𝐓𝐡𝐞 𝐒𝐡𝐢𝐞𝐥𝐝 𝐨𝐟 𝐑𝐞𝐬𝐢𝐥𝐢𝐞𝐧𝐜𝐞: 𝐂𝐚𝐩𝐢𝐭𝐚𝐥 𝐑𝐞𝐪𝐮𝐢𝐫𝐞𝐦𝐞𝐧𝐭 𝐚𝐠𝐚𝐢𝐧𝐬𝐭 𝐂𝐫𝐞𝐝𝐢𝐭 𝐋𝐨𝐬𝐬𝐞𝐬

𝐓𝐡𝐞 𝐒𝐡𝐢𝐞𝐥𝐝 𝐨𝐟 𝐑𝐞𝐬𝐢𝐥𝐢𝐞𝐧𝐜𝐞: 𝐂𝐚𝐩𝐢𝐭𝐚𝐥 𝐑𝐞𝐪𝐮𝐢𝐫𝐞𝐦𝐞𝐧𝐭 𝐚𝐠𝐚𝐢𝐧𝐬𝐭 𝐂𝐫𝐞𝐝𝐢𝐭 𝐋𝐨𝐬𝐬𝐞𝐬

While deposits and debts fund a bank's operations and lending activities, they also represent liabilities that must be repaid. The nature of the lending business expects a few borrowers to default, and the bank is expected to lose a certain percentage of the money lent. To illustrate in a simple example, let’s assume a bank operates purely on deposits and lends USD 100.00. If 5% of borrowers default, the bank can collect USD 95.00, wherein it needs to repay USD 100.00 to depositors. Assuming the income from operations equals the business expenses, the bank cannot repay the debt, leading to bankruptcy. In contrast, capital represents the bank's funds, providing a cushion to absorb losses before depositors and creditors are at risk. Capital acts as a financial buffer, protecting the bank and its stakeholders from unexpected losses due to loan defaults, market downturns, or other adverse events.

To understand the amount and type of capital the bank should ideally have to safeguard against credit loss, let us understand the types of losses and how banks measure them.

Types of Credit losses and measurement approaches

Credit risk for banks or lending institutions can be classified into an expected, unexpected, and stressed loss. The expected loss is anticipated on a loan or portfolio of loans and is mitigated using underwriting policies and processes, risk-adjusted pricing, and provisions. Unexpected loss is mitigated through regulatory capital, and stressed loss is mitigated through buffer capital.

Article content
Credit Loss Distribution

Expected Credit Loss

Expected Credit Loss accounting for credit risk is based on the loss likely to occur on a loan or portfolio. The accounting method estimates potential future losses on financial assets and recognises those losses in the financial statements. It represents the probability-weighted estimate of the present value of future cash inflow shortfalls from an instrument or portfolio of loans.

ECL is calculated by estimating the forward-looking Probability of default (PD) for each loan and then multiplying it with Loss given default (LGD), the percentage loss expected to occur if the borrower defaults and Exposure at default (EAD), the expected loss for each loan. The sum of loss values for each loan for a portfolio of loans is the expected loss for the entire portfolio.

The probability of default (PD) is the likelihood that a borrower will default on contractual payments during a predetermined period.

Loss Given Default (LGD) is the percentage loss incurred in case a borrower defaults despite the best collection effort, including collateral liquidation, if applicable.

The LGD can be expressed as:

Article content

Exposure at Default (EAD) is the loss of exposure at the time of default. EAD is dynamic and keeps changing as a borrower makes payments in case of a term loan or utilises in case of a revolving facility.

Article content

Approaches for Recognition of Expected Credit Loss

Various approaches and methodologies are used for loss recognition in the banking sector based on the prevailing regulatory guidelines of the respective countries.

1.   Incurred Loss Model:

This traditional approach recognises losses only when they are incurred and there is evidence of a loss event. It is based on historical loss experience and identifies specific loans or assets deemed impaired. Financial institutions in many geographies follow the incurred loss approach, IRAC norms prescribed by the regulator in accounting for losses on loans and other financial assets, which is not aligned with IFRS 9 and Ind AS 109. The main drawbacks associated with the incurred loss recognition model are:

a)    Pro-cyclicality: Recognising credit losses when they are "probable" or have been incurred during economic downturns or periods of financial stress will delay recognising losses until they are already severe, exacerbating the downward spiral of credit conditions and economic activity.

b)    Delayed Recognition of Losses: Books may not adequately reflect the deterioration of credit quality in their financial statements until it is too late, leading to potential underestimation of credit losses and misrepresentation of the true risk profile of the loan portfolio.

c)    Inadequate Risk Management: By only recognising credit losses when incurred, the incurred loss model incentivises engaging in riskier lending practices or insufficient provision for expected losses, which weakens risk management practices and undermines the financial system's stability.

 2.   Expected Loss Model:

The expected loss model is forward-looking and is a key component of the current Expected Credit Loss (ECL) accounting standard, such as IFRS 9. It requires banks to recognise losses based on the expected credit losses over the entire life of a financial instrument. This approach offers several advantages over the traditional incurred loss model:

a)    Forward-Looking: The ECL model takes a forward-looking approach to estimating credit losses, allowing banks to recognise expected losses based on current information and future economic conditions, market trends, and borrower-specific factors for more timely and accurate insights into credit risk.

b)    Timely Recognition of Risks: Unlike the incurred loss model, the ECL model enables banks to recognise expected losses as soon as they become foreseeable. This proactive approach ensures banks promptly anticipate and prepare for potential credit losses, enhancing risk management practices and financial stability.

c)    Improved Decision-Making: The ECL model provides banks with more reliable and actionable information for decision-making purposes, such as credit underwriting, portfolio management, and capital allocation. By aligning credit loss estimates with the underlying risk profile of the loan portfolio, the ECL model enables banks to make more informed decisions and allocate resources more efficiently.

3.   Scenario Analysis and Stress Testing:

Scenario analysis and stress testing help assess the potential impact of adverse economic scenarios on their credit portfolios. By modelling different economic scenarios, banks can evaluate the sensitivity of their expected credit losses to changes in economic conditions and make informed decisions about risk management and capital planning.

 4.   Management Judgment and Experience: 

Management's judgment and experience play a crucial role in loss recognition. Experienced risk managers often contribute qualitative insights into the creditworthiness of borrowers and potential loss scenarios. 

 It is essential to adopt a combination of these approaches, depending on the nature of their portfolios and the regulatory requirements they must adhere to. The choice of methodology often depends on factors such as the type of financial instruments held, the risk profile of the bank, and the prevailing accounting standards. 

Worst Case Credit Loss

The worst-case loss represents a bank's maximum potential loss, considering a worst-case default rate(WCDR) assuming the economy is at its worst at a 99.9% confidence level.

Unexpected Credit Loss

Unexpected Loss (UL) is the difference between worst-case loss and expected loss. The unexpected loss is the expected loss plus the potential adverse volatility or variation in the expected loss, and it is calculated as the standard deviation from the mean at a certain confidence level.

Article content

Where:

Article content

As a default, it is a Bernoulli variable with a binomial distribution.

Unexpected Loss is a formal Risk Measure introduced in the Basel II regulatory reforms. It is used primarily for estimating Risk Capital using internal models and aims to separate the related Expected Loss. Expected losses are provisioned for, and unexpected losses must be explicitly insured against other forms of capital.

Stressed Loss:

Stressed loss refers to a bank's potential losses under severe or stressed economic conditions. Stressed Loss is calculated by subjecting the portfolio to challenging and unlikely scenarios, often as part of stress testing exercises. Banks conduct stress tests that impose adverse conditions on various risk factors (e.g., interest rates, economic indicators) to evaluate the impact on their portfolios.

It's important to note that these calculations are part of a broader risk management framework, and banks may use additional metrics and methodologies to assess and mitigate various risks, including market risk, credit risk, and operational risk. Additionally, the specific calculations and methods may vary based on regulatory requirements and the complexity of the bank's activities.

In the context of risk management for banks, the terms "Expected Loss," "Unexpected Loss," and "Stressed Loss" are often associated with the measurement and management of credit risk. These concepts are commonly used in calculating Regulatory Capital or Economic Capital, the capital a bank sets aside to cover unexpected and extreme case losses.

Regulatory Capital

As per Basel norms, regulatory capital refers to the minimum amount of capital banks must maintain to ensure their financial stability and ability to absorb losses. The regulatory capital framework under Basel III defines three tiers of capital:

Tier 1 Capital:

Common Equity Tier 1 (CET1): CET1 includes common equity instruments like commo shares and retained earnings.

Additional Tier 1 (AT1): Instruments that may include non-cumulative preferred stock and other qualifying instruments.

Tier 2 Capital:

Tier 2 capital includes subordinated debt and other qualifying instruments that provide additional loss absorption capacity.

Total Capital:

Tier 1 and Tier 2 capital constitute the total regulatory capital.

Regulatory capital requirements are a percentage of a bank's risk-weighted assets (RWA). The risk-weighted assets are calculated by assigning different risk weights to various types of assets based on their credit risk. This is intended to ensure that banks hold more capital for riskier assets.

Basel III introduced more stringent capital requirements compared to its predecessor, Basel II, to enhance the resilience of banks in the face of financial stress and economic downturns. It also introduced additional capital buffers, such as the Capital Conservation Buffer (CCB) and the Countercyclical Capital Buffer (CCyB), to strengthen the financial system further. Regulatory capital can also be computed through IRB Models.

Economic Capital:

Economic capital, sometimes called risk capital, represents the amount of capital a bank needs to hold to cover its various risks, including credit, market, and operational risks. Unlike regulatory capital, economic capital is not mandated by regulators; instead, it is an internal measure the bank uses.

Banks use economic capital to ensure sufficient capital to cover unexpected losses beyond what regulators require. It provides a more comprehensive view of a bank's risk profile and allows for a more nuanced understanding of the potential impact of various risks.

Economic capital is often divided by risk type, with separate allocations for credit, market, and operational risks. Each component reflects the potential losses associated with a specific risk category.

Economic capital is calculated using internal risk models and assessments. These models consider the bank's risk appetite, risk tolerance, and the specific characteristics of its portfolio.

Economic capital is more flexible and can be tailored to the bank's risk profile and business strategy. It provides management with insights into the level of risk-taking that aligns with the institution's goals.

In summary, while regulatory capital is a minimum requirement set by regulators to ensure the financial system's stability, economic capital is an internal measure that allows banks to understand and manage their risks more comprehensively. Both concepts play crucial roles in a bank's overall risk management framework.

We'll go into more detail in the subsequent articles on regulatory norms and requirements in the following articles.

_____________________________________________________________________________________

Acknowledgement: I acknowledge the support of Sukanta Swain , CQF , reviewing and sharing his input on this article.

Disclaimer: The opinion shared through the article is not professional financial advice. Consulting a financial advisor about your particular circumstances is advisable.



Hello sir, very helpful and insightful article. Waiting for your next article on regulatory capital requirements and RWA

Like
Reply

Excellent insights! It would have been interesting to get more details on the USD 100.0 that the bank is lending (maturity, products, etc) to understand how its interest rate policy should be determined in order to cover for the costs and risks and post a profit. Thanks

Hello. Thanks for this very good summary. There is just a little detail in your post which is confusing me. In the graph in the beginning you show that Regulatory Capital is meant to cover for unexpected loss, up to the 99.9% confidence level. Beyond that, you say, stressed loss should be further covered by Economic Capital (more exactly, the piece of Economic Capital in excess of Regulatory Capital). You also mention that "Regulatory Capital can also be computed by IRB models". This implies that the graph should be true under both Standardized and IRB frameworks. However, the RWA approach for Regulatory Capital hardly captures a 99.9% scenario (at least under the Standardized method). From my experience, I would rather say that Regulatory Capital is supposed to cover a portion of the unexpected loss while stress testing models are used to compute Economic Capital which will cover the remainder of this unexpected loss, up to a (internally) chosen confidence level which happens to be 99.9% in most cases. In other words, my point is: if you calibrate your Economic Capital model to the 99.9th percentile, you don't cover for losses beyond that point (what you called stressed losses). Happy to have your views. Thanks

Like
Reply

Thank you for the summary. One reason that recovery rate might be a better variable to start from compared to LGD is that the discounting becomes simpler. For recovery rate you simply discount all recoverable cash flows. A common mistake seen is that people discount the losses in LGD, which instead makes the loss too small. A long workout period should increase LGD, not reduce it.

To view or add a comment, sign in

Others also viewed

Explore content categories