Introduction
In our last lesson, we explored one-factor models like Vasicek and CIR, which describe the "risk-free" term structure of interest rates. These models operate under an important simplifying assumption: that the borrower (e.g., a government) will never fail to repay its debt. Now, we step into the real world where this assumption does not always hold.
This lesson introduces credit risk, the risk of financial loss arising from a borrower's failure to meet their obligations. Our goal is to formally define this crucial concept and describe the specific "credit events," such as default, that trigger these losses. Understanding these definitions is the essential first step before we can begin to build the quantitative models used to measure and manage this risk, which will be our focus in the lessons to come.
What is Credit Risk?
At its core, credit risk is the potential for a lender to suffer a financial loss because a borrower or counterparty fails to meet their contractual obligations. This applies to a wide range of financial arrangements, from an individual's mortgage to a corporation's bonds or complex derivative contracts between institutions.
The following video provides a concise and formal introduction to the concept.
This video from Quant Next, "Credit Risk: An Introduction," clearly defines credit risk and provides examples of the different types of borrowers and debt instruments involved.
Please watch the first part of the video, from the beginning until the end of the introduction. Focus on how credit risk is defined as a potential for loss and the variety of contexts in which it appears.
As the video explains, the borrower can be an individual, a company, or even a government, and the debt can take many forms. The unifying theme is the risk of loss due to non-fulfillment of an agreement.
The Building Blocks of Credit Loss
From a statistician's perspective, managing risk begins with quantifying it. For credit risk, the most fundamental measure is the Expected Credit Loss (ECL). This is the average loss a lender anticipates over a specific period. It is not an unexpected, catastrophic event but rather the statistically predictable "cost of doing business" in lending.
The ECL is calculated by multiplying three key components. This decomposition is central to nearly all modern credit risk modeling.
Let's break down each term:
- Probability of Default (PD): This is the likelihood that the borrower will default on their obligation within a specified time horizon (typically one year). Estimating the PD is a major focus of credit risk modeling, and we will explore methods to do this in our next lesson.
- Exposure at Default (EAD): This is the total value that the lender is exposed to when the borrower defaults. For a simple term loan, it might be the outstanding principal. For a credit card or line of credit, it could be the full credit limit, as a borrower in distress might draw down all available funds before defaulting.
- Loss Given Default (LGD): This is the proportion of the EAD that is ultimately lost after all recovery efforts have been exhausted. It is expressed as a percentage. For example, if a bank repossesses and sells a house after a mortgage default, the proceeds from the sale reduce the total loss. The LGD is what remains. It is calculated as .
The following reading provides a clear definition of this formula and a practical example of its calculation.
Measuring Credit Risk | FRM Part 1 Study Notes - AnalystPrep
This article from AnalystPrep provides the mathematical formulation for Expected Loss and walks through a numerical example.
Please read the section titled "Calculation of the Expected Loss" and then work through the accompanying example. This will solidify your understanding of how the three components combine.
By breaking the problem down this way, financial institutions can model and manage each component separately. For example, they might use statistical models for PD, set contractual limits to manage EAD, and improve their legal and collections processes to reduce LGD.
Common Credit Events
We have used the term "default" frequently, but what does it technically mean for a credit event to occur? A default is not a single, universally defined event but rather a category of triggers specified in a loan agreement or financial contract. The International Swaps and Derivatives Association (ISDA) provides a standardized list that is widely used, especially in the derivatives market.
The following resource provides clear definitions for the most common credit events.
Credit Event - Corporate Finance Institute
This article clearly defines what a credit event is and explains the most common types with simple, illustrative examples.
Please read the following sections: Start with the list of six ISDA credit events to get an overview. Then, read the detailed explanations for the three most common events: Bankruptcy, Payment Default, and Debt Restructuring. Pay close attention to the examples, which help distinguish between these related concepts.
To summarize the key distinctions from the reading:
- Bankruptcy: A formal legal status declared by a court when an entity cannot repay its debts.
- Payment Default: A failure to make a specific interest or principal payment on its due date. This is often a precursor to bankruptcy.
- Debt Restructuring: A modification of the debt terms that is unfavorable to the lender, such as reducing the interest rate, forgiving some of the principal, or extending the maturity date. This is considered a "soft" default, as the borrower has not technically failed to pay but has unilaterally changed the rules.
Other credit events defined by ISDA, such as Obligation Acceleration (when a default on one debt triggers a demand for immediate repayment on other debts) and Repudiation/Moratorium (rejection of the debt obligation, often by a sovereign government), are also important but less frequently encountered in typical corporate lending.
Beyond the Individual: Portfolio and Systemic Risk
So far, we have focused on the risk of a single borrower. However, in a large portfolio of loans, the risks are interconnected. As a statistician, you know that the assumption of independence is powerful but often violated in real-world systems. This is particularly true in finance.
Loan defaults are not independent events. They tend to be correlated because they are often driven by common underlying factors, primarily the overall health of the economy.
- Systematic Risk: During an economic recession, many companies may see their revenues decline simultaneously, increasing the probability of default across the board. This shared exposure to macroeconomic factors is a form of systematic risk that cannot be diversified away simply by holding more loans.
- Credit Contagion: This describes a domino effect where the default of one major company causes its suppliers and creditors to face financial distress, potentially leading to their defaults as well.
This means that a portfolio of loans is exposed to different layers of risk. The video below categorizes these risks, distinguishing between those affecting a single borrower and those affecting an entire portfolio.
Let's return to the "Credit Risk: An Introduction" video from Quant Next to explore the different types of credit risk.
Watch the segment from Types of credit risk. Note the distinction between individual credit risk (like default and downgrade risk) and portfolio credit risk (like concentration, sector, and systemic risk).
Understanding this correlation structure is one of the most challenging aspects of credit risk management and is a major area of research in financial engineering. The simple ECL formula assumes you can determine PD for a single borrower, but in reality, the PD of one borrower is often conditional on the behavior of others and the state of the economy.
Conclusion
In this lesson, we have formally defined the concept of credit risk and its fundamental components. We've moved beyond the idealized world of risk-free assets to acknowledge the very real possibility of loss due to a counterparty's failure to meet its obligations.
Key Takeaways:
- Credit Risk is the potential for loss when a borrower or counterparty fails to meet their contractual obligations.
- The Expected Credit Loss (ECL) provides a foundational framework for quantifying this risk and is calculated as .
- A Credit Event is a specific trigger for a default. The most common events are Bankruptcy, Payment Default, and Debt Restructuring.
- Credit risks within a portfolio are often correlated due to shared systematic factors (like the economy) and contagion effects, meaning defaults are not independent events.
Now that we have established what credit risk is, our next lesson will address how to model it. We will explore the Merton structural model, our first quantitative tool for estimating a company's Probability of Default (PD) by analyzing its capital structure.