رامین مجاب

According to the Basel III regulatory framework, uniform minimum liquidity requirements have been imposed on all types of banks. Using an agent-based model of a banking system, we investigate the effects of liquidity requirements on banks' insolvency under two policy experiments in one of which the minimum liquidity requirements are applied uniformly and in the other differentially across banks. The model introduces a banking system with 12 heterogeneous banks that must also comply with two liquidity requirements while performing their daily activities of taking deposits and making loans. The model is applied to the Iranian banking system. Results illustrate that because banks respond differently to liquidity requirements, applying one size minimum liquidity requirements to all kinds of banks, strengthens the likelihood of a liquidity shock turns into banks' insolvency and could increase banking system instability. Thus our findings highlight that to achieve financial stability at the national level, policymakers should revise the current one size fits all approach when designing liquidity requirements.

Relying on the Area Under the Curve (AUC) measure, we compare the performance of the Logit regression model and the LightGBM algorithm. Despite these methods being common in the literature, our study emphasizes the role of statistical inference to evaluate and compare the results comprehensively. We use the training set of the Vesta (2018) dataset, provided by Vesta—a global fraud prevention company headquartered in the United States specializing in payment solutions and risk management. Originally released as part of a Kaggle competition focused on credit card fraud detection, this dataset comprises diverse transaction records, representing a rich source for exploring advanced fraud detection methods. Our analysis reveals that while the LightGBM algorithm generally yields higher predictive accuracy, the differences between the calculated AUCs of the two methods are not statistically significant. This underscores the importance of using inferential techniques to validate model performance differences in fraud detection.

Technological progress has transformed the payment industry. Although alternative payment instruments such as bank cards share some characteristics with banknotes, they are not direct competitors to cash. First, cash possesses unique attributes—its functionality without electrical power and its inherent anonymity. Second, statistical data indicate that the ratio of cash to M1 has stabilized at approximately 10% in recent years, primarily due to its substitution with demand deposits. In this paper, we model the demand for banknotes by incorporating technology-related characteristics alongside conventional theories of money demand. Our approach has important implications for central bank, particularly in liquidity management, currency issuance, effective management of physical currency, and continued investment in digital payment infrastructures. We adopt a macroeconomic perspective that considers banknote demand from both transactional and asset viewpoints. To this end, we estimate an autoregressive distributed lag (ARDL) model, which allows us to analyze both short-term and long-term coefficients within a framework of model uncertainty. A key focus of our study is the estimation of a payment system development index using logistic growth and Gompertz models, through which we assess its impact on cash demand. Our empirical evidence supports the notion of S-shaped growth in the adoption of payment technologies. The findings demonstrate that advancements in the payment sector have led to a significant decline in cash demand. Moreover, while increased real economic activity exerts a positive and significant influence on cash demand, a higher opportunity cost of holding cash has a pronounced negative impact.