🏦 Internal Ratings-Based (IRB) Approach in Banking

Advanced credit risk measurement under Basel norms

🌟 Introduction

Credit risk—the risk that borrowers may fail to meet their obligations—is the largest risk faced by banks. To ensure banks hold sufficient capital against credit risk, global regulators introduced standardized and advanced risk measurement frameworks under the Basel Accords.

One of the most sophisticated of these is the Internal Ratings-Based (IRB) Approach, which allows banks to use their own internal credit risk models—subject to regulatory approval—to estimate capital requirements.

The IRB approach represents a shift from rule-based regulation to risk-sensitive, data-driven supervision.

📌 What Is the IRB Approach?

The Internal Ratings-Based (IRB) Approach is a method under the Basel framework that allows banks to estimate key credit risk parameters internally and use them to calculate Regulatory Capital for Credit Risk.

Under IRB, banks estimate:

PD (Probability of Default)
LGD (Loss Given Default) (in some cases)
EAD (Exposure at Default) (in some cases)
M (Effective Maturity)

These inputs are plugged into Basel-specified risk-weight functions to compute Risk-Weighted Assets (RWA).

🧱 Evolution of the IRB Approach

The IRB approach was formally introduced under:

Basel II – Advanced credit risk measurement
Strengthened under Basel III – Higher capital quality, buffers, and model governance

It is governed globally by the Basel Committee on Banking Supervision, operating under the Bank for International Settlements.

🧭 Types of IRB Approaches

1️⃣ Foundation IRB (F-IRB)

Bank estimates: PD
Regulators provide: LGD, EAD, M

📌 Suitable for banks transitioning from standardized approaches.

2️⃣ Advanced IRB (A-IRB)

Bank estimates: PD, LGD, EAD, M
Requires highly sophisticated risk systems

📌 Used by large, internationally active banks.

🧮 Key Credit Risk Components in IRB

🔹 Probability of Default (PD)

Likelihood that a borrower defaults within 1 year
Estimated using internal rating models

📌 Example:
A corporate borrower rated “BBB” has

PD = 1.5%

🔹 Loss Given Default (LGD)

Percentage of exposure lost if default occurs
Depends on collateral, seniority, recovery rates

📌 Example:
Secured loan → LGD = 40%
Unsecured loan → LGD = 60%

🔹 Exposure at Default (EAD)

Amount outstanding at the time of default
Includes drawn + expected future drawings

📌 Example:
Loan limit = ₹10 crore
Drawn = ₹7 crore
Estimated EAD = ₹8.5 crore

🔹 Maturity (M)

Remaining economic maturity of exposure
Impacts capital requirement for long-term loans

🔄 IRB Credit Risk Calculation Flow

Flow:

Internal borrower rating
PD estimation
LGD & EAD estimation
Basel risk-weight function
RWA calculation
Capital requirement (8% of RWA)

📊 Worked Example: IRB Capital Calculation

Assumptions (Corporate Loan):

Exposure (EAD) = ₹100 crore
PD = 2%
LGD = 45%
Maturity = 2.5 years

Using Basel’s IRB formula (simplified):

Expected Loss (EL) = PD × LGD × EAD

EL = 0.02 × 0.45 × 100 = ₹0.9 crore

The Basel risk-weight function converts PD, LGD, and M into Risk-Weighted Assets (RWA).

If:

RWA = ₹80 crore

Then minimum capital required:

Capital = 8% × 80 = ₹6.4 crore

📌 IRB typically produces more risk-sensitive capital than standardized methods.

📈 Benefits of the IRB Approach

✔ Risk-sensitive capital allocation
✔ Better differentiation between good and bad borrowers
✔ Encourages improved data, analytics, and governance
✔ Aligns regulatory capital with economic risk

⚠️ Challenges and Risks of IRB

✘ Model risk and estimation errors
✘ Data quality and historical depth requirements
✘ Procyclicality (capital rises in downturns)
✘ Regulatory scrutiny and approval complexity
✘ High implementation and maintenance cost

🏛️ Regulatory Oversight and Governance

Supervisors require banks to demonstrate:

Robust model validation
Independent risk governance
Back-testing and stress testing
Clear use-test (models used in business decisions)

In India, IRB adoption is overseen by the Reserve Bank of India, which has taken a cautious and phased approach.

🔍 IRB vs Standardized Approach

Aspect	Standardized	IRB
Risk sensitivity	Low	High
Data requirement	Low	Very high
Regulatory approval	Not required	Mandatory
Model usage	None	Extensive
Capital efficiency	Lower	Potentially higher

🌍 Real-World Use Cases

Corporate & SME lending
Retail credit (home loans, credit cards)
Project finance
Wholesale banking portfolios

Large global banks in Europe and the US primarily use Advanced IRB for major portfolios.

🔮 Post-Basel III & Basel IV Developments

Reduced model flexibility (output floors)
Constraints on IRB for certain asset classes
Greater emphasis on stress testing
Increased transparency and comparability

📌 Regulators aim to balance risk sensitivity with simplicity.

🔁 IRB vs Expected Credit Loss (ECL) under IFRS 9

Regulatory capital vs accounting impairment

Although both the Internal Ratings-Based (IRB) approach and Expected Credit Loss (ECL) under IFRS 9 rely on similar credit risk concepts (PD, LGD, EAD), their objectives and usage are fundamentally different.

IRB → Regulatory capital adequacy
IFRS 9 ECL → Financial reporting and loan loss provisioning

Understanding this distinction is crucial for bankers, risk managers, auditors, and analysts.

🎯 Purpose: Why They Exist

Framework	Primary Purpose
IRB (Basel)	Ensure banks hold sufficient regulatory capital to absorb unexpected losses
IFRS 9 (ECL)	Ensure timely recognition of expected credit losses in financial statements

📌 Key insight:

IRB focuses on unexpected loss (capital buffer), while IFRS 9 focuses on expected loss (provisions).

🧮 Core Risk Parameters: Similar Inputs, Different Use

Both frameworks use:

PD (Probability of Default)
LGD (Loss Given Default)
EAD (Exposure at Default)

But they differ in time horizon and treatment.

Aspect	IRB	IFRS 9 ECL
PD Horizon	12-month PD	12-month or Lifetime PD
LGD	Downturn LGD	Point-in-time LGD
EAD	Regulatory estimate	Accounting estimate
Forward-looking info	Limited	Mandatory (macroeconomic scenarios)

📆 Time Horizon Difference (Very Important)

🔹 IRB

Uses 1-year PD for capital calculation
Focused on through-the-cycle risk
Less sensitive to short-term economic fluctuations

🔹 IFRS 9

Uses a staging approach:

Stage	Credit Quality	Loss Recognition
Stage 1	Performing	12-month ECL
Stage 2	Significant credit deterioration	Lifetime ECL
Stage 3	Credit-impaired	Lifetime ECL + interest on net basis

📌 IFRS 9 is much more forward-looking and procyclical.

📊 Numerical Example: IRB vs IFRS 9

Loan details:

EAD = ₹100 crore
PD (12-month) = 2%
LGD = 45%

🔹 Expected Credit Loss (IFRS 9 – Stage 1)

ECL = PD × LGD × EAD

ECL = 0.02 × 0.45 × 100 = ₹0.9 crore

➡ Recognized as provision in P&L

🔹 IRB Capital Treatment

Expected Loss (EL) = ₹0.9 crore
Unexpected Loss → converted into Risk-Weighted Assets (RWA)
Capital required = 8% of RWA

📌 EL is not expensed, but absorbed through provisions and capital buffers.

🏦 Accounting vs Regulatory Treatment

Dimension	IRB	IFRS 9
Framework type	Regulatory	Accounting
Governing body	Basel Committee	International Accounting Standards Board
Affects	Capital Adequacy Ratio	Profit & Loss
Volatility	Lower	Higher
Macro sensitivity	Low	High

⚖️ Procyclicality: A Key Policy Concern

IFRS 9 increases provisions sharply during downturns
IRB smooths risk through long-term averages

📌 This can create tension:

Higher ECL → lower profits
Higher risk → higher capital requirements

Banks must manage capital + provisioning simultaneously.

🔄 Alignment Challenges for Banks

Banks face challenges in aligning:

IRB PD models (through-the-cycle)
IFRS 9 PD models (point-in-time)
Data consistency
Model governance and validation
Regulatory vs audit expectations

📌 Many banks maintain parallel model frameworks.

🧠 Role of Analytics & AI

Advanced analytics helps by:

Translating TTC PDs into PIT PDs
Scenario-based ECL estimation
Stress testing capital and provisions together
Improving explainability for regulators and auditors

🧾 Key Takeaways

IRB is a data-driven, advanced credit risk framework
Relies on PD, LGD, EAD, and M
Enables more accurate capital allocation
Requires strong analytics, governance, and regulatory oversight
Central to modern banking risk management
IFRS 9 is more volatile and forward-looking
Both IRB and IFRS 9 use PD, LGD, and EAD—but for different purposes
IRB protects the system via capital buffers
IFRS 9 protects transparency via early loss recognition
Effective bank risk management integrates both frameworks

📚 References & Further Reading

Basel Committee on Banking Supervision – Basel II & Basel III Credit Risk Framework
Bank for International Settlements – Global Regulatory Standards
Reserve Bank of India – Guidelines on Capital Adequacy
International Monetary Fund – Financial Stability Reports
Saunders & Allen – Credit Risk Management in and out of the Financial Crisis
Hull, J. – Risk Management and Financial Institutions
GARP – FRM Curriculum (Credit Risk)