tierney says the SaaS solution will be especially useful for international banks that only have a small outpost in Singapore, typically with a small finance and IT team, and where there is insufficient local capability to deal with Singapore-specific regulation. It will also enable banks to benefit from enhanced economies of scale, information consistency and operational transparency, he says.

As regulators demand more in terms of reporting of data from financial institutions, Teo feels that technology is key in helping the institutions cope with not only local regulatory requirements, but also with requirements in other jurisdictions. “The fact is, the majority of MAS returns are being prepared using Excel spreadsheets today, and with the MAS’ drive towards automation, the introduction of software to assist with the preparation of the returns is inevitable,” he says.

But this requires investment, which could be an issue for smaller banks that rely on their head office for this, particularly for an on-site deployment.

Tierney explains, for instance, that to deploy a solution, a bank might need to spin up three or four servers, and maintain a production support team, a change-management team, and an infrastructure team, as well as build and maintain internal know-how on how to run a solution.

“With the SaaS model it is all taken care of by the vendor, and the bank focuses more on the analysis and understanding of the data being processed and the forms being produced,” he says.