An extensive body of literature explains features of the highly co‐operative relationship between Japanese firms, particularly Toyota, and their supplier‐firm network. The literature does not explore when these networks emerged or the factors conducive to their emergence. This paper uses transaction cost analysis and game theory to address these issues. It indicates that the networks emerged in the 1950s, and that the initiating factors were exogenous to the networks, centring on the unusual business environment that then prevailed in Japan. The analysis indicates why firms like Toyota gained an advantage over competitors in America. They were relieved of two major transaction costs: those linked to internalization and decomposed subcontracting. In addition, successful networks developed intragroup understandings that led to significant reductions in both interfirm co‐ordination costs and direct production costs per unit of output. The paper questions the extent to which Toyota‐type networks are prevalent in Japan. It questions the advantage that firms in America or elsewhere (even Japan) can gain from emulating Toyota practices such as unguarded subcontracting. It offers a different interpretation to standard business practices in firms like Toyota, such as relational contracting. It provides a basis for a reinterpretation of Japanese business history and the role of ‘Japan Inc.’.