Embarking on a comprehensive exploration of Probability of Default (PD) modeling, we embark on a journey that traces its origins from the pioneering works of Beaver (1966), Altman (1968), and Ohlson (1980). The landscape has evolved, incorporating machine learning (ML) techniques, as showcased in modern studies like Moscatelli et al. (2020). Yet, a regulatory conundrum arises, with regulators hesitating to fully embrace ML models for compliance purposes, as discussed in EBA (2021). A potential middle ground, proposed by the Bank of Spain, suggests utilizing conventional models for regulatory compliance and ML for validation, adding an intriguing layer to the PD modeling narrative.
Global Perspectives on PD Modeling: PD models have been crafted for various countries, reflecting the unique economic nuances of each. From France and China to the UK, Japan, the EU, and Hungary, these models have evolved to address the specificities of each financial landscape. The Italian case, as a G7 member with systemic importance within Europe and globally, beckons a closer examination of its banking dynamics.
Banking Sector Nuances: PD models tailored for banks face challenges, especially when dealing with so-called “low default portfolios (LDP).” The Lehman Brothers case serves as a stark reminder that even seemingly negligible default rates can have profound consequences, emphasizing the need for specialized PD models for the banking sector. Noteworthy studies on bank defaults and distress span continents, from the EU and Turkey to India, Japan, the USA, the Persian Gulf region, and BRICS countries, showcasing the global relevance of PD modeling in financial stability.
Moody’s RiskCalc Model and Russian Banking Data: The RiskCalc model by Moody’s emerges as a significant player in the realm of banking PD modeling, covering both the USA and the global banking landscape. Surprisingly, Russian banking data have become a focal point for PD model development, driven by the detailed monthly financial statements provided by the Central Bank of the Russian Federation.
Focus on Italy’s Banking Sector: Zooming in on Italy, previous studies have explored factors influencing bank failures, with a particular emphasis on smaller cooperative banks. The introduction of a co-operative banks (BCC) dummy variable and the subsequent banking reform in 2016 have added layers of complexity and distinctiveness to Italy’s banking sector. This begs the question of whether existing PD models adequately address Italy’s unique characteristics.
Small Banks and Cooperative Groups: Delving deeper into the intricacies of cooperative groups, where smaller member banks wield decision-making power alongside major banks, challenges conventional PD models. The need for a nuanced approach arises – should a new PD model be developed, or can existing solutions like Moody’s Analytics (2016) cater to Italy’s distinct characteristics?
Conclusion: As we navigate through the multifaceted landscape of PD modeling in the Italian banking sector, it becomes evident that the intricacies and uniqueness of Italy’s financial institutions warrant tailored approaches. This journey promises not only to fill the existing research gap but also to shed light on the subtle nuances that set Italy apart in the ever-evolving world of PD modeling.
Source: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4141518
Senior Business Analyst with a track record of spearheading international operations between Europe, America, UAE, and Asia. Specializing in risk management and financial modeling in the financial services sector. Fluent in English, Italian (native), with basic knowledge of German and French. Seeking a role to apply strategic insights and leadership skills.