This article develops Agentic Capital as a measurable and econometrically testable productive category of the agentic economy. The core argument is that models, data, robots, compute, protocols, and human-sovereignty systems are no longer auxiliary digital inputs; they increasingly constitute a composite stock of productive assets that enables economic action to be selected, executed, verified, and governed. To operationalise this concept, the study constructs a six-pillar Agentic Capital framework composed of model capital, data capital, robotic physical capital, compute capital, protocol capital, and human sovereignty capital. The methodology adapts the diagnostics-first logic of the Gondauri Index by using robust normalisation, additive and geometric aggregation, balance and gap diagnostics, non-compensatory adjustment and scenario pathways. The key transformation is the Gondauri-adjusted Agentic Capital Index, which integrates macro-financial resilience into the agentic-capital measurement architecture so that technological readiness is not interpreted separately from systemic resilience. Empirically, the article applies a country-year panel design for nine economies over 2005-2024 and estimates the relationship between agentic capital and labour productivity growth through a layered econometric strategy. The analysis combines pooled OLS, random effects, fixed effects, Hausman model selection, panel unit-root diagnostics, Pesaran cross-sectional dependence testing, Driscoll-Kraay robust standard errors, dynamic panel GMM, IV-2SLS, panel quantile regression, Hansen-style threshold estimation, mediation analysis, component-level fixed effects, and robustness checks. The results indicate that geometric agentic capital is positively associated with labour productivity performance, that structural imbalance weakens the maturity of the system, and that human sovereignty and governance conditions shape the productivity value of agentic systems. The article contributes to capital theory by shifting the analytical focus from accumulated productive assets toward accumulated capacity for governed economic action. It also provides a reproducible empirical architecture for assessing how countries, sectors, and firms can transform AI readiness into legitimate, auditable, and productivity-relevant agentic capacity.
Gondauri et al. (Fri,) studied this question.