Garment District Performance: S(weatshop)-Efficiency vs. I(nnovation)-Efficiency

Abstract

Garment districts have always had a form of distinctive industrial organization arising from strong agglomeration economies based on common skills and the need for direct coordination among fashion designers, manufacturers, contractors, and buyers. These local clusters of apparel-related firms are often populated by small and medium sized enterprises whose production is coordinated by large manufacturers, jobbers, or other lead firms. However, large manufacturers are disappearing from garment districts in the United States and other industrialized countries as import competition forces downsizing and business closures. Increasingly it is the smaller firms that tend to survive. Based on field research, this paper will compare the different evolutionary paths of the economic structures, contracting arrangements, and innovation patterns in the garment districts of New York City and Cholet, France. It finds that the two districts had similar economic organizations through the early 1980s, but that they have subsequently responded in very different ways to continuing import shocks. New York City has adopted an enhanced version of its traditional hierarchical contracting mechanisms between jobbers and contractors, whereas Cholet has abandoned hierarchical contracting in favor of more collaborative relationships with external supply chains and a greater reliance on co-contracting relationships within local production networks. Both New York City and Cholet continue to draw upon standard sources of agglomeration economies, but have adopted a different set of arrangements for promoting dynamic efficiency. The experience of these two districts leads us to posit two distinctive models for improving garment district performance - a sweatshop model (S-efficiency) and an innovation model (I-efficiency). These models are fruitful for understanding the performance of other garment districts in France, Italy, and the UK that are part of our study.