The Modular Kimono
| Dick Langlois |
I recently ran across this interesting paper on vertical integration and subcontracting in the Japanese kimono industry of the late nineteenth and early twentieth centuries. By this period, most of the Japanese silk (and cotton) industries had adopted the factory system. But there remained a few industrial districts that relied on the putting-out system. This paper is most interested in presenting a risk-aversion model that explains why “premier subcontractors” got relational contracts in the putting-out system. I’m not sure I buy it, but in any case what caught my eye was something else — a modularity story:
In the weaving industry of Kiryu, the factory system equipped with hand looms had been chosen to weave the luxury fabrics, while the putting-out system had been used for most other fabrics, until the factory system equipped with power looms became dominant for most kinds of fabrics in the 1910s and later. Instead of being replaced, the putting-out system developed and dispersed within Kiryu, especially from the 1860s to the 1900s, when the main products of Kiryu were yarn-dyed silk fabrics. “Yarn dying” means material yarn is dyed before weaving. For the luxury fabrics that were dyed after weaving, the cleaning and finishing processes undertaken after weaving were important, and those processes were conducted inside the manufacturers’ workshops. In contrast, in the production of the yarn-dyed fabrics, dying, arranging warps, cleaning yarn, throwing, re-reeling, and other preparation processes were essential. Because those processes needed special skills, the craftsmen who specialized in each process were organized as subcontractors by manufacturers. … With the moving weight from production of traditional piece-dyed (dyed-after-weaving) fabrics to production of yarn-dyed (dyed-before-weaving) silk fabrics, the throwing process, the finishing process, and the designing process, as well as the weaving process, came to be put-out. Manufacturers decreased the production inside of their workshops and established subcontracting relations with independent artisans. This case suggests that the technological change induced by the change of products from piece-dyed fabrics to the yarn-dyed fabrics affected production organization.
This has a bit of a Christensen flavor to it. When “performance” needs were high — high-end kimonos — the industry used a non-modular technology (dyed-after-weaving) and an integrated organization. When performance needs were lower — lower-quality kimonos — it used a modular technology (dyed-before-weaving) and a vertically disintegrated structure.