Non-Equilibrium Dynamics of Digital Expansion: A Stability–Fragility Trade-off in Vietnamese Commercial Banks
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
This paper develops a non-equilibrium framework to analyze the systemic implications of digital expansion in banking networks. Departing from the conventional view that digitalization is monotonically stabilizing, we model the banking system as a stochastic dynamical process in which digital expansion acts as a structural perturbation to both the drift and covariance operators. Within this framework, we formally establish the existence of a stability–fragility trade-off and derive a critical threshold beyond which marginal stability gains are reversed. The theoretical mechanism is characterized by a spectral phase transition: as digital expansion increases, the system’s covariance structure undergoes concentration, leading to the detachment of a dominant eigenmode. This transition is identified using random matrix theory through deviations of the largest eigenvalue from the Marchenko–Pastur bound, providing a rigorous diagnostic of endogenous systemic risk. Empirically, we test the model using panel data from 27 Vietnamese commercial banks over the period 2010–2024. A dynamic panel GMM specification is employed as a reduced-form representation of the underlying stochastic system, allowing us to estimate the nonlinear impact of digital expansion on banking stability. The results reveal a statistically significant turning point, beyond which digital expansion amplifies fragility rather than resilience. Complementary rolling-window spectral analysis confirms the persistent detachment of a systemic mode, consistent with the theoretical predictions. These findings demonstrate that digital transformation is not an additive technological improvement but a structural reconfiguration of financial networks. In emerging markets, unbalanced digital expansion, particularly when technological capacity outpaces risk-processing capability, can endogenously generate systemic vulnerability. The results have direct policy implications for the design of digitalization metrics and the sequencing of financial inclusion strategies.