By Giorgio Baldassarri, Global Head of Analytic Development Group, S&P Global Market Intelligence
Can you please tell the Risk Insights readers a little bit about yourself, your experiences, and the nature of your current professional focus?
I lead the Analytic Development Group at S&P Global Market Intelligence. My team employs state-of-the-art statistical and machine learning techniques to develop quantitative models that assess credit risk and supply chain risks of corporations and small- and medium-sized enterprises globally.
What, for you, are the benefits of attending a conference like Risk EMEA 2019, and what can attendees expect to learn from your session?
In general, attendance at Risk EMEA enables people to gain useful insights and an up-to-date view on major risk perspectives and trends in the financial industry. My session focuses on a modelling framework that helps link climate-related transition risks with credit risks of corporations. This is a very hot topic (no pun intended), as it is a major focus for financial regulators worldwide to better understand risks, as well as opportunities, and how these may affect banks’ lending books.
In your opinion, why are climate-linked scenarios and credit modelling a critical issue for 2019/2020?
Reducing carbon emissions worldwide will require an unprecedented and concerted effort that involves governments, corporations, and even individuals. The financial impact of climate-related transition risks on companies in brown sectors will intensify in the coming decades, as most countries will increase current carbon tax levels, or enact new carbon pricing policies. Financial regulators want to understand the risks and opportunities this transition may pose to the stability of the global economy, and are planning to include stress scenarios in banks’ annual stress-testing exercises as early as this year.
How could the introduction of climate-related scenarios into annual stress testing impact financial institutions?
There is an intense debate among regulators on whether to introduce a brown-penalizing factor that increases capital requirements of banks with material exposures to high carbon-intensive industries, or a green-supporting factor to facilitate investments in green exposures. The main challenge in facilitating the transition to a low-carbon economy will be to strike a balance between redirecting investments towards greener investments and ensuring sound risk management practices are in place to account for regulatory and policy risk that will necessarily affect carbon-emissions produced in the next decades.
Without giving too much away, how can firms look to effectively manage climate-linked scenarios and credit risk modelling?
Scenario analysis remains a critical pillar in any sound risk management practice. However, to properly manage climate-linked scenarios, the risk perspective needs to shift towards much longer time horizons (from the traditional two to three year timeframe to several decades). In addition, a sound approach needs to leverage a robust set of data. We have developed two complementary methodologies that help gauge the impact of climate-related scenarios on credit risk: a bottom-up/fundamentals-driven approach and a top-down/market-driven company perspective. These models are complementary and can help enrich the toolkit of risk managers.
How do you see credit risk modelling evolving over the next six to 12 months?
The advent of the “Big Data era” allows banks and other financial institutions to access big and alternative datasets that can offer a complementary view on credit risk (for example, digital fingerprints and sentiment analysis). The application of machine learning techniques supports automating the analytical process and distilling insights from noisy data. However, a successful implementation of machine learning techniques will require combining the power of unsupervised techniques with a sound validation based on common sense and an a-priori intuition or, in the absence of it, with interpretation of model outputs. This can help financial practitioners avoid (to put it in Nassim Taleb’s words being “fooled by randomness”.