Decision-Relevant Information in Partially Observed Production Networks

2026-09-24

AI Summary

This study addresses the challenge of identifiable economic decisions with unknown individual exposures in partially observed production networks. It proposes a novel paradigm that evaluates data based on decision utility rather than full network disclosure. Methodologically, exposure information is characterized through sufficient statistics and systematically assessed using input-output account analysis, maximum regret computation, and frozen design techniques. The results demonstrate that optimal decision performance can be achieved with only a small set of key statistics, effectively distinguishing baseline performance from decision guarantees. This work establishes a new theoretical framework for evaluating the value of network data under limited information.

Abstract

A production network can remain largely unidentified even when the economic decision it supports is identified. We characterize sufficient measurements for exposure-based decisions and compute sharp maximum regret over networks consistent with released totals. Using earlier and later vintages of Japan's interregional input-output accounts, we select measurements from the 1995 table and evaluate the frozen design against the 2005 benchmark. At roughly half the statistics required for full disclosure, the resulting monitoring set loses only 0.07 percentage points of average exposure relative to the benchmark optimum, yet its sharp maximum regret across compatible networks is 4.83 points. In U.S. coal deliveries surrounding a 2005 Wyoming rail disruption, additional shipment measurements identify the optimal set of plants to monitor for inventory risk, even though four monitored plants' exposures to the affected coal supply remain unidentified. The results distinguish good benchmark performance from a decision guarantee and show that decisions can be identified before individual exposures. They suggest evaluating network data by the economic decisions they support.

Problem

• production networks
• partial observation
• decision-relevant information
• exposure-based decisions
• maximum regret

Innovation

• partially observed production networks
• sufficient measurements
• maximum regret
• exposure-based decisions
• network data evaluation