Financial firms struggling to quantify climate-related risk

Financial firms struggling to quantify climate-related risk

Financial firms are struggling to quantify climate-related risk, according to the sector watchdog the Prudential Regulatory Authority (PRA). After analysing audit reports for 2021-2022, it wrote to firms in October highlighting key areas of note. Firms have taken action to enhance governance, data and risk assessments relating to climate change. However, auditors failed to identify specific risks…

Financial firms are struggling to quantify climate-related risk, according to the sector watchdog the Prudential Regulatory Authority (PRA).

After analysing audit reports for 2021-2022, it wrote to firms in October highlighting key areas of note. Firms have taken action to enhance governance, data and risk assessments relating to climate change. However, auditors failed to identify specific risks of material misstatements in financial reporting.

Lack of data

In fact, lack of reliable data was a major concern. “Auditors noted a lack of reasonable and supportable data available to management about their exposure to climate risk, and available to auditors to substantiate the impact of climate risks on the balance sheets,” Victoria Saporta, executive director of PRA’s prudential policy. “Where data was available, firms’ control environments around the quality of new data sources and use of proxies were immature.”

Some firms had made good progress in identifying the relevant risks so that they can be accurately reflected on balance sheets. PRA said it was important to link better risk management practices with accounting disclosures.

Increase quantitative analysis

PRA wanted to see more quantitative analysis of climate-related risk. Currently, much of the data available to audit committees and auditors is qualitative.

It wanted to see firms to embed quantitative analysis on the impact of climate risk on balance sheet valuations. This analysis should be reported to the audit committee to support decision-making. In addition, PRA recommended firms set up management information on their progress towards creating more effective reporting models in this area.

PRA recommended that firms develop “…a centralised process to source, manage, and enhance data needed to factor climate risk into balance sheet valuations.” Increasing automated data collection to improve controls over data would help, it said.

Model risk

The regulator also spelt out weaknesses in firms’ modelling for risk.  In particular, organisations were often too slow to react to rapid shifts in the economy.

“We continue to regard it as essential that firms develop capabilities to perform more comprehensive economic sensitivity analysis more quickly, and improve their use of timely, granular, and comparable peer benchmarking data, to support robust governance,” it said.

Read the letter here





← Previous

Cyberattack preparations may not be robust enough
Cyberattack preparations may not be robust enough to withstand real attacks at many organisations, according to…






Next →

<strong>Businesses failing the digital trust test</strong>
Businesses are failing the digital trust test as their ability to protect customers lags expectations,…

7 OCTOBER 2026

Key elements of a mature programme risk capability

Hosted by Vinay Shrivastava. Vinay's presentation will cover lessons he has learned over the course of his career. These insights will be shared using the principles of ISO31000 risk management.

Find out more

8 OCTOBER 2026

Pre-deployment agentic risk management

Delivered by Adam Grainger, the outcome of this session is to be able to support an agentic implementation with effective pre-deployment risk identification and mitigation.

Find out more

14 OCTOBER 2026

AI Transforming Enterprise Risk Management Activities

Risk management has changed considerably over the years, with new frameworks, standards and expectations shaping the way organisations manage risk. However, many of the day-to-day activities within Enterprise Risk Management (ERM) functions have remained much the same.

Find out more

Advertisement