A well-established regional retail player with a spotless 10-year repayment record approached the bank for a ₹50 crore working capital loan. While the track was clean, there were subtle warning signs —e-comm players were investing heavily and hyperlocal services were gaining traction, sales were stagnant last year and some stretch in payment cycle was visible. These issues seemed marginal, and nothing appeared glaringly out of order.
The Credit Risk Officer reasoned: Hasn’t the company survived competition for a decade? Doesn’t its track record prove it can navigate these challenges? Confident in the company’s reliability and past performance, reassured by financial ratios and satisfied with the recent plans for starting home delivery, the officer approved the loan.
However, the decision proved costly. The apparent risks—stagnant sales, stretched payment cycles, and rising competition—were downplayed while even the plans on drawing board provided high comfort. The officer had fallen into Confirmation Bias trap, focusing on information that supported pre-existing beliefs while underplaying contradictory evidence.
We are all Biased!
Bias is an inherent human tendency, and recognising and minimising it requires deliberate effort. No matter how advanced the models or how experienced the analysts, decisions are inevitably influenced by personal experiences and context. The challenge lies in identifying when this reliance on past-experience crosses the line into bias.
In lending there’s a significant focus on financial modeling and developing complex risk metrics, but one critical aspect remains largely overlooked: idiosyncratic cognitive biases. These biases could arise from individuals handling credit risk function or from a systemic weakness owing to culture and processes. While Credit Committees help mitigate these biases by bringing multiple viewpoints, they too often fall victim to collective bias, influenced by the authority or reputation of those endorsing a decision.
The first step is acknowledging that these biases exist. We look at 7 scenarios to illustrate most common cognitive biases and fallacies that tend to impact our credit judgement. (Note: I use Credit Officer in these scenarios only for ease to represent both individual and systemic biases within the bank.)
Scenario 1: Fallacy of Composition (What’s true for part is true for whole)
A few regions in a large retail chain are showing rapid revenue growth. The company applies for a loan to expand its operations nationwide.
The officer approves a ₹50 crore loan seeing the performance as the select regions are doing very well and therefore the company would have thought through its national expansion plan.
What went wrong?
The officer overlooked that the retail chain is performing well only in specific high-demand areas, while the broader market is saturated and highly competitive. Some regions were showing signs of stagnancy for other players. The revenue growth in only a few growth regions is not indicative of how the entire chain will perform across diverse regions with different consumer behaviors and competition levels.
Bias/Fallacy at Play:
Fallacy of Composition: Assuming some parts (a few regions) doing well can conclusively indicate that the whole (national expansion) will also succeed.
Scenario 2: Fallacy of Division (What’s true for whole is true for part)
The Indian pharmaceutical industry reports 9% growth, driven by increasing demand for branded generics and exports of bulk drugs. A mid-sized Indian pharmaceutical company that manufactures APIs (active pharmaceutical ingredients) applies for a ₹50 crore loan to expand capacity.
The officer approves the loan, giving due weight to the recent results of other pharma players and as the sector is seeing robust growth in exports and domestic demand for generics.
What went wrong?
While the pharmaceutical industry was growing, it was primarily driven by finished formulations and exports of high-margin branded generics. The borrower manufactures APIs with low economies of scale, where it faces stiff competition from Chinese manufacturers, which dominate the global API market. Additionally, raw material costs for APIs have been volatile due to disruptions in supply chains. The officer assumed all parts of the pharmaceutical sector will grow, missing the entity specific challenges.
Bias/Fallacy at Play:
Fallacy of Division: Assuming that the overall growth of the pharmaceutical industry (driven by branded generics and formulations) applies equally to smaller API manufacturers, even though the company faced distinct competitive and supply chain challenges.
Scenario 3: Hindsight Bias (It was so obvious!)
A large Indian retail company defaults on a ₹300 crore loan after a rapid expansion into tier-2 and tier-3 cities.
A review committee is formed led by a senior officer, who wasn’t part of the approving committee. In the review discussion, the senior officer claims that the company’s strategy to open stores in smaller cities was a clear red flag from the beginning considering the viability of such stores, and the company was bound to fail. The challenges seemed so obvious.
What went wrong?
At the time of approval, the company’s expansion plan was supported by strong growth in the retail sector and rising demand in smaller cities. The company had already adjusted store formats. The real issue was around proper planning for supply chain complexities and over-reliance on premium segment products in tier-2 and tier-3 cities. This execution failure wasn’t obvious at the time of assessment but with the benefit of hindsight, this was highlighted as a glaring miss..
Bias/Fallacy at Play:
Hindsight Bias: Claiming that the company’s expansion and failure were inevitable, even though these factors are clear only in hindsight.
Scenario 4: Availability Heuristic (Over reliance on the most recent information)
A renewable energy company applies for a ₹150 crore loan to develop a solar power park in Rajasthan.
The officer recalls a recent default in a wind energy project in another state, where significant delays in land acquisition and regulatory approvals led to the project’s failure, resulting in losses for the lenders. The officer assumes that all renewable energy projects face similar obstacles in the current regulatory environment.
What went wrong?
The officer fails to account for the fact that unlike the wind project, solar energy parks in Rajasthan had a faster regulatory process, more predictable energy output, and fewer land acquisition issues. The solar energy sector has received strong government support, including recent policies like the PM-KUSUM scheme. The project also had 25% Viability Gap Funding (VGF), significantly reducing financial risk. The officer allowed the vivid memory of the wind energy project’s failure to overshadow the strong structural and project specific mitigants.
Bias/Fallacy at Play:
Availability Heuristic: The recency effect leading to over-weight on failure-risk based on wind energy project despite strong project specific mitigants of the current case. (If you have noticed ‘knee-jerk’ reactions of banks to smallest of the events, you have seen this at play)
Scenario 5: Sunk Cost Fallacy (Having invested so much, just a little more will take it through)
A construction project had already consumed ₹100 crore and the borrower asks for an additional ₹25 crores to complete the project. The project is slightly behind schedule and as a result the inventory movement is slow.
The officer approves additional ₹25 crore funding, citing the bank’s significant prior investment and this incremental exposure will help take the project through. The premise is that supporting this last leg of completion will compensate for the initial costs.
What went wrong?
Rising material costs (up 15% in the last 6 months) and delays due to labour shortages have pushed the project’s ROI from an expected 20% to 10%. Costs are only rising upwards and were likely to make returns difficult for the developer. By ignoring the current costs and only focusing on sunk costs, the officer underestimated the risk and future viability of the project.
Bias/Fallacy at Play:
Sunk Cost Fallacy: Continuing investment based on past costs, rather than re-evaluating the project’s current and future potential or risks.
Scenario 6: Overconfidence Bias (It is so obvious, I know it for sure)
A fast growing logistics company that specializes in last-mile delivery applies for a ₹150 crore loan to invest in a new fleet of electric vehicles. The company has grown rapidly over the past three years using its existing CNG and ICE fleet due to the surge in e-commerce and presents ambitious plans to expand its network across the country.
The credit officer, buoyed by the company’s rapid growth, order book, and trend of increasing EV adoption, approves the loan. The officer has confidence in the aggressive projections given that company’s expertise in delivery services and rising demand.
What went wrong?
The officer fails to account for difference in infrastructure augmentation that will be required for large electric vehicle fleet, including charging infrastructure and battery maintenance. Additionally, the useful life and resale value of these vehicles is still not established, impacting the ownership IRR. The company faced unexpected delays in rolling out its new fleet due to limited production capacity of EV cargo vehicles. As a result, the company’s expansion moved at a much slower pace than what the projections indicated.
Bias/Fallacy at Play:
Overconfidence Bias: The overconfidence in the company’s rapid growth trajectory and industry trend leads to an underestimation of the operational complexities, resulting in an overly optimistic decision.
Scenario 7: Confirmation Bias (Over-weigh the data that aligns with experience)
A long-standing borrower with solid repayment track in the textile industry, with a ₹100 crore turnover, applies for additional working capital. Recent reports indicated that the company has faced a decline in exports due to weaker demand from Europe and is also seeing stretch in its cycle.
The officer approves the loan, relying on the borrower’s 10 year track record, customer relationships and considering these temporary challenges will resolve as the demand seems to be recovering.
What went wrong?
The officer ignored key red flags: increased in receivables period 90 to 150 days, indicating potential cash flow issues, underplaying it as company’s tactical move to retain sales. The officer considers the recent 25% decline in exports as a short-term fluctuation. The demand recovery in key markets was very slow. Competition from Bangladesh and Vietnam in the textile export market was squeezing margins and profitability was under threat.
Bias/Fallacy at Play:
Confirmation Bias: Overweighing past experiences (10 years of solid repayment history) and discounting recent negative indicators (declining exports, competition, slow recovery) that contradict the officer’s prior belief in the borrower’s reliability.
End Note
The key to minimising these biases is identifying their source. At the individual level, training and increased awareness can help, but for systemic biases, banks need to dive into cultural and process-based issues- simple governance structures or training won’t suffice. Acknowledging that these biases exist and can impact credit decisions is the first step toward overcoming them.
Disclaimer: The opinions expressed here are those of the author and does not reflect the views of FrankBanker.com