Menu
Banking Exchange Magazine Logo
Menu

Pandemic challenges bank stewardship

Scenario impact analysis brings new utility but challenges existing stress test programs. Here are four keys to a modern scenario-based stress testing infrastructure.

  • |
  • Written by  John Voigt, Principal Business Solutions Manager at SAS
  • |
  • Comments:   DISQUS_COMMENTS
Pandemic challenges bank stewardship

The COVID-19 outbreak has brought extraordinary disruption to communities and economies across the globe. The novel coronavirus, alongside preventative actions taken to stop its spread, has stalled economic production and wreaked havoc on supply chains and traditional working arrangements.

As the world struggles through the pandemic and prepares for recovery, financial institutions face myriad operational and financial challenges, including:

  • The massive shutdown of regional economies, resulting in unemployment spikes that have elevated both retail and corporate credit risks and fraud risks.
  • Commodity price shocks and their resulting impacts on credit demand and default risk.
  • Unprecedented levels of fiscal and monetary interventions – implemented at breakneck speed, with unknown effectiveness and unclear long-term consequences.
  • The sudden closure of offices, creating a distributed workforce and seismic disruption to internal operations.

The undetermined trajectory of the virus and timing of potential treatments or a vaccine has created significant forecast incertitude. Moreover, the public health crisis has brought volatility and unsurety to global economies and the credit cycle, forcing financial institutions to continually adapt to the ever-changing environment. Avoiding missteps relies on well-informed decisions – dependent largely on timely analysis.

Though years of regulatory stress tests have improved risk management practices, many banks now find their stress testing programs too slow and inflexible to keep pace in a real-world crisis. However, banks that can meet the new demands on their stress test systems, processes and people are better positioned to respond with confidence in the face of uncertainty.

Organizations must examine and answer the following questions:

  • Is the capital plan robust over a wide range of possible scenarios?
  • What are the risks to liquidity?
  • What is the firm’s exposure to at-risk concentrations, and what can be done to mitigate losses?
  • What level of provisioning may be needed?
  • How should portfolio growth strategies be adjusted?

Detect and avoid roadblocks with scenario impact analysis

Since the financial crisis of 2008, stress testing has become widespread for supervisory purposes, and many financial institutions have incorporated these regulatory exercises into their business-as-usual. It seems a natural progression, then, to utilize similar scenario analyses to support strategic decisioning during today’s crisis.

Importantly, strategic use of scenario impact analysis is a significantly different use case than that of regulatory stress testing. First, it requires much faster turnarounds – i.e., hours/days vs. weeks – with timeliness more important than absolute precision. It also requires more frequent analysis over a much wider range of possible scenarios.

And, due to the uniqueness of the crisis, analytical processes will need to be adjusted. Additional levels of expert judgment may be required on inputs and outputs. Alternative modeling approaches may be needed where traditional models falter. Rather than relying on standardized disclosures, the outputs must be tailorable to focus on specific concerns.

Often constructed with a singular focus on compliance, many banks’ stress test programs lack the necessary flexibility and responsiveness to adjust to today’s circumstances. The systems and processes in place often present limitations, including:

  • An overreliance on manual subprocesses. This creates process inefficiencies that can stretch analysis of a single scenario for weeks. Key person dependencies create bottlenecks and heighten operational risks. Additionally, inefficiencies may be further magnified in a remote working environment.

  • Disparate and fragmented systems. The fast-paced environment, combined with a distributed workforce, can lead to inconsistent data and incoherent results.

  • Rigid modeling frameworks. While supervisory stress scenarios are fairly constrained, the pandemic’s dubiety requires consideration of a much wider set of scenarios. This may strain the models’ operating range and require additional judgmental adjustments and overlays.

  • Underpowered processing capabilities. Limited processing power effectively constrains the number of scenarios that can be considered and limits analytical granularity.

  • Nonintegrated control processes. As people and processes are tested in a quick paced, iterative environment, the risk of operational errors is magnified.

  • Inflexible reporting. While deliverables can be well defined for regulatory stress tests, strategic use requires a significant degree of ad hoc analysis and alternative reporting.

Relieving stress: four keys to success

To address these pitfalls, a modern, comprehensive scenario-based risk management infrastructure can improve the organization’s effectiveness through a crisis by providing the following key functionalities:

1. Flexible workflow automation with cloud access

Successful response in a crisis requires timely information. A menu-driven, highly automated workflow reduces bottlenecks and helps maintain a consistent, error-free process through multiple iterations and tight deadlines. Cloud access allows analysts to collaborate while working remotely.

2. Increased centralization of data, models and processes

The collection, normalization and reconciliation of data can require considerable time and effort, hindering an institution’s ability to assess and respond quickly. A unified platform helps ensure data integrity and comparability of results across scenarios and over time.

3. High-performance analytics

Effective response requires the consideration of numerous scenarios and a range of modeling assumptions. Significant increases in processing power gained through in‐memory parallel processing supports a more granular analysis with much faster cycle times. This allows management to continually explore multiple iterations of analysis to inform their decision-making process.

4. Modularization

Unique circumstances may require processes and models to be quickly modified for ad hoc analysis. A transparent, modular framework allows for the substitution of models, data, or process steps without undue disruption of the overall workflow.

Once adopted, a scenario-based risk management program can help organizations respond more swiftly and confidently to the changing environment and growing obstacles.

The road to recovery

COVID-19 has brought sudden and significant financial and operational challenges to the banking industry. Amid the pandemic’s ambiguity, this much is clear: planning around a single baseline scenario is a potential recipe for disaster for banks as the virus continues its global advance.

The unpredictability of the pandemic’s impacts and trajectories, shifting with the daily headlines, signals that financial institutions must be prepared to respond to a wide range of plausible scenarios. It demands that banks update their analyses frequently. Come what may, robust, enterprise-wide scenario analysis prepares firms to adapt to the evolving situation – and it gives their leaders greater confidence in their stewardship through these trying times.


John Voigt, CFA, is Principal Business Solutions Manager at SAS. He has nearly 20 years of risk experience, including model development, risk strategy and consulting. Through his work, he has developed practical domain expertise across financial risk management, including regulatory and economic capital, stress testing, credit loss forecasting and CECL/IFRS 9.

back to top

Sections

About Us

Connect With Us

Resources