What Capital After Labor? Forecasting the Talent ROI Transition in the Human-AI Era

Abstract: AI augmentation breaks the accounting link between labor time and productive contribution, yet firms continue to evaluate talent through time-based overhead bundles. This paper develops a forecasting framework for the transition from time-based talent accounting to output-based talent ROI in the human-AI era. The framework centres on Theorem 3 (ROI Inversion at τ) as the empirical spine, with four mechanism theorems: overhead non-additivity, augmentation-saved-time pathways, innovation-premium amplification, and human-AI dyad attribution uncertainty. Korea's staged 52-hour workweek mandate provides an empirical early-warning case. In a DART panel of 365 listed firms (2,281 firm-year observations), the SG&A-to-revenue ratio rose from 18.26 percent in 2018 to 20.06 percent in 2020, corrected mildly in 2021-2022, and peaked at 20.10 percent in 2024. Under the revenue-percentile cohort proxy, two-way fixed effects (+1.56 pp, p = 0.049), pooled event-study estimates (+4.21 pp at t = +3, p = 0.001), and Callaway-Sant'Anna doubly-robust staggered DiD estimates (+4.51 pp at t = +4) converge on a positive overhead-pressure signature. A 2015-2017 backward extension (224 firms, 601 observations) supplies pre-treatment data, providing evidence against pre-existing upward-trend confounds. We read the Korean evidence not as a direct τ estimate or a point causal magnitude, but as, to our knowledge, the first empirically documented signature of the pre-τ overhead-pressure regime, where time-based accounting still dominates while AI augmentation and labor-time compression jointly raise overhead. Output-based firms are forecast to outperform time-based peers by 1.5-2.0 percentage points in firm-level TFP growth by 2032. The contribution is a forecasting model and managerial planning tool for the shift to AI-augmented talent ROI accounting.

• The paper introduces a forecasting framework that identifies a critical threshold (τ*) where output-based talent ROI overtakes traditional time-based accounting.
• It integrates macroeconomic, sociotechnical, and agency theories with empirical analysis of 365 Korean firms to reveal rising overhead pressures due to AI augmentation.
• Findings suggest firms transitioning to output-based evaluation can boost firm-level TFP by 1.5–2.0pp by 2032, offering actionable insights for strategic planning.

Forecasting the Talent ROI Regime Transition in the Human-AI Era

Introduction and Context

The paper "What Capital After Labor? Forecasting the Talent ROI Transition in the Human-AI Era" (2606.19846) develops a forecasting and planning framework for a structural shift in how firms account for talent ROI in the context of AI augmentation. Historically, labor time served as the canonical unit of productive accounting. The proliferation of human-AI dyads undermines this assumption: the accounting link between labor time and productive output is severed, yet organizational evaluation and overhead structures remain time-based.

Drawing from macroeconomic augmentation frameworks (e.g., Shin's ICH model), evolutionary and sociotechnical regime theories, and multitask agency formulations, the paper articulates a transition threshold—denoted $\tau^*$—at which output-based talent ROI accounting overtakes time-based models. The Korean 52-hour workweek mandate, coupled with accelerated enterprise AI adoption, is analyzed as an empirical prototype of the pre-$\tau$ overhead-pressure regime, supplying leading-indicator evidence rather than point-causal estimation.

Theoretical Architecture

Five Theorems Hierarchy

The forecasting model is structured as a chain of five theorems:

1. Overhead Decomposition Non-Additivity: AI augmentation renders the marginal ROI contributions of talent overhead components (e.g., wage, insurance, space, management, training, communication, motivation) non-additive. Cost-accounting frameworks misestimate productive returns when cross-partial interactions emerge, especially as convergence capacity and AI utilization rise.
2. Slack-Augmentation Synergy: Time savings gained via augmentation are distributed across four mutually exclusive pathways—work intensification, hidden leisure, overemployment, and creative-slack reinvestment; only the latter amplifies innovative ROI. Its probability is monotonically increasing in firms' output-oriented evaluation regimes and employee autonomy.
3. ROI Inversion at $\tau^*$: The intersection of talent ROI curves under time-based and output-based accounting is formalized: as AI utilization increases, ROI under time-based regimes falls, while output-based ROI rises. The critical threshold $\tau^*$, a function of output orientation, autonomy, and convergence capacity, determines optimal regime transition.
4. Innovative ROI Premium: Firms with high convergence capacity and large augmentable cognitive capital (as specified in Espinal Maya's decomposition) realize an amplification factor $k > 1$ for creative-slack returns, providing a vector-theoretic anchor for empirical results such as the Google and 3M slack experiments.
5. Information Asymmetry in Human-AI Dyads: Attributed multitask agency cost is a monotone function of dyadic contribution uncertainty $\gamma$. Output-based evaluation regimes offset this cost, while time-based evaluation magnifies it, especially as attribution opacity rises.

The five-theorem structure allows independent falsifiability and boundary specification, with Theorem 3 constituting the empirical spine anchored by Korean corporate panel data.

Empirical Analysis: Korea as the Pre-$\tau$ Prototype

The panel analysis incorporates 365 listed Korean firms (2,281 firm-year observations, 2018–2024), exploiting externally staged labor-time compression via the 52-hour mandate. The SG&A-to-revenue ratio increased from 18.26% (2018) to 20.10% (2024), with statistical diagnostics across TWFE, event-study, and staggered DiD converging on a positive overhead-pressure coefficient (+1.56–+4.51pp, $p < 0.05$), after controlling for pre-trend confounds.

Key empirical inferences:

• The pattern is interpreted not as a direct $\tau^*$ estimate but as the first documented signature of the pre-$\tau$ overhead-pressure regime—time-based accounting persists, while AI augmentation and labor-time compression jointly raise overhead.
• Firms that transition to output-based talent ROI accounting are forecast to outperform time-based peers in firm-level TFP by 1.5–2.0pp by 2032.
• Cross-country comparisons identify Denmark as a polar comparison (output-based, high convergence capacity, no overhead-pressure surge) and Japan as a cultural twin (time-based, low AI adoption, partial overhead-pressure).

Robustness checks, sectoral stratifications, and backward panel extensions reinforce the directional evidence. The empirical strategy is constrained by disclosure variability and joint shocks, with future identification paths specified for component-level decomposition and regime separation.

Practical and Managerial Implications

At the firm level, the theoretical apparatus supplies actionable planning prescriptions. Firms below $\tau$0 should monitor and develop convergence capacity; those approaching $\tau$1 should pilot output-based evaluation in selected divisions; post-inversion, firms should institutionalize output-based accounting and creative slack regimes. The framework cautions against antipatterns: surveillance-driven productivity paranoia, the premature shift to output-based evaluation absent convergence capacity investment, and nominal slack policies without autonomy or output-orientation.

For national and sectoral policy, the Korean case demonstrates how rigid labor-time regulation acts as a timing anchor for overhead-pressure accumulation. The framework predicts heterogeneous regime transitions across OECD economies, contingent on institutional evaluation structures and the diffusion curve of enterprise AI utilization.

Theoretical Contributions and Future Directions

The paper synthesizes production-function genealogy, augmented human capital theory, regime-transition forecasting, and multitask agency extensions into a formal, falsifiable forecasting apparatus. The explicit identification of the pre-$\tau$2 regime, empirical early-warning signature, and the Foresight Matrix planning tool constitute practical advances for managerial, organizational, and policy decision-making. It advances beyond occupation-level displacement forecasting to regime-transition forecasting—the inversion of accounting categories for talent in the human-AI era.

Future directions include dynamic tracking of $\tau$3 crossing, sector-specific theorem magnitude estimation, and longitudinal evaluation of TFP separation following regime transition. Empirical methodologies must be expanded to supply component-level overhead granularity, direct firm-level AI utilization statistics, and multidimensional convergence capacity measurement.

Conclusion

The onset of AI augmentation fundamentally disrupts the labor time–productivity nexus embedded in contemporary talent accounting. The five-theorem architecture forecasts that current time-based overhead regimes will accumulate pressure until regime inversion occurs at a critical utilization threshold. Korea’s empirical trajectory provides a leading indicator template, with sectoral and cross-country variations anticipated over the coming decade. Theoretical and managerial implications converge on the necessity to reconstruct accounting categories, evaluation mechanisms, and overhead structures to realize the productivity potential of augmented human capital. The framework is falsifiable; if regime transition fails to yield the forecasted TFP separation, revision or replacement will be required.