One of the key innovations introduced by the new accounting standards has to do with lifetime losses. Though this concept is not new in risk management, its implementation in the financial industry is extremely contemporary, as described in Section 3.1. Account-level information is usually required to develop a comprehensive modelling framework, as detailed in Section 3.2. Based on data availability, one may consider few alternative methods.

Exposure at default (EAD) represents the balance on which expected credit loss (ECL) is computed. As detailed in Section 5.1, a key distinction operates between committed products (for example, loans) and uncommitted facilities (for instance, overdrafts).

Expected credit losses (ECLs) are usually computed by relying on the product of three main risk parameters: probability of default (PD), loss given default (LGD) and exposure at default (EAD). A forward-looking view is central for the overall estimate, as detailed in Section 6.1. In this regard, the first goal of this chapter is to study multivariate time series to analyse and forecast macroeconomic scenarios. The focus of Section 6.2 is on vector auto-regression (VAR) and vector error-correction (VEC) models. Information bearing on global vector auto-regression (GVAR) modelling is also provided.

Loss given default (LGD) represents the portion of a non-recovered credit in case of default. Two extremes can easily be identified as detailed in Section 4.1. On the one hand, a full recovery is associated with 0% LGD. On the contrary, a zero recovery scenario leads to a 100% LGD.

One-year probability of default (PD) can be seen as a snapshot of lifetime PD. Two main reasons suggest treating one-year and lifetime ECLs separately. Firstly, banks have been developing one-year PD models over the last two decades for Basel II regulatory requirements as detailed in Section 2.1. Secondly, a building-block-structure split in one-year and lifetime PD facilitates the learning process. As a starting point, Section 2.2 focuses on how default events are defined for accounting purposes. As a following step, the process to build a consistent PD database is investigated in Section 2.3. Generalized linear models (GLMs) are explored as a paradigm for one-year PD estimates. Indeed, this approach is the most commonly adopted in the banking industry. A two-step scheme based on scorecard development and its calibration allows us to estimate point-in-time (PIT) PDs. In recent years, growing attention has been devoted to big data. Machine learning (ML) algorithms play a key role in this field, as detailed in Section 2.4. Classification and regression trees (CARTs), bagging, random forest, and boosting are studied both to challenge existing models and explore new PD modelling solutions. If a wide data availability encourages the use of data driven methods, low default portfolios and data scarcity are other challenges one may need to face. Indeed, limited number of defaults, requires bespoke methods. Likewise, lack of deep historical data forces banks to develop ad hoc approaches for one-year PD models as detailed in Section 2.5. Examples and case studies are explored throughout the chapter by means of R software. The laboratory Section 2.6 provides details on how to perform the implementation by means of SAS language.

As a response to incurred losses criticisms, both the International Accounting Standard Board (IASB) and Financial Accounting Standard Board (FASB) worked to redesign accounting standards towards an expected credit loss paradigm. The aim was to anticipate loss recognition by avoiding issues experienced—in particular—during the 2007–2009 financial crisis. Starting from an initial joint effort for a unique solution, IASB and FASB agreed on common principles, but then issued two separated standards. IASB's International Financial Reporting Standard number 9 (IFRS 9), issued in 2014, relies on a three-bucket classification, where one-year or lifetime expected credit losses are computed. On the contrary, FASB's Current Expected Credit Loss (CECL) accounting standard update 2016–13 (topic 326: credit losses) follows a lifetime perspective as a general rule. IFRS 9 and CECL are separately introduced in Sections 1.2 and 1.3 to point out their similarities and differences. Then the focus is on the link connecting expected credit loss estimates and capital requirements as detailed in Section 1.4. As a final step, a book overview is provided in Section 1.5 as a guide for the reader willing to grasp on overview of the entire expected credit loss modelling and validation journey.