Search for a command to run...
The correct valuation of collateral supporting real estate loans has always been a key issue for the stability of the credit system. Substandard lending practices and the absence of uniform valuation approaches have historically contributed to the accumulation of non-performing loans. In recent years, several regulatory measures operating at both the European and national level have introduced principles, rules and procedures aimed at standardizing the valuation of properties pledged as collateral for credit exposures. These interventions seek to promote greater transparency, consistency, and prudence in property appraisals, thereby enhancing the soundness and resilience of the financial system. In January 2025, the updated Regulation (EU) 575/2013 came into force, incorporating the Basel III reform (also referred to as Basel 3+ or Basel IV). Among the innovations introduced, the concept of property value (PV) is particularly relevant, a prudential value that excludes expectations of price growth and considers the sustainability of the value over time in relation to the duration of the loan. PV is defined as a derived value with respect to market value (MV), determined by considering the main current and forward-looking risk factors that may arise during the life of the loan, including environmental, social and governance (ESG) risks, the intrinsic characteristics of the property and expectations regarding the economic cycle. This paper proposes a quantitative model for the determination of PV, applied to a practical case involving a residential property located in a medium-sized city in Italy’s Veneto region. The model adopts a deterministic and a probabilistic approach, the latter implemented through Monte Carlo simulation, which is indeed a generalization of the deterministic one. The model links the assessment of PV to the possible evolution of the property’s key parameters and the real estate cycle over the duration of the loan. It was tested under the assumption of a twenty-year mortgage originated in 2025 for the purchase of a residential property in Italy, considering two alternative locations: a suburban area and a city-centre area. The analysis conducted showed a substantially higher MV haircut for the suburban property compared with the central location. This difference reflects the fact that PV is less sensitive to real estate cycle fluctuations in more premium, central locations. Furthermore, the use of Monte Carlo simulation in the probabilistic approach enabled the calibration of the haircut according to a predefined confidence level, confirming the pattern observed in the deterministic framework. The combined evidence strengthens the empirical robustness of the model and highlights the importance of locational and cyclical dynamics in collateral valuation.