End-to-end advantage of bounded local paths over longer cycles

Determine whether bounded three-firm path clearing retains a larger share of centralized payment-mass reduction than longer cycle clearing when participation declines or consent becomes less reliable.

Background

The paper hypothesizes that the limited coalition size of a two-edge path may make path clearing more robust than longer-cycle clearing in decentralized settings. Longer cycles require evidence and agreement from more firms, potentially increasing rejection, timeout, stale-reservation, and disclosure risks.

Although the centralized experiments show policy-level advantages for the evaluated path policy in most cohorts, they do not establish that paths will outperform cycles after asynchronous communication, partial participation, strategic behavior, and consent frictions are introduced. The authors identify this as a future empirical test rather than a conclusion supported by the present experiments.

References

Whether it produces lower end-to-end communication cost or higher realized PMR is an empirical question for the next study.

— Atomic Common-Day Invoice Clearing under Causal Daily Scheduling: Path-Enabled and Bounded-Cycle Policies  (2608.23060 - Rosa et al., 24 Aug 2026) in Section 6.3, “Path-enabled clearing as an agentic research agenda”