Working Paper: NBER ID: w25441
Authors: Andrew B. Bernard; Emmanuel Dhyne; Glenn Magerman; Kalina Manova; Andreas Moxnes
Abstract: This paper quantifies the origins of firm size heterogeneity when firms are interconnected in a production network. Using the universe of buyer-supplier relationships in Belgium, the paper develops a set of stylized facts that motivate a model in which firms buy inputs from upstream suppliers and sell to downstream buyers and final demand. Larger firm size can come from high production capability, more or better buyers and suppliers, and/or better matches between buyers and suppliers. Downstream factors explain the vast majority of firm size heterogeneity. Firms with higher production capability have greater market shares among their customers, but also higher input costs and fewer customers. As a result, high production capability firms have lower sales unconditionally and higher sales conditional on their input prices. Counterfactual analysis suggests that the production network accounts for more than half of firm size dispersion. Taken together, our results suggest that multiple firm attributes underpin their success or failure, and that models with only one source of firm heterogeneity fail to capture the majority of firm size dispersion.
Keywords: firm size heterogeneity; production networks; buyer-supplier relationships
JEL Codes: F10; F12; F16; L23; L25
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Firm Size (L25) | Input Prices (L11) |
| Firm Size (L25) | Number of Customers (C69) |
| Firm Size (L25) | Market Share (L17) |
| Production Capability (D24) | Market Share (L17) |
| Production Capability (D24) | Input Costs (D24) |
| Production Capability (D24) | Number of Customers (C69) |
| Production Network (D85) | Firm Sales Variance (L21) |
| Downstream Component (D39) | Size Dispersion (D39) |