Banking Model Audit
Executive Summary
Key Takeaways
- ✓ A structural audit of a bank model tests whether the model's formulas and logic as actually built calculate correctly, distinct from model validation's broader test of whether the model's design and calibration are fit for purpose.
- ✓ The segmented balance sheet, interest income and NIM bridge, credit loss provisioning, and capital adequacy modules built across this domain should each be checked for internal consistency — for example, that the volumes feeding the interest income build match the volumes in the loan portfolio and deposit modules, not a separately maintained set of figures.
- ✓ Common structural findings in bank models include broken links between the balance sheet forecast and the capital adequacy module, hardcoded values overriding what should be live formula-driven figures, and inconsistent formulas across segment rows that should follow an identical calculation pattern.
- ✓ A structural audit does not itself validate whether a specific regulatory capital or liquidity calculation is correct under the applicable framework — that determination requires model validation and, ultimately, the bank's own regulatory reporting process.
- ✓ A bank model audit's value is highest when performed on the specific model actually used for the decision in question — a credit approval, a capital forecast — not a simplified or illustrative version of the model that may not reflect the same structural risks.
Objective¶
This guide covers how a structural audit should be applied to a bank financial model, within Banking Model Risk, extending the general Financial Model Auditing pillar to the specific modules built across this domain.
What a Banking Structural Audit Checks¶
A structural audit tests whether the model's formulas and logic as actually built calculate correctly, checking internal consistency across the modules this domain has covered:
| Module | Structural Consistency Check |
|---|---|
| Segmented balance sheet | Do loan and deposit segment volumes reconcile across the balance sheet, interest income, and provisioning modules? |
| Interest income and NIM bridge | Does the NIM bridge's volume/rate/mix decomposition reconcile to the actual period-over-period change? |
| Credit loss provisioning | Does the allowance roll-forward reconcile provision charges, write-offs, and recoveries to the opening and closing balance? |
| Capital adequacy | Do the capital ratios calculate as live formulas from the capital tier build and risk-weighted asset calculation, not a disconnected figure? |
Common Structural Findings¶
- Broken links between the balance sheet forecast and the capital adequacy module, where the risk-weighted asset calculation references a stale or disconnected set of balance figures rather than the live forecast.
- Hardcoded values overriding what should be live formula-driven figures, particularly in regulatory ratio calculations where a specific ratio has been manually entered rather than calculated.
- Inconsistent formulas across segment rows that should follow an identical calculation pattern (the same yield or loss-rate formula structure per segment) but have diverged through ad hoc manual editing over time.
What a Structural Audit Does Not Do¶
A structural audit tests the model's own internal formula correctness. It does not validate whether a specific regulatory capital or liquidity calculation is correct under the applicable regulatory framework — that determination requires model validation and, ultimately, the bank's own regulatory reporting and compliance process, consistent with the scope boundary described throughout this domain's capital and liquidity guides.
Auditing the Model Actually in Use¶
A bank model audit's value is highest when performed on the specific model version actually used for the decision in question — a credit approval, a capital forecast for a board presentation — rather than a simplified or illustrative version that may have been built with a different structure and would not surface the same structural risks present in the live decision-making model.
Common Construction Pitfalls¶
- Auditing a simplified or illustrative version of a model rather than the specific version actually used for the underlying decision.
- Treating a structural audit finding as evidence about the correctness of the underlying regulatory calculation, rather than only the model's own formula integrity.
- Overlooking segment-level formula consistency, checking only the aggregate totals rather than confirming each segment row follows the intended calculation pattern.
- Conflating a structural audit with model validation, or assuming either discipline substitutes for the other.
Continue Reading¶
Prerequisites¶
- Banking Model Risk — the parent guide
- Financial Model Auditing
Related Technical Guides¶
Related Checklists¶
How OXXON tests thisRun a free structural check with FMAE
Frequently Asked Questions
What does a structural audit of a bank model check?
Whether the model's formulas and logic as actually built calculate correctly — internal consistency across the segmented balance sheet, interest income build, credit loss provisioning, and capital adequacy modules, and the absence of structural errors like broken links, hardcodes, and inconsistent formulas.
How is a structural audit different from model validation?
A structural audit tests whether the model's formulas as actually built calculate correctly; model validation tests the broader question of whether the model's design, assumptions, and calibration are fit for its intended purpose — see Banking Model Validation for that broader discipline.
What are common structural findings in bank models?
Broken links between the balance sheet forecast and the capital adequacy module, hardcoded values overriding what should be live formula-driven figures (particularly in regulatory ratio calculations), and inconsistent formulas across segment rows that should follow an identical calculation pattern but have diverged through manual editing.
Does a structural audit validate a specific regulatory capital or liquidity calculation?
No — a structural audit tests the model's own internal formula correctness, not whether a specific regulatory capital or liquidity calculation is correct under the applicable framework, which requires model validation and the bank's own regulatory reporting and compliance process.
Why does it matter which specific model version is audited?
Because a bank model audit's value is highest when performed on the specific model actually used for the decision in question, not a simplified or illustrative version that may have been built differently and would not reflect the same structural risks present in the live decision-making model.
How does this guide relate to the general Financial Model Auditing pillar?
This guide extends that general pillar with the specific modules and structural risks particular to bank models — the segmented balance sheet, NIM bridge, provisioning, and capital adequacy modules built across this domain.
Related Articles
Banking Model Risk
Model risk in banking is a distinct, heavily formalized discipline, because banks rely on models for decisions with direct regulatory and financial stability consequences — credit decisions, capital adequacy, and liquidity management chief among them. This guide extends the general Model Risk pillar with the banking-specific model taxonomy (credit, valuation, capital, liquidity models), the three-lines-of-defense structure common to bank model risk management frameworks, and why banking model risk management is typically more formalized than in most other industries.
Banking Model Validation
Banking model validation is the independent, second-line function that tests a bank model's conceptual soundness, implementation accuracy, and ongoing performance against actual outcomes. This guide covers the three pillars of a banking model validation exercise: conceptual soundness review (does the model's design make sense for its intended use), implementation testing (does the model as built actually implement its intended design), and outcomes analysis (does the model's output track what actually happens over time) — and why validation is a distinct discipline from a structural audit.
What Is a Financial Model Audit?
A financial model audit is an independent, structured examination of an Excel based financial model to confirm that its mechanics, logic, and outputs are reliable enough to support a decision. It is not a check of whether the assumptions are optimistic or conservative. It is a check of whether the model actually calculates what its author believes it calculates. Every year, lenders extend debt, investment committees approve capital, and boards sign off on transactions using numbers that came out of a spreadsheet nobody outside the immediate deal team has independently verified. A financial model audit exists to close that gap before it becomes expensive.
Bank Capital Adequacy Checklist
This checklist covers the structural construction of a bank model's capital adequacy build, from capital tier segmentation and deductions through risk-weighted asset calculation, minimum ratio and buffer thresholds, and the live connection between the balance sheet forecast and the resulting capital ratios. It is a construction-discipline checklist, distinct from validating whether any specific regulatory capital calculation itself is correct.
Independent Review in Banking
Independent review of a bank model — performed by a party outside the bank's own model risk management function — provides a perspective distinct from internal model validation, even when both disciplines cover similar technical ground. This guide covers why independence from the institution itself matters beyond independence from the first-line business unit, when a bank should seek external independent review in addition to its internal second-line validation function, and how a lending syndicate or regulator might rely on independent review differently than the bank's own governance process.