Working Paper: NBER ID: w18739
Authors: Allan Collard-Wexler; Jan De Loecker
Abstract: We measure the impact of a drastic new technology for producing steel - the minimill - on the aggregate productivity of U.S. steel producers, using unique plant-level data between 1963 and 2002. We find that the sharp increase in the industry's productivity is linked to this new technology, and operates through two distinct mechanisms. First, minimills displaced the older technology, called vertically integrated production, and this reallocation of output was responsible for a third of the increase in the industry's productivity. Second, increased competition, due to the expansion of minimills, drove a substantial reallocation process within the group of vertically integrated producers, driving a resurgence in their productivity, and consequently of the industry's productivity as a whole.
Keywords: productivity; minimills; steel industry; reallocation; technology
JEL Codes: L1; O3
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| introduction of minimills (L61) | productivity growth (O49) |
| reallocation of output due to minimills (L23) | increase in total factor productivity (TFP) (O49) |
| increased competition from minimills (L19) | reallocation process within vertically integrated producers (L23) |
| reallocation process within vertically integrated producers (L23) | resurgence of productivity among incumbents (O49) |
| exit of less efficient firms (L19) | reallocation of resources to more productive plants (O49) |
| reallocation of resources to more productive plants (O49) | catching-up effect among vertically integrated producers (L23) |