Defining the Machine-Economy Era: Economic Agency, Necessary Human Inputs, and the Renewal of Productive Capacity

Rising automation alone does not identify when the economically consequential organization of production changes from predominantly human-selected decisions to sustained machine-led decisions, qualified output and essential productive-capital renewal. This conceptual and measurement paper proposes—not establishes—a historical-economic identification contract for that possible transition. It defines a candidate macroeconomic machine-predominant formation by joint machine predominance in substantive economic discretion (D), accepted production (P), and both the choice and realization of necessary capital replacement (K1-g/K1-r), verified along the same service–asset lineages. The claim is narrower than the existence of machine economies, algorithmic markets or autonomous AI, and distinct from general-purpose-technology paradigms, task displacement and real-authority theories on which it builds. Its proposed incremental use is to prevent high but disconnected automation shares from being mistaken for a network-wide shift in productive organization; whether this improves historical research beyond strong existing approaches remains empirically unproven. The paper compares K1 formation, K2-dependent formation, local productive self-sustenance and macroeconomic maturity under the same constructed facts; it distinguishes structural ERA from the independent legality of actual conduct. An event-and-weight contract separates decision materiality from nonduplicated production weights, treats pending asset replacement as right-censored, and specifies macro scope, dependency links, three clocks, missingness and threshold sensitivity before classification. Two documentary historical reference cases show what early automated systems actually establish and which decisive records remain unavailable. The illustrative Python rules and synthetic attacks validate only conditional logical behavior, not real national era classifications, prediction gains or priority over earlier theories. If K2, historical counterexamples, measurement unreliability or a stronger existing framework eliminate the value of this classification, its “era” label should be revised or downgraded. Conceptual and Methodological Research Preprint. Not peer reviewed.