Reputation and Disclosure in Dynamic Networks

TL;DR

This paper models how public custody and disclosure protocols affect reputation and information flow in dynamic networks, using Bayesian and stochastic process tools.

Contribution

It introduces a novel disclosure protocol with a public custody process and analyzes its equilibrium properties and network implications.

Findings

Retention influences disclosure timing and strategic behavior.

Finite-time resolution is achieved under reputational discipline.

Shared record censoring impacts network formation and value.

Abstract

Docketed hard evidence changes the meaning of delay: at a public review date, retention is a censoring event. I study a disclosure protocol in which a dated sealed record moves through public custody; review dates are public, and content becomes public only at terminal disclosure. Retention rules out the states in which the holder would have relayed or disclosed the record. With custody separated from audit information, Bayes' rule gives a Gaussian reference likelihood restricted to the remaining support; under interval strategies, that support is tracked by finitely many endpoints. An Ornstein-Uhlenbeck quadratic benchmark verifies Markov perfect Bayesian equilibria in such strategies and gives finite-time resolution under reputational discipline, with tail bounds beyond compact support. With two certified routes for the same record, retention on one route changes what is feasible on the other before it acts. This common-record censoring creates a network value that pairwise formation can miss.