Controversies and ESG event risk

Most of an ESG rating is built from slow-moving information: policies, targets and yearly data. But some of the biggest ESG risks arrive suddenly. A gas leak, a deadly protest, a regulator's order or a short-seller's report can change a company's prospects in a day.

Rating providers track these events separately as controversies. A controversy score reflects how serious an event is, how many people or how much of the environment it affects, how long it lasts and how well the company responds.

For analysts, controversies are a reminder that ESG risk is partly event risk. The question is not only "how good are this company's policies?" but also "what could go badly wrong, and what would it cost?"

Assessing a controversy

• Severity: harm to people, the environment or investors.
• Scale: how many people, sites or markets are affected.
• Duration and recurrence: a one-off event or a repeated pattern.
• Company response: speed, openness, compensation and changes to systems.
• Legal status: allegation, investigation, court case, settlement or final finding.

Providers express this in different ways. Some use levels from 1 (low) to 5 (severe); some use colour flags; some use a 0 to 100 score where lower means worse. India's Nifty ESG indices, for example, now exclude companies with a controversy score below 70. Many investors also screen for breaches of the UN Global Compact principles.

Types of ESG events

Four Indian examples

LG Polymers, Visakhapatnam (2020). On 7 May 2020, styrene gas leaked from LG Polymers India's plant. At least 11 people died and about 1,000 were exposed. The National Green Tribunal held the company to have "absolute liability" and ordered an interim deposit of ₹50 crore for compensation and environmental restoration.

Sterlite Copper, Thoothukudi (2018). Protests against alleged pollution from Vedanta's copper smelter ended in police firing in May 2018, in which 13 people were killed. The plant has been shut since May 2018. On 29 February 2024, the Supreme Court dismissed Vedanta's plea against the closure, saying the health and welfare of local residents was of "utmost concern".

Paytm Payments Bank (2024). On 31 January 2024, RBI barred Paytm Payments Bank from taking new deposits, credit transactions and top-ups after 29 February 2024, citing "persistent non-compliances and continued material supervisory concerns". RBI later extended the deadline to 15 March 2024.

Adani group and Hindenburg (2023 to 2025). In January 2023, short-seller Hindenburg Research accused the Adani group of fraud, which the group denied. The combined market value of Adani group companies fell from about ₹19.2 lakh crore before the report to about ₹6.8 lakh crore on 27 February 2023. On 18 September 2025, SEBI issued two orders finding that the related-party and fraud allegations it had examined were not established, and it imposed no penalties.

Lessons from these cases

1. Markets move faster than ratings. Prices react within hours; ratings and legal processes take months or years.
2. Allegations are not findings. The Adani case shows how a final regulatory outcome can differ from the initial claims.
3. Some events end the business. For Sterlite, the cost was not a fine but the loss of its licence to operate the plant.
4. Regulatory risk is an ESG risk. In financial services, persistent compliance failures can shut down a business line.
5. Process safety matters. Industrial accidents bring human costs, legal liability and long-term damage to trust.

Building event risk into a valuation (illustrative)

A hypothetical company is worth ₹10,000 crore before considering a pending ESG issue. The analyst sees two possible outcomes over the next year:

• A 10% chance that a plant is shut, destroying ₹2,000 crore of value: expected loss = 10% × 2,000 = ₹200 crore
• A 30% chance of a fine and clean-up costs of ₹200 crore: expected loss = 30% × 200 = ₹60 crore
• Total expected loss = 200 + 60 = ₹260 crore, or 2.6% of value
• Adjusted value = 10,000 minus 260 = ₹9,740 crore

Modelling outcomes this way is usually clearer than adding an arbitrary premium to the discount rate.