Abstract
The global financial crisis of 2007–2009 placed a spotlight on financial risks, particularly credit risk, which played a central role in triggering the crisis. The collapse of mortgage-backed securities, driven by widespread defaults, exposed vulnerabilities in the financial system, leading to severe economic disruptions worldwide. This crisis underscored the importance of understanding, quantifying, and managing credit risk to prevent similar systemic failures in the future. In addition to traditional financial risks, climate change has emerged as one of the most pressing global challenges, posing unprecedented threats to the stability of socioeconomic and financial systems. The increasing frequency and severity of climate-related events underscore the need to identify, quantify, and mitigate the financial risks associated with climate change. Addressing these risks is crucial for ensuring long-term financial stability, especially in emerging markets, which are often more susceptible to both climate impacts and financial volatility. In line with these concerns, the thesis focuses on credit and climate change in emerging markets. The first empirical chapter examines the macroeconomic determinants of credit risk in Brazil, Russia, India, and China (BRICS countries). To analyse this, a Markov Switching Model (MSM) is employed, which allows for identifying distinct economic regimes, capturing the heterogeneous effects of macroeconomic variables on credit risk across different conditions. The empirical results show that slower economic growth, rising inflation, an appreciating currency, and higher interest rates are associated with increasing credit risk across the BRICS countries. The empirical results reveal that the effects of these macroeconomic determinants are not homogeneous across different regimes. For example, slower economic growth, rising inflation, and higher interest rates increase credit risk. Still, the impact of these variables varies depending on whether the economy is in a stable or volatile state. This heterogeneity underscores the importance of understanding credit risk through a regime-switching lens, as traditional models may overlook the distinct ways macroeconomic conditions affect risk during different economic phases. The study further employs feature importance analysis using a random forest model to identify the most influential macroeconomic variables explaining credit risk. From the feature importance results across BRICS countries:
viii | P a g e
Interest rates are the most important variable in Russia, India, and South Africa – high borrowing costs elevates NPLs. Stock market returns are most critical in Brazil, while the current account balance dominates in China, suggesting distinct macro-financial linkages driving credit risk in each country. The second empirical chapter examined the effects of climate change on sovereign credit risk in developed and emerging countries. Using climate vulnerability and resilience indices as proxies for climate change and bond yields and spreads as proxies for sovereign credit risk, the empirical results show that climate change vulnerability has a significant positive impact on sovereign credit risk particularly in low-risk economies where investors react more strongly to climate risks. High-risk emerging markets do not show a strong link between climate risk and sovereign credit risk. In contrast, climate change resilience has a negative effect on sovereign credit risk. A comparison between developed and emerging countries indicates that the impact of climate change is more significant, especially in emerging countries. Developed markets are more resilient overall, but climate risk plays a larger role in high-risk developed countries, which amplifies financial distress Our findings consistently demonstrate that developed countries are more prepared to mitigate climate-related when compared to emerging countries. The third empirical chapter investigates the role of climate change in driving systemic risk within South Africa’s banking sector, focusing on asset volatility as a mediating factor. The escalating impact of climate change on financial systems signals an urgent need to enhance risk management within the banking sector. Addressing this challenge is particularly critical for South Africa, where climate-induced systemic risks are increasingly evident. Quarterly data from 2002 to 2020 was used for this study. The study utilizes Bayesian Model Averaging (BMA) and Structural Equation Modeling (SEM) along with the Baron and Kenny mediation approach. The findings reveal a positive relationship between climate change and bank systemic risk, with asset volatility acting as a partial mediator suggesting that climate-induced risk elevates bank systemic risk in South Africa. The study underscores the need for cohesive risk management strategies that integrate both macro-prudential regulatory perspectives and micro-risk management practices to mitigate
ix | P a g e
climate-induced systemic risks. This study contributes to the understanding of climate change’s impact on systemic risk in South Africa’s financial system by using the Component Expected Shortfall (CES) method to quantify risk. By using asset volatility as a mediator and the ND-GAIN Climate Vulnerability Index, the study offers a nuanced, multidimensional view of how climate risks affect financial stability. In the last empirical chapter, we examined the role of climate finance in promoting climate resilience across BRICS nations using a panel quantile regression model. The findings confirm a significant and positive influence of climate finance on resilience, indicating that increased financial support for climate-related projects enhances adaptive capacity. Additionally, the Green Growth Index (GGI) demonstrates a consistently positive relationship with resilience, highlighting the crucial role of green growth efforts, such as investments in renewable energy, sustainable infrastructure, natural resource management, and social inclusion, in fostering climate adaptation.