China’s AI pilot zones lower listed firms’ cost of equity by improving human capital, ESG and green innovation signals; private, high‑polluting and high‑tech firms benefit most. The regional AI policy appears to reduce investor uncertainty and required returns, potentially unlocking investment in innovative activities.

Regional AI development and firms’ cost of equity capital: Evidence from China’s AI pilot zones

Xiaoping Zhang, Pu Zhao, Jing Shi, Yiwen Gao · September 09, 2026 · Economic Analysis and Policy

openalex quasi_experimental medium evidence 8/10 relevance Summary only summary available; pdf_status=paywall DOI Source PDF

Structured author observations

Linked only from stored provider relations; the raw author line above is never matched by name.

OpenAlex

Latest observation: September 13, 2026

Xiaoping Zhang provider ID
Pu Zhao provider ID
Jing Shi provider ID
Yiwen Gao provider ID

Exposure to China’s AI pilot zones (AIIDPZ) leads to significant reductions in listed firms’ cost of equity, with effects operating through improved human capital composition, stronger ESG performance, and increased green innovation.

Citation observations

Cumulative provider counts captured on specific dates; providers are never combined.

0 cumulative citations

OpenAlex · Observed September 13, 2026

View corpus context

0 cumulative citations

Semantic Scholar · Observed September 13, 2026

View corpus context

In this study, we investigate whether AI-oriented regional policy reduces firms’ cost of equity capital. Using a sample of Chinese A-share listed firms from 2006 to 2024, we exploit the establishment of China’s National New Generation Artificial Intelligence Innovation and Development Pilot Zones (AIIDPZ) as a quasi-natural experiment. Our findings indicate that AIIDPZ exposure significantly reduces firms’ cost of equity capital. The results remain robust after a series of robustness tests. The mechanism through which AIIDPZ reduces the cost of equity capital is by improving firms’ human capital structure, enhancing ESG performance, and promoting green innovation, thereby strengthening firms’ growth prospects and the information environment perceived by investors. Furthermore, the effects of AIIDPZ are more pronounced among non-state-owned firms, high-polluting firms, and high-tech firms. Overall, our study contributes to the literature on AI development and the cost of equity capital by showing that a regional AI innovation ecosystem can shape investors’ required returns, providing new evidence that technology-oriented place-based policies have important capital-market consequences beyond innovation and industrial upgrading.

Summary

Main Finding

Exposure to China’s National New Generation Artificial Intelligence Innovation and Development Pilot Zones (AIIDPZ) significantly reduces listed firms’ cost of equity capital. The effect is robust and operates through improvements in firms’ human capital structure, ESG performance, and green innovation, which together strengthen growth prospects and the information environment investors use to set required returns.

Key Points

Sample: Chinese A-share listed firms, 2006–2024.
Identification: AIIDPZ establishment used as a quasi‑natural experiment to assess regional AI policy impact on firms.
Main outcome: AIIDPZ exposure lowers firms’ cost of equity capital.
Mechanisms:
- Better human capital structure (skills, workforce composition) improves expected future cash flows.
- Enhanced ESG performance reduces perceived risk and information asymmetry.
- Increased green innovation signals long-term viability and reduces environmental risk exposure.
Heterogeneous effects: stronger reductions in cost of equity for non-state-owned firms, high-polluting firms, and high-tech firms.
Robustness: results hold across a battery of robustness checks (alternative specifications and tests reported).

Data & Methods

Data: Panel of A-share listed firms in China spanning 2006–2024.
Empirical strategy: Exploit the staggered/regionally targeted rollout of AIIDPZ as a quasi-natural experiment (difference-in-differences style identification implied).
Outcome variable: Firm-level cost of equity capital (standard estimation methods for cost of equity used).
Mechanism tests: Empirical analyses linking AIIDPZ exposure to firm-level measures of human capital composition, ESG scores/performance, and green innovation activity; mediation-style evidence that these channels account for part of the effect.
Robustness checks: Multiple robustness tests reported (alternative controls/specifications, sample splits, sensitivity analyses).

Implications for AI Economics

Policy: Place-based AI innovation policies can affect capital-market pricing, not just local innovation output. By lowering firms’ equity costs, regional AI ecosystems can facilitate investment, scaling, and reallocation toward innovative activities.
Market signaling: AI-oriented clusters improve information environments (via ESG, human capital, green innovation signals), reducing investor uncertainty and required returns—highlighting a non-technology pathway through which AI policy influences finance.
Distributional effects: Benefits are concentrated among privately owned, polluting, and high-tech firms, suggesting heterogeneous impacts across ownership types and industry exposures; policymakers should consider equity and sectoral distribution when designing regional AI initiatives.
Research directions: Further work can quantify long-term welfare impacts, map firm-level investment responses to lower equity costs, explore causal microchannels in more detail, and test external validity in other countries or with alternative AI policy instruments.

Assessment

Paper Typequasi_experimental Evidence Strengthmedium — Strengths include a long panel of listed firms, plausibly exogenous staggered policy rollout, multiple mechanism tests (human capital, ESG, green innovation), and numerous robustness checks; limitations include potential non-random selection of pilot zones, spillovers across regions, staggered DiD biases and heterogeneous treatment timing concerns, and limited detail here on pre-trends and placebo tests. Methods Rigormedium — The design leverages quasi-experimental variation and tests mechanisms, which is good practice, but without full details there are open concerns about identification validity (selection into pilot status, parallel trends, heterogeneous timing in staggered DiD) and whether alternative explanations (concurrent local policies, investor composition changes) are fully ruled out. SamplePanel of Chinese A-share listed firms observed annually from 2006 to 2024, matched to firm location/region to determine exposure to AIIDPZ; firm-level outcomes include estimated cost of equity, firm controls, ESG scores, measures of human capital structure and green innovation activity (exact sample sizes and exclusions not reported in the summary). Themesinnovation human_ai_collab IdentificationStaggered, regionally targeted rollout of China’s National New Generation Artificial Intelligence Innovation and Development Pilot Zones (AIIDPZ) used as a quasi-natural experiment in a difference-in-differences/event-study framework: firms are coded as treated if located in pilot zones and compared to firms in non-pilot regions over 2006–2024, with firm and year fixed effects and controls to isolate the policy effect on firm-level cost of equity. GeneralizabilityFindings limited to publicly listed Chinese A-share firms (large/regulated firms) and may not generalize to private firms or SMEs., China-specific institutional context (state involvement, capital‑market structure, regional industrial policy) may limit external validity to other countries., AIIDPZ is a specific place-based policy—results may not transfer to non-place-based AI interventions or to different policy designs., Possible selection of pilots into already-favored regions means effects could differ where zones are assigned differently., Outcome is cost of equity; results speak directly to financial pricing but may not map directly to real-sector productivity or employment effects.

Claims (7)

Claim	Direction	Outcome	Confidence & Evidence	Details
Exposure to China’s National New Generation Artificial Intelligence Innovation and Development Pilot Zones significantly reduces listed firms’ cost of equity capital. Other	negative	Firm-level cost of equity capital	Reading fidelity high Study strength medium	not reported 0.48
The reduction in firms’ cost of equity capital operates partly through improvements in human capital structure. Skill Acquisition	positive	Firm human capital structure, including skills and workforce composition	Reading fidelity high Study strength medium	not reported 0.48
The reduction in firms’ cost of equity capital operates partly through enhanced ESG performance. Ai Safety And Ethics	positive	Firm ESG performance	Reading fidelity high Study strength medium	not reported 0.48
The reduction in firms’ cost of equity capital operates partly through increased green innovation. Innovation Output	positive	Firm green innovation activity	Reading fidelity high Study strength medium	not reported 0.48
The reduction in cost of equity capital is stronger for non-state-owned firms than for state-owned firms. Other	negative	Firm-level cost of equity capital	Reading fidelity high Study strength medium	not reported 0.48
The reduction in cost of equity capital is stronger for high-polluting firms and high-tech firms. Other	negative	Firm-level cost of equity capital	Reading fidelity high Study strength medium	not reported 0.48
The estimated negative relationship between AIIDPZ exposure and firms’ cost of equity capital remains under multiple robustness checks. Other	negative	Firm-level cost of equity capital	Reading fidelity high Study strength medium	not reported 0.48