Credit Risk Models: PD, LGD & EAD — 2026 IIBF Risk Guide

Credit risk models — this guide gives you the latest 2026 understanding of how banks quantify expected loss through PD, LGD and EAD, and exactly what IIBF Risk Management candidates must remember.

For anyone preparing the Risk Management certification, credit risk models sit at the heart of the syllabus. They turn the vague idea of "this borrower might default" into measurable numbers a bank can price, provision and hold capital against. The three building blocks are the Probability of Default, the Loss Given Default and the Exposure at Default.

In this guide we unpack each component, the expected-loss equation that ties them together, how they map to the Basel framework, and how candidates should approach this quantitative topic in the exam with confidence.

What Are Credit Risk Models and Why They Matter

Credit risk models are quantitative frameworks that estimate the loss a bank may suffer if a borrower fails to meet its obligations. Rather than treating every loan as equally risky, they assign a measurable risk to each exposure, allowing a bank to price loans correctly, set provisions, and calculate the regulatory capital it must hold.

The central output is expected loss — the average loss a bank can anticipate over a given horizon. Expected loss is built from three parameters: the Probability of Default (PD), the Loss Given Default (LGD) and the Exposure at Default (EAD). Together these convert credit quality into a rupee figure.

For a banker, mastering credit risk models means understanding why two loans of the same size can require very different capital. Sharpen these distinctions with our IIBF mock tests as you revise.

Probability of Default (PD)

Within credit risk models, the Probability of Default is the likelihood that a borrower will default over a specified time horizon, usually one year. It is expressed as a percentage between 0 and 100. A blue-chip corporate may carry a PD of a fraction of a percent, while a stressed borrower may carry a double-digit PD.

PD is typically estimated from a combination of internal rating models, historical default experience for similar borrowers, and financial and behavioural indicators. Under the Basel framework, default is generally defined as the borrower being unlikely to pay, or being more than 90 days past due on a material obligation — the same 90-day idea that underpins NPA recognition.

For the exam, remember PD measures the chance of the event, not the size of the loss. Be ready to distinguish a through-the-cycle PD from a point-in-time PD. Broaden your reading with related guides on our blog.

Loss Given Default (LGD) and Exposure at Default (EAD)

The second parameter in credit risk models is Loss Given Default, the proportion of the exposure a bank expects to lose if default actually occurs, after accounting for recoveries from collateral, guarantees and workout. LGD is expressed as a percentage; its complement, the recovery rate, equals one minus LGD. A well-secured loan has a low LGD, an unsecured one a high LGD.

The third parameter is Exposure at Default, the total amount the bank is likely to be owed at the moment of default. For a fully drawn term loan this is straightforward, but for revolving facilities such as credit cards or overdrafts, EAD must allow for the borrower drawing down further undrawn limits as they approach distress, often captured through a credit conversion factor.

Understanding how collateral lowers LGD, and how undrawn commitments raise EAD, is exactly the applied insight examiners reward. Keep policy and rate context current via our RBI rates and resources page.

Expected Loss and the Basel Link

Credit risk models bring the three parameters together in a single, much-loved equation: Expected Loss = PD x LGD x EAD. If a bank faces a 2% probability of default, a 40% loss given default, and an exposure of 100 units, its expected loss is 0.02 x 0.40 x 100 = 0.8 units. This simple multiplication is one of the most frequently tested calculations in the paper.

Expected loss is meant to be covered by provisions and pricing — it is the cost of doing business. Unexpected loss, the volatility around that average, is what regulatory capital is designed to absorb. Under the Basel framework, banks using the Internal Ratings-Based approach estimate PD (and, in advanced IRB, also LGD and EAD) to compute capital requirements, subject to supervisory floors and validation.

For exam scenarios, practise plugging numbers into the expected-loss formula and interpreting how a change in any one parameter moves the result. Track regulatory and Basel-related updates on our IIBF updates page.

Exam Strategy for Risk Management Candidates

Questions on credit risk models typically test the definitions of PD, LGD and EAD, the expected-loss formula, the distinction between expected and unexpected loss, and short numericals applying EL = PD x LGD x EAD. Build a one-page sheet defining each parameter with its unit and a worked example, and rehearse the calculation until it is automatic.

Pair conceptual study with timed practice and connect the parameters to provisioning and capital so your answers show genuine understanding. Review weak areas after every mock and keep your formula notes crisp. Begin your free IIBF practice tests today and track your progress on iibf.store.

Source: Reserve Bank of India — rbi.org.in

Frequently Asked Questions

Master credit risk models and the rest of the Risk Management syllabus by combining conceptual notes with timed practice. Start your free IIBF mock tests today and track your progress on iibf.store.