Case 084Operational risk and loss eventsHard

A bank's branches sold insurance-linked savings plans to elderly fixed deposit customers, and 4,000 complaints have arrived. Estimate the remediation cost, add an illustrative penalty, and trace the root cause.

The situation

Over two years, Dravika Bank's branches sold a five-year insurance-linked savings plan to customers whose fixed deposits were maturing. Many buyers were over 65. The plan locked money in for five years and charged heavy surrender penalties. Branch staff earned about three times the incentive on the plan that they earned on renewing a fixed deposit.

4,000 complaints have arrived. The complaints team estimates the average premium paid at Rs 3 lakh, and the bank expects to refund premiums with interest at 8% a year, compounded, for the two years the money was tied up. File review will cost about Rs 5,000 a case. The regulator may also impose a penalty; use Rs 10 crore to Rs 50 crore as an illustrative range, not a forecast. The bank sold 18,000 such plans to customers over 60.

Your task

What will remediation cost, what is the realistic total including a penalty, and what is the root cause the bank has to fix?

Quick check

Before any penalty, roughly what does remediating the 4,000 complainants cost?

Worked solution

Try it on paper, then open one step at a time.

30-second answerThe answer to give first

Remediation for the 4,000 complainants is about Rs 140 crore, and Rs 152 crore to Rs 192 crore with review costs and an illustrative penalty. The bigger risk is the other 14,000 similar sales: each further 1,000 found adds about Rs 35 crore. The root cause is the incentive, which made the unsuitable sale the best-paid one, with product approval, point-of-sale checks and complaint monitoring all failing to catch it.

Step 1How do you size the remediation?

Remediation puts the customer back where they would have been without the sale. If a shop sells you a faulty fridge, a fair fix is your money back plus what you lost in the meantime, not just a discount on the next one. Each customer gets Rs 3 lakh back plus two years of 8% interest, Rs 3 lakh times 1.08 squared, about Rs 3.50 lakh; across 4,000 customers that is Rs 120 crore of premium and Rs 20.0 crore of interest, about Rs 140 crore. Add Rs 2 crore of file review, at Rs 5,000 a case.

Rs croreLowHigh
Premium refunds, 4,000 x Rs 3 lakh120.0120.0
Interest at 8% compounded for two years20.020.0
File review at Rs 5,000 a case2.02.0
Regulatory penalty, illustrative10.050.0
Total152.0192.0
Refunds of Rs 120 crore, interest of Rs 20.0 crore and review costs of Rs 2 crore come to about Rs 142 crore, and an illustrative Rs 10 crore to Rs 50 crore penalty lifts the total to Rs 152 crore to Rs 192 crore.
What it costs: remediation first, penalty on topRefunds 120interest 20review 2penalty 10 to 50, illustrative1521920Look-back risk: 4,000 complaints out of 18,000 similar sales.Each further 1,000 customers found adds about Rs 35 crore.
Dravika's cost stacks from Rs 120 crore of refunds and Rs 20 crore of interest to a total of Rs 152 crore to Rs 192 crore with review and an illustrative penalty, and every further 1,000 mis-sold customers found adds about Rs 35 crore.
Step 2Why is 4,000 probably not the final number?

Complaints come from customers who noticed, understood and chose to complain. The bank sold 18,000 of these plans to over-60 customers, so the 4,000 complaints are a floor, and a regulator will expect a look-back reviewA proactive review of every similar sale, not only those that drew complaints, to find and repay customers who were harmed but did not complain. of all of them. If a file review finds another 3,000 unsuitable sales, the bill rises by about Rs 105 crore. That uncertainty belongs in the provision discussion with finance now, as a range, not after the review.

Step 3What is the root cause, as against the symptoms?

Follow the chain back from the complaints and stop at the first link that made the others likely. The root cause is the incentive: staff earned three times as much for moving a maturing deposit into a locked-in plan, so the unsuitable sale was the best-paid sale. Product approval should have limited the plan to customers who could lock money away for five years, the sales process should have tested suitability, and complaint monitoring should have spotted the pattern early. Each was a control, and each failed.

Four failures in a row, each one letting the next through1 PayInsurance plan paid staff3x the incentive of an FD2 ApproveProduct signed off withno age or target market3 SellPitched at FD maturity,no suitability check4 WatchComplaints logged, nottrended or escalated4,000 complaints from elderly FD customersabout Rs 140 crore to put them back where they wereRoot cause sits at step 1: the incentive made the wrong sale the best-paid sale.Steps 2 to 4 were the controls that should have caught it, and each one failed.
Dravika's losses trace back through four failures: an incentive paying three times more for the plan, a product approved with no age limit, sales with no suitability check and complaints never trended, which together produced 4,000 complaints and about Rs 140 crore of remediation.

Close with fixes matched to each link, and say which comes first. Change pay so incentives on third-party products do not exceed deposit renewals and can be clawed back if a sale is found unsuitable. Re-approve the product with a defined target market. Add a cooling-off call for customers over 60. Trend complaints by product and branch monthly. Fixing the controls without fixing pay leaves the pressure that produced the failure in place.

Where candidates lose it

The common miss is sizing only the complaints received. A risk interviewer wants to hear that complaints are the visible part, and that a look-back across all similar sales is what the regulator, and the provision, will be based on.

The second is naming staff misconduct as the root cause. Individual staff behaved as the incentive paid them to; blaming them without changing pay is how the same loss recurs under a different product.

What the interviewer asks next

  • How would you record this in the bank's operational loss data, and in which event type?
  • Should the provision be booked at the low end, the midpoint or the high end? Why?
  • What key risk indicator would have flagged this a year earlier?

Company names and figures are illustrative.