Maximize credit card portfolio profit
Credit Card Approval and Profit Optimization (12-month horizon)
You receive 10,000 credit card applications per month. Applicants are categorized into risk bands, each with a specified portfolio share and annual probability of default (PD). The expected credit loss (ECL) for an approved account can be calculated using:
- Exposure at Default (EAD) = 80% of credit limit
- Loss Given Default (LGD) = 90%
For each approved account over a 12-month period, the revenues and costs are:
- Interchange revenue = 1.8% of annual spend
- Expected annual spend per approved account = $6,000
- Variable servicing cost = $24/year
- Acquisition cost = $50 per approved account
- Baseline credit limit = $2,500
The risk bands are defined as follows (share of applicants; PD):
- A: 10%; PD = 0.2%
- B: 20%; PD = 1%
- C: 30%; PD = 3%
- D: 25%; PD = 6%
- E: 15%; PD = 10%
Assume that PDs are for a 12-month period, annual spend is not limited by the baseline credit limit, and other revenue streams (like interest) are to be ignored.
Tasks:
- For each risk band, compute the expected credit loss (ECL) per approved account and the expected profit per approved account over a 12-month horizon.
- Determine an approval cutoff strategy (e.g., approve applicants from bands A through X) that maximizes the total monthly profit for the 10,000 applicants. Report the monthly portfolio ECL and the overall approval rate for your chosen cutoff.
- If credit limits can be adjusted per band (e.g., by halving the limits for bands D and E), re-evaluate and state whether the optimal approval cutoff changes.
- Propose a safe experimental design to validate the chosen decisioning policy before its full implementation.
AI interview prep
Core dilemma
Balancing risk (defaults and associated losses) against revenue generation (spend and interchange fees) to maximize overall portfolio profit, while also considering the operational costs and the need for a robust validation strategy.
3 tradeoffs and 3 stress tests locked
Interview prep
Unlock the interviewer X-Ray rubric, tradeoff matrix, stress tests, clarifying questions, and key concepts.
Maximize credit card portfolio profit
Credit Card Approval and Profit Optimization (12-month horizon)
You receive 10,000 credit card applications per month. Applicants are categorized into risk bands, each with a specified portfolio share and annual probability of default (PD). The expected credit loss (ECL) for an approved account can be calculated using:
- Exposure at Default (EAD) = 80% of credit limit
- Loss Given Default (LGD) = 90%
For each approved account over a 12-month period, the revenues and costs are:
- Interchange revenue = 1.8% of annual spend
- Expected annual spend per approved account = $6,000
- Variable servicing cost = $24/year
- Acquisition cost = $50 per approved account
- Baseline credit limit = $2,500
The risk bands are defined as follows (share of applicants; PD):
- A: 10%; PD = 0.2%
- B: 20%; PD = 1%
- C: 30%; PD = 3%
- D: 25%; PD = 6%
- E: 15%; PD = 10%
Assume that PDs are for a 12-month period, annual spend is not limited by the baseline credit limit, and other revenue streams (like interest) are to be ignored.
Tasks:
- For each risk band, compute the expected credit loss (ECL) per approved account and the expected profit per approved account over a 12-month horizon.
- Determine an approval cutoff strategy (e.g., approve applicants from bands A through X) that maximizes the total monthly profit for the 10,000 applicants. Report the monthly portfolio ECL and the overall approval rate for your chosen cutoff.
- If credit limits can be adjusted per band (e.g., by halving the limits for bands D and E), re-evaluate and state whether the optimal approval cutoff changes.
- Propose a safe experimental design to validate the chosen decisioning policy before its full implementation.