CCAR and DFAST Stress Testing Survey Results

The Fed has emphasized that banks could leverage the stress testing results to inform their business decisions, but when survey participants were asked if they used stress testing for other purposes beyond regulatory compliance and capital planning, the results quickly drop.

Source: Moody’s Analytics Stress Testing and Capital Planning Roundtable

Both CCAR and DFAST banks ranked loss estimation as the top challenge, with model validation coming in second. The biggest gap was with documentation, with regulators asking larger banks for more details and clearer explanations. Unsurprisingly, small banks are also not as concerned about reporting and Pre-Provision Net Revenue (PPNR) modeling at this point.

Source: Moody’s Analytics Stress Testing and Capital Planning Roundtable

Across 30 different banks, large and small, enhancing credit modeling was the top priority for improvement in our surveys. They also identified a need to upgrade data infrastructure, and improve stress testing processes and workflow. Indeed, CCAR banks are anticipating a significant increase in stress testing spending, with an estimated operating cost of $10 million per year, and with a technology spend of $30-50 million on top of that.

1. Developing improved primary models for credit risk stresses
2. Enhancing documentation and documentation standards
3. Improving econometric estimation of interest income and expense
4. Enhancing estimation of new business volumes under adverse and more adverse stress conditions
5. Developing a centralized environment for managing all CCAR report assembly
6. Estimating non-interest income and expense in a more rigorous, econometric fashion
7. Calculating pro-forma RWAs and regulatory capital numbers/ratios more effectively
8. Increasing the usefulness of the stress testing outputs/results
9. Improving challenger and benchmark models

For DFAST banks, developing primary and challenger models were the top priorities for 2015 in our surveys. Smaller banks will focus on migrating from top-down, statistically-based models to bottom-up models with more granular data, but they will need to clean up their data and data infrastructure as a first step.

Source: Moody’s Analytics

More than 85% of CCAR institutions surveyed also noted that multiple departments contribute to stress testing, with central coordination provided by several business units.

Source: Moody’s Analytics

In general, CCAR institutions are preparing for their annual regulatory submission by performing key activities in a certain order on average. Participants noted that they need to overcome organizational obstacles to efficiently coordinate stress testing activities in disparate areas of their institutions.

Source: Moody’s Analytics

Percentage of DFAST banks with a roadmap for improving stress testing over the next few years.

Source: Moody’s Analytics

Only about a third of DFAST institutions use loan-level data for their modeling. That number should grow as banks improve their data sources and data capturing processes.

Source: Moody’s Analytics