ECL models developed before the pandemic based on information in relatively calm situations were unlikely to capture escalating credit risks in a timely manner. For above reasons, the provisioning amount of these models suddenly seems underestimated. Statistical models built on sunny days suddenly proved useless.
Worse, these funding models also contained forward-looking elements. Macroeconomic variables such as GDP growth, unemployment rate, consumer price index (CPI) or energy price with a designated level of portfolio relationships were incorporated into these models to obtain the provisioning amount as accurate as possible. This characteristic suddenly becomes a liability since the variables have deviated greatly from reasonable levels. Almost every country posted negative GDP growth, unemployment soared, the price of crude oil plummeted and the heavily commodity-weighted CPI hit multi-year lows.
As the world moved into uncharted territory, financial institutions were hastily hoarding provisions to hedge against the worst possible scenarios. Leaders of financial institutions developed very critical overlays based on pessimistic assumptions to calculate the additional provisioning immediately needed for their institutions, and the amounts of provisioning, including these overlays, were reaching exorbitant levels that ultimately caused their profits to plummet.
This has prompted central banks around the world to aggressively ease monetary policies, partnering with governments to stimulate the economy to provide a lifeline for businesses and individuals and the engine of business has come to a head. re-ignited.
Unfortunately, we are unlikely to see a drop in provisioning anytime soon as financial institutions worry about the second-order impact of the pandemic and the emergence of geopolitical tensions that are disrupting supply chains and increase global inflation. They are reluctant to unravel the overlays put in place during the pandemic as uncertainty looms. Governments have been running budget deficits for years and are unlikely to continue to provide support, leaving pandemic-survived corporate and personal borrowers fully exposed to unprecedented inflation which, in turn, is affecting their ability to repay.
Some lending institutions are now seeing a direct impact as people disburse their available lines of credit while reducing repayments. This is the predictor of potential defaults as the financial situation at the individual level deteriorates, closing the possibilities for immediate release of provisioning.
There are also ECL model risks that cause financial institutions to believe that it is still premature to release provisioning and overlays, as the parameters used in the model development process have mostly been frozen during the pandemic and need to be recalibrated to reflect the most recent situations. Forward-looking elements are again a cause of deviations in the performance of the ECL model. Inflation-related factors such as the CPI or commodity prices are likely to remain highly volatile as market participants balance current supply chain situations with a potential recession due to inflation-driven inflation. costs, which distorts the provisioning results. We expect that once normality is achieved, ECL patterns will be rectified and management overlays removed.
The purpose of provisioning models and overlays is to ensure that financial institutions set the right amount of cushion to safely manage business while allowing financial statement audiences to assess the true health of portfolios, it is so in everyone’s interest to keep the accuracy of these models a top priority, but for now be prepared for a high level of provisioning and a level of suppressed profit.
By Sarun Boonchalakulkosol
Director I Risk Advisory – Financial Industry Risk & Regulatory