Latent Fragility and Clustered Withdrawals in Dynamic Banks Runs

Battle Count: 99

The paper is primarily about banking theory and depositor behavior rather than trading strategies. However, understanding bank run dynamics and clustered withdrawals could inform risk management for financial institutions, credit risk assessment, and understanding liquidity events that affect market conditions. The mean-field game framework and optimal stopping techniques have methodological relevance but the paper does not directly address trading, portfolio construction, or market microstructure.

The paper involves advanced mathematical economics: mean-field games, optimal stopping theory, free-boundary PDE problems, Tarski fixed point theorems, Bank-El Karoui representation, strong maximum principles, and Fokker-Planck equations. Implementing the numerical examples requires solving coupled HJB and Fokker-Planck equations with free boundaries. The continuous-state model with common noise involves infinite-dimensional master equations. The theoretical proofs are highly technical and the numerical implementation would require expertise in stochastic control and PDE methods.

The paper is purely theoretical with analytical proofs and numerical illustrations (Figures 2, 3, F.1). No code or computational scripts are provided. The numerical examples use specific parameter values but no reproducible code is available. The mathematical framework is fully specified but implementing the mean-field equilibrium computation would require significant effort.