Agile Business

IFRS 9 Data Challenges and Solutions for Risk and Finance

 
April 23, 2018

The International Accounting Standards Board (IASB) has already issued the final version of International Finance Reporting Standard 9 (IFRS9) to bring together the classification and measurement of assets and liabilities, impairment of assets based on expected losses and hedge accounting. It will replace the IAS 39: Financial Instruments: Recognition and Measurement (IAS 39) for accounting periods beginning on or after January 1, 2018. Per IFRS9,  the measurement of impairment loss allowances is to be based on an expected credit loss (ECL) accounting model rather than on an incurred loss accounting model – a forward looking approach for finance. For banks however, this presents a number of challenges, specifically in terms of ECL modeling and risk and finance accounting data management.

Challenges

Accounting is a rule-based or standard-oriented process whereas risk assesment follows defined principals with some liberties. As business functions, their goals and motivations are different and subsequently their respective data infrastructure is arranged differently in the IT environment. With the introduction of IFRS9 banks will need to ensure that the data  used by risk and accounting is coherent. In a way, IFRS9 enforces the pace of risk integration with accounting, factoring in the neccesity to calculate expected credit losses for one year or lifetime, depending on the asset impairment characteristics across stages one, two, and three. This has certain implications for data used for preparing the downstream IFRS9 compliant financial statements.

Once IFRS9 standard is in place, banks will experience an increase in data volume and granularity due to the inflow of risk parameters, macro-economic factors, market, and external information required for point in time models. These are more data-intensive as compared to through the cycle (TTC) models.

Data collection, sufficiency, enrichment, provisioning, and quality needs to improve in order to tackle the challenge of data management. Regulatory  reports, which mainly comprise of quantitative data based on business rules and calculations, are as good as the data used. Therefore, uncompromising importance should be attached to data integrity.

According to the Basel Committee on Banking Supervision regulation 239 (BCBS 239) progress report, banks are also grappling with the challenges posed by the lack of robust data architecture and IT infrastructure, low accuracy and reliability of data, and weak adaptability of the existing risk data framework.

Setting the Gold Standard for Data

  • Overcoming these hurdles, and seamlessly churning out regulatory and finance reports will require coordination and harmonization between various business and functional units, data repository process, reporting process, and final submission. Business, modeling teams, and IT will need to closely collaborate and work in an agile environment. It will enable them to have meaningful discussions, take the right decisions together, and create business value for the organisation.
  • An integrated data architecture program that handles risk and finance data efficiently is the need of the hour. Such an architecture should enable validation and reconciliation of finance and risk numbers at various stages of data life-cycle, simplify data delivery process and reduce data redundancy. The data models developed for the integration program should be flexible, scalable, and consistent to address demands of different banking units such as corporate, retail, treasury, legacy infrastructure, and more.
  • Further data control processes like collecting and storing different metadata in a central repositry should be implemented to make the end-to-end process auditable as well as traceable till the originating source This will  automatically ensure better data governance.
  • Banks will also need to develop a common taxonomy due to differences between IFRS 9 and Regulations like Basel. Some of the measures for the same are:
  • A comprehensive business glossary for the attributes used in the integrated data model
  • Relationship between these attributes for dependency and traceability
  • Business and techincal rules associated with these attributes

For example, The table below explains  terms which can differ or stand in common between IFRS9 and Basel

IFRS 9 standard Basel III regulation
ECL for one year or lifetime depending on the impairment stage of the asset ECL for one year in case of Basel and 3 years in case of Stress Testing
EAD is the net present value of future cash flows EAD is the forecasted value
Different LGDs & PDs(Loss given default, Probability of default) Downturn LGDs & PDs
Probability weighted macro economic scenarios Baseline and adverse macroeconomic scenarios
Combination of standards and principles Based purely on principles

The Evolving Regulatory Landscape

Regulators today are demanding an increasing level of visibility into data managed by banks to practice tighter control.  IFRS9, along with other upcoming as well as existing regulations such as IFRS15, LCR (Liquidity coverage ratio), NSFR (Net Stable funding ratio),  Anacredit, Basel IV, etc., will force banks’ risk, finance, and IT functions to develop a common framework for regulatory and accounting functions to ensure common data language and single source of data. The operating model of these three functions should enable seamless exchange of quality data in compliance with reporting timelines and the principles and standards of IFRS 9.

 

Pratish Pusal is a Business Analyst in TCS working for Analytics & Insights unit. His areas of expertise include Risk analytics, Credit risk and CCAR/BASEL regulation. He has an MBA from NITIE, Mumbai, a BE in Electronics from SPCE, Mumbai University.