This paper proposes a new mechanism for the emergence of collective exuberance in socio-economic systems, arguing that non-symmetric and hierarchical network structures are key drivers of transient bubbles, even in subcritical regimes. This contrasts with existing models that rely on fine-tuning to critical points. The authors empirically validate their framework by analyzing meme stocks and demonstrate a link between bubble size and network non-normality, quantified by the Kreiss constant. The findings suggest that financial crises are an intrinsic feature of systems with non-normal interactions.

Didier Sornette, Sandro Claudio Lera, Jianhong Lin, Ke Wu