Reconstructing firm-level interactions in the Dutch input-output network from production constraints

Area of Science:

• Economics
• Network Science
• Economic Sociology

Background:

• Economic resilience analysis requires understanding firm-level input-output networks.
• Empirical inter-firm network data is scarce due to confidentiality issues.
• Existing network reconstruction methods overestimate system resilience by assuming full connectivity.

Purpose of the Study:

• To develop a novel method for reconstructing inter-firm networks from transaction data.
• To accurately model firm-level input-output structures and their impact on economic resilience.
• To provide a realistic generative model for inter-firm networks considering production constraints.

Main Methods:

• Utilized two large transaction datasets from the Netherlands to map the inter-firm network.
• Introduced a generalized maximum-entropy reconstruction method.
• Ensured the method preserves individual firm production functions (input/output flows per product type).

Main Results:

• The maximum-entropy method's accuracy in network reconstruction improves with finer product resolution.
• The method generates realistic inter-firm networks that respect specific production constraints.
• The model's likelihood quantifies alternative network configurations and estimates reduced rewiring capabilities.

Conclusions:

• The proposed method offers a more accurate approach to reconstructing empirical inter-firm networks.
• This approach provides a realistic generative model for studying economic resilience under supply constraints.
• The findings highlight the importance of detailed production information in assessing network adaptability.