Which framework for a risk management case?

Hi,

I have worked for 5 years in the financial services management consulting(incl. risk management).

I would suggest the standard risk management framework:

EAD x PD x LGD = Expected Loss

EAD is the predicted amount of loss a bank may be exposed to when a debtor defaults on a loan

PD represents as a percentage the likelihood of default.

LGD represents the amount unrecovered by the lender after selling the underlying asset if a borrower defaults on a loan

Thus you can significantly improve your risk profile by optimizing each metric. A few obvious ideas:

EAD - use more data sources to establish a better understanding of how much money you can afford to lend to each particular borrower

PD - improve the quality of the borrowers via more strict scoring systems/ underwriting

LGD - improve the value of collateral that you accept for the loans, improve loan-to-value ratio

I suggest that you should also study other important risk metrics: RAROC, Basel capital ratios.

Hope this helps!

Anton

(editiert)

Hi!

1. Clarify the objective (e.g., minimize risk of bank insolvency due to loan defaults)

2. Define/operationalize this objective (e.g., [loan default rate * loan volume] should not exceed x% of the bank's equity (this is just an indicative example; it needs to be clarified with the interviewer! Never just assume something!)

3. Once the definition is clear, disaggregate each element of the definition (loan default rate, loan volume, bank equity) into its conceptual components (use driver trees)

4. Discuss how you can influence each element into the desired direction (i.e., how to decrease loan default rate? Is it feasible to decrease the overall loan volume? can we beef up the equity? What is the priority lever here (probably the first one!))

This is how you think through issues. Be it Risk or anything else. This will impress an interviewer a million times more than if a candidates comes up with a "bucket framework" which outlines some areas to look into.

Cheers, Sidi

(editiert)

Hi,

This highlights the problem that so many candidates run into when they think that using generic frameworks will allow them to solve business problems. This is fundamentally the wrong approach.

At best, these frameworks could be used as a mental checklist to ensure you are covering all the potential angles from which a problem can be produced.

I would recommend that you start with the client's objective and then think about what are the drivers that will allow the client to achieve their objective.

For the specific problem that you are struggling to solve my tip would be:

1. Establish the target. In this case, it could be to reduce the NPL (non-performing loans %)

2. Understand and breakdown the drivers of loan non-repayment or NPL metric

3. Build a hypothesis on what would be the most likely cause in this specific case given the situation/complication provided in the case prompt and prioritize

It would be great if you can share your proposed approach and share it here. I am happy to discuss this with you since I have experience in the lending business.

Hi there,

And here we run into the inevitable problem: You've thought all along that just memorizing frameworks will let you ace the case...WRONG!

You need to learn how to think. Frameworks are a good guide. They are not the end all be all.

When you learn a framework, you're meant to focus on the "ah ha! THIS is how they thoght about THIS problem. They want to look at THIS because of THIS reason". Rather than getting the takeaway: "For x question, I can use y case"

_____________________________________________________

I'm not answering this question - I want you to think through this and propose something.

As a guide:

1) What matters? What does the bank care about? (Objective)

2) What are the areas we might want to look at? What might help us understand the landscape better? What might drive us towards an answer/resolution? (Hypothesis)

Dear A,

I used to work for boutique consulting company called zeb, specialized only on financial services. And I also have plenty of cases, including the risk cases.

If you need any help to prepare for specific interview, feel free to reach out.

Best of luck,

André

Hello!

It´s good you ask this since it surfaces a big problem when preparing for MBB.

Memorizing classical fremworks it´s only a guarantee to fail.

For sure it´s useuful to know about them, particularly when you start your prep. However, you will never gain an offer by leveraging pre-stablished frameworks. It would be too easy, right?

Leverage them only to build your own customized analisys (analysis is not framework!!!), so you can demostrate that you are unique, creative, think out of the box, bring new ideas to the table... those are the attributes to be proven!

PM me if you want to discuss further how a coach could help.

Hope it helps!

Cheers,

Clara

Hi,

Searching for frameworks is not always an optimal solution/approach. Often times you can just use a more common sense approach to breakdown a problem

Lets start with the original question and I can walk you through one approach -

"Bank’s clients are not paying back their loans. What the bank could do?”

Objective - I will assume the bank wants to stem losses from this portfolio in the near term and improve performance in the longer term

Short term

1. If unsecured - Take a hit on the P&L statement - loan write offs where applicable to ensure investors are aware of the losses
2. If it is a secured loan then re-possess assets from the debtors and sell them to recover losses (e.g., home foreclosures)
3. Stop issuing new loans immediately for the impacted portfolio
4. Cut the line of credit (where applicable) to customers with a similar risk profile (e.g., lower credit limit on credit cards)

Long term - Process improvement to ensure this doesn’t happen again

1. Re-evaluate criteria for risk assessment - update model to include new risk factors that may have affected outcomes (e.g., overreliance on tourism industry)
2. Use newer forms of assessment - e.g., partner with platforms that incorporate data you don’t have access to (e.g., A.I tools that help detect pattern recognition/predict outcomes)
3. Stop loans for this segment completely (e.g., low income housing)

Solution here will depend on risk appetite and level of resources available

As you can see this was not framework driven, largely logical approach to a problem like most case examples

All the best,

Udayan

Hello

Risk management cases can be tackled using 3 different mandates of this function.

• Regulatory compliance / regulatory reporting (risk departmenet has hundreds of regulatory constraints to comply with) But be carefull, regulation is only one part of the risk management picture !
• Independent control. Risk is defined as the second line of defense in an internal control set-up. Thus, risk management has to control what first line of defense (e.g. business ) is doing
• Business support. Risk management also works on high value added / business processes such as the granting process. A good credit risk departmenent has good tools / processes allowing fast credit granting approval.

Also, it's important to know that there are different risk types managed by the risk department.

• Credit risk
• Market risk
• Operational risk
• Liquidity risk
• Climate risk (new)

Bottom line, a potential framework could be a matrix mandates x risk types but it depends on the case of course ...

Regarding your specific question, you have to build a tailor made framework and do not go towards EAD x PD x LGD because it's too complex and too indirect. I would suggest a very simple & direct approach

1. Recovery process (= how to minimize your loss leveraging collateral or guarantees provided by clients ?)
2. Review risk appetite (e.g. ensure that your risk policy is well defined and not too agressive vs. your pricing policy, ...)

Do not hesitate to contact me for more information on risk management, I have conducted tens of projects on those topics.

Best,

David

(editiert)