Large Bank Examination Workshop February 2026

Summary Findings

Based on requirements for measuring and managing interest rate risk (IRR) as established by

regulatory expectations and industry best practices, the Model and supporting processes are

categorized as aligning with Best Practice (see Model Validation Scorecard ™ ). The conceptual

framework is well-designed for an institution of this size and complexity. There are some findings

and suggestions noted to provide the Bank with an enhanced view of interest rate risk.

I - Data and Assumptions

A review of the model procedures indicated that the extract, transform, and load (ETL) process

comprises primarily a simple data transfer, which is appropriate given the relatively standard output

provided from the Bank ’s core systems . SFS reviewed the various source system extracts used to

feed the Model. The Bank ’s prepayment and core deposit behavior data were rigorously scrutinized

to evaluate the reasonableness of their assumptions regarding their optionality. We reviewed the

processes that tie out the chart of accounts balances to general ledger balances and the

methodology used to reconcile differences.

The Model includes an automated Data Input and Source Integrity check. Among other factors,

key balances within the model are compared to source data to ensure that the ETL process has been

conducted appropriately. The results are as follows for the 9/30/22 period:

SFS also identified and reviewed custom inputs, such as Excel spreadsheets, which were used to

feed the Model. No material concerns regarding the reconciliation process were noted.

ALMTech ALM is used to calculate Net Interest Income (NII, or Earnings at Risk) and Economic

Value of Equity (EVE, or Capital at Risk) at the consolidated bank holding company level. Financial

4

Page 16 of 39