LO 19.2: Describe the experts-based approaches, statistical-based models, and

LO 19.2: Describe the experts-based approaches, statistical-based models, and numerical approaches to predicting default.
Although the consequences of default can be substantial, fortunately a default itself is a relatively rare occurrence (the default rate during deep recessions peaks in the range of 2% to 3%). A credit analyst whose job it is to assess the potential for default is typically an individual with a great deal of experience who can balance his knowledge with perception and intuition when evaluating default scenarios.
An early model for assessing default was created by Wilcox (1971)1 using what was called gamblers ruin theory. His model for predicting the probability of default was dependent on assessing the probability of gains and losses as well as the level of profits relative to a companys initial capital endowment. Another theory applied to corporate finance is the point of no return theory, which implies that business operations must produce enough cash to cover required interest and principal payments on debt. As long as the operational flow of funds exceeds interest and principal payments needed, the company will be successful. The balance needed represents the no-return point, as a company can only be sustainable as long as it can meet its debt payments.
Credit quality analysis from an experts-based approach will apply frameworks such as the four Cs of credit (Character, Capital, Coverage, Collateral) proposed by Altman/NYU, LAPS (Liquidity, Activity, Profitability, Structure) from Goldman Sachs, and CAMELS (Capital Adequacy, Asset Quality, Management, Earnings, Liquidity, Sensitivity) from JP Morgan. As Porter (1980, 1985)23 emphasized, qualitative features need to be factored into any analysis along with quantitative components.
A statistical-based classification centers on the fact that a quantitative model is essentially just a description of the real world within a controlled environment. Models are simply used to express a viewpoint of how the world will likely behave given certain criteria. A quantitative model will have a qualitative (formal) formulation that describes the basic view of the world we are trying to capture in the model; it will also have the underlying assumptions needed to build the model. The assumptions, which serve to simplify the process, should cover organizational behavior, possible economic events, and predictions on how market participants will react to these events. Statistical-based models are primarily focused on assessing the default risk associated with unlisted firms, even though they certainly can be useful in managing default risk for many other entities and organizations. Here, the model is based on quantitative and qualitative variables, as well as publicly unavailable and low-frequency data.
As will be described later in the topic, numerical approaches have the objective of deriving optimal solutions using trained algorithms and incorporating decisions based on relatively weak information in very complex environments. An example is a neural network, which is able to continuously update itself for changes to the environment.
1. Wilcox, J. W. (1971), A Gamblers Ruin Prediction of Business Failure Using Accounting Data,
Sloan Management Review, 12 (3).
2. Porter, M. (1980), Competitive Strategy, Free Press. 3. Porter, M. (1985), Competitive Advantage: Creating and Sustaining Superior Performance, Free Press.
2018 Kaplan, Inc.
Page 39
Topic 19 Cross Reference to GARP Assigned Reading – De Laurentis et al., Chapter 3
R a t i n g M i g r a t i o n M a t r i x