Co-location of skill related suppliers: Advancing coagglomeration research using firm-to-firm network data
This repository contains data and code to reproduce the results and visualizations of the following research paper:
Juhász, S., Elekes, Z., Ilyés, V. and Neffke, F. (2026) Co-location of skill related suppliers: Advancing coagglomeration research using firm-to-firm network data. Journal of Economic Geography, lbag006. https://doi.org/10.1093/jeg/lbag006
Abstract
Firms in industry clusters benefit from locating close to suppliers and customers. However, the pervasiveness of global value chains questions the need for co-location in buyer-supplier relationships. We propose that supply-chain partners are more likely to co-locate if they exchange not only goods but also know-how, implying superadditivity of Marshallian agglomeration channels. We test this in a coagglomeration framework using microdata for Hungary -- a small, open economy deeply embedded in global value chains -- examining co-location, labor flows and value chains between firms and industries. We find that supply chains foster co-location primarily among firms in skill related industries.
Repository structure
├── code/
│ ├── databank/ # Scripts executed in the secured Databank environment
│ ├── regressions/ # Reproducible code for all regression analyses
│ ├── visuals/ # Notebooks for figures and visualizations
│
├── data/
│ ├── databank_oc/ # Output-checked aggregate data tables from the Databank
│ ├── outputs/ # Model outputs and intermediate data tables
│ ├── shape_files/ # Shapefiles used for visualizations
│
└── README.md
Explore the geography of labor flows and input-output connections
Interactive data visualization site:
https://vis.csh.ac.at/colocation-suppliers/
Designed by Tobias Batik (Complexity Science Hub, Vienna).
Data repository
Aggregate data on skill relatedness, input–output relatedness, and the underlying industry–region level networks visualized on the interactive site are available at:
https://zenodo.org/records/13753930