Integration of climate risk in internal credit models

« The bank’s core business remains credit. It represents an important part of the income, but it also generates a very high cost in case of non-repayment. Therefore, there is a risk related to the non-repayment of the loan granted: this is the credit risk. Credit risk is defined as the risk of loss on a debt instrument resulting from the failure of the borrower to make required payments to the debtor such as a bank. We generally distinguish two types of credit risk:

• Default risk, which arises when the borrower is not able to settle an obligation in full when due. An example is a mortgage loan where the borrower defaults on his payment.
• Downgrading risk, which concerns debt securities. This is the risk of a change in the debt market value caused by the increase in the obligor’s credit risk.

Against this background, the Basel Committee on Banking Supervision (BCBS) requires banks to have sufficient capital and reserves to absorb losses in the event of default. Historically, to calculate the probability of default, banks use traditional risk drivers. However, today with climate change, new drivers have appeared in the calculation of the probability of default in internal models: these are the climate risk factors.

Climate risk, driven by floods, droughts, or hurricanes, is the exposure of economic agents to climate impacts and their repercussions on their activities. These climatic phenomena are not new in the history of humanity, but they have increased in recent years to the point of being the focus of all concerns.

There are three main types of climate risks:

• Physical risk can be divided into two types: « acute » and « chronic ». The former corresponds to the direct losses associated with the damage caused by climatic hazards on economic actors. These effects can be direct as damage to property or reduction in productivity, or indirect, such as the disruption of supply chains. The chronic physical risk corresponds to losses caused by a progressive change in the climate.
• Transition risk, which corresponds to the economic consequences of implementing a low-carbon economic model. It may result, for example, from the rapid adoption of climate policies unfavorable to certain sectors of activity (fossil fuels, transport, etc.), from the acceleration of technological progress, or from changes in customer preferences.
• Legal risk, which corresponds to the compensation that a legal entity could be required to pay in the event that it is deemed responsible for damage caused by the consequences of climate change.

In this study, we will first conduct a literature review to understand what has already been done around the integration of climate risk into internal credit models. This literature review is a critical assessment of research developments in the area of integrating climate risk into the calculation of the probability of default. It will allow us to better understand the issues at stake. Then, in the second step, we will integrate the physical risk into the calculation of the probability of default. We will conclude with an analysis of our modeling results and a final conclusion. »