Sequential Defaulting in Financial Networks
summary
Paper models discrete-time updates of announced recovery rates in networks with debts and CDSs and shows concrete, surprising behaviors: small examples produce multiple equilibria (three in one case) but some equilibria are unreachable, others cause infinite periodic updates, and nodes can exit default later. Core technical contributions are a constructed binary counter gadget that forces exponential-time stabilization (then halts) and reductions from MAX-SAT proving the problem of finding the best update ordering or best time to default is NP-hard. A modified model using optimistic updates plus liability freezing becomes monotone and guarantees quadratic-time stabilization.
AI-written summary · about