Why do organizations exist, and why is their form becoming increasingly volatile? This paper proposes a temporal hierarchy theory arguing that organizations are coordination overhead made structural: mechanisms for managing the transaction costs of executing processes across human agents. The theory rests on a three-level temporal stability hierarchy in which value goals persist across centuries, processes across decades, and organizational form across years. This ordering extends the pace layering framework (Beane and Leonardi, 2025) by specifying the stability ranking and deriving AI-era consequences. From it we derive organizational metamerism: the condition under which structurally distinct configurations produce functionally equivalent value outputs, extending equifinality from path equivalence to simultaneous state equivalence. We argue that AI exposes the metadata nature of organizational form: as coordination costs approach zero, the organizational layer becomes freely reconfigurable. Each technology revolution — from Toyota’s pull systems through API-first architectures to AI-mediated coordination — compresses organizational overhead by absorbing coordination into process execution. We identify the tacit knowledge boundary as the primary constraint on this compression: where processes depend on non-codifiable knowledge, organizational form remains thick. Finally, organizational form functions as an external signal, shaping stakeholder evaluations through legitimacy and identity heuristics independently of process output. Six testable propositions are derived with implications for organizational theory, firm boundaries, and AI governance.
Dmitry Zharnikov (Sat,) studied this question.