What it means
Bank capital rules decide how much loss-absorbing money a bank must hold against its loans. The AIRB approach is the most sophisticated version: instead of using regulators' standard risk weights, qualifying banks estimate the key risk parameters from their own data and models.
Three parameters drive the calculation, namely the probability that a borrower defaults, the share of exposure the bank expects to lose if default happens, and the amount likely to be outstanding at that moment. Regulatory formulas turn those parameters into capital.
The Basel framework supplies the risk-weight functions and the bank's internal estimates feed them, producing risk-weighted assets and therefore the minimum capital each loan requires. The prize is capital efficiency, because banks with genuinely better risk data and safer books can justify lower capital than standardised weights demand, which is why the largest banks invested heavily in internal models.
The price is supervision. Supervisors must approve a bank's models before use, validate them regularly and can impose add-ons, because a bank grading its own homework has an obvious incentive to flatter itself.
The foundation variant splits the difference: under the foundation internal ratings approach, banks estimate only default probabilities while regulators fix the loss and exposure parameters, a halfway step for banks not yet ready for full self-estimation. The approach reshaped bank behaviour.
Because low-risk loans consume less capital, the framework pushed banks toward better borrower data, more granular rating systems and pricing that reflects each loan's capital cost. Reforms have trimmed the freedom.
After the financial crisis exposed model optimism, regulators introduced output floors limiting how far internal-model capital can fall below the standardised result, and restricted modelled approaches for some exposures. For a manager dealing with a large bank, the approach explains pricing differences, since a bank's capital model feeds the return it demands on each loan and credit terms partly reflect how the lender's internal ratings view your business.
The concept also generalises as risk-sensitive regulation. The same philosophy, capital set by measured risk rather than fixed buckets, runs through modern banking and insurance supervision worldwide.
In practice
Real-world examples.
Example
A global bank estimates a corporate borrower's default probability at 0.3% and loss given default at 40%. Its approved model assigns the loan far less capital than the standardised weight would, which lets the bank price the loan more keenly than a rival using standard weights.
Example
A supervisor reviewing a bank's AIRB models finds loss-given-default estimates too optimistic for one portfolio. The supervisor requires an add-on that raises the bank's reported capital need, and the bank must also rebuild the estimates using longer data history.
Example
Under the output floor, a bank's internal models imply $60 billion of risk-weighted assets, while the standardised approach gives $100 billion. A floor of 72.5% of the standardised figure, once fully applied, is $72.5 billion, so the bank must hold capital on $72.5 billion rather than $60 billion.
Formula
Calculation
Expected loss = PD x LGD x EAD, where PD is the probability of default, LGD is loss given default and EAD is exposure at default. Capital = EAD x risk weight x capital ratio, where the risk weight comes from a Basel risk-weight function that uses PD, LGD and the framework's maturity and correlation adjustments to cover unexpected losses.
Worked example: with PD of 1%, LGD of 45% and EAD of $10,000,000, expected loss is 1% x 45% x $10,000,000 = $45,000. Suppose, purely for illustration, the risk-weight function returns 50% for this loan, so risk-weighted assets are $10,000,000 x 50% = $5,000,000. At an 8% capital ratio, capital is $5,000,000 x 8% = $400,000, compared with $10,000,000 x 100% x 8% = $800,000 if a standardised 100% weight applied.Case study
Seen in the real world.
A made-up regional bank spends three years building rating models and data history to win AIRB approval. This case study is fictional and illustrative. Approval cuts capital on its mortgage book, but its corporate estimates are floored by the supervisor, and the bank learns that model quality, not model existence, drives the benefit.
The finance team then tracks a single measure each quarter: capital saved against the standardised result, split by portfolio. Mortgages show a large saving, while corporate lending shows almost none. The board uses that split to direct new data-collection effort toward the corporate book, where better evidence could justify lower estimates.
Watch out
Common mistakes.
- Assuming internal models mean self-regulation; supervisors approve, validate and override AIRB models, and output floors cap how far modelled capital can fall below standardised results.
- Treating parameter estimates as fixed truths; default and loss estimates drift with the cycle, so models require continuous validation, back-testing and conservative margins.
- Confusing expected loss with capital; provisions cover expected losses from PD times LGD times EAD, while capital exists for unexpected losses, and the two must not be double-counted or swapped.
Questions
People also ask.
What is the advanced internal rating-based approach?
A Basel framework method where approved banks calculate credit-risk capital from their own estimates of probability of default, loss given default and exposure at default, fed through regulatory risk-weight formulas.
How does AIRB differ from the foundation approach?
Under the foundation approach, banks estimate only default probabilities while regulators fix the loss and exposure parameters. AIRB banks estimate all three, requiring deeper data and passing stricter supervisory approval.
Why did regulators add output floors?
Because post-crisis reviews found some banks' models produced implausibly low capital. Floors limit modelled capital to a percentage of the standardised result, keeping internal models honest against an external yardstick.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%Related