Financial digital intelligence in China is linked to stronger innovation in strategic emerging industries, with gains driven by improved disclosure, lower transaction costs and deeper industry–university–research ties; effects vary substantially across regions and industry types.

Main Finding

Financial digital intelligence (FiDI) — a region-level measure of digital + intelligent financial development emphasizing AI, big data, blockchain and related tools — significantly and positively promotes the innovative development of strategic emerging industries (SEIs) in China (sample 2015–2022). The effect operates partly through three mediating channels: improving firms’ information disclosure quality, lowering transaction costs, and strengthening regional industry–university–research (IUR) collaboration. The positive impact is heterogeneous across regions and industries (not uniformly present everywhere).

Key Points

• Core result: Higher regional FiDI is associated with greater firm-level SEI innovation (composite index of R&D inputs and outputs).
• Mediating mechanisms:
  • Information disclosure: FiDI improves the quality/visibility of firm information, relaxing financing frictions and encouraging R&D.
  • Transaction costs: FiDI reduces search/monitoring/contracting costs (via data, smart contracts, monitoring), enabling longer‑horizon/high‑quality innovation.
  • Industry–university–research cooperation: FiDI fosters coordination and resource flows among firms, universities and research institutes, accelerating innovation cycles.
• Heterogeneity:
  • Significant FiDI effects found in central, eastern and western regions (varying magnitudes), and in capital‑intensive and technology‑intensive sectors.
  • No significant effect detected in some other regions/industry groups (authors note uneven spatial and industrial diffusion).
• Limitations acknowledged by authors:
  • Sample restricted to listed SEI firms (789 firms; 5,731 firm-year observations), so effects on startups, private SMEs, and university spin‑offs may be understated.
  • Industry classification coarse (labour-, capital-, technology‑intensive) and may mask intra‑industry differences.
  • Measurement and causality: while panel models and mediators are used, causal identification beyond fixed effects is limited.

Data & Methods

• Sample: 5,731 unbalanced firm-year observations (2015–2022) covering 789 listed SEI firms across 114 prefecture‑level Chinese cities. Firms selected by National Bureau of Statistics’ SEI classification; excluded ST/ ST* firms and required >20% main-business revenue in SEI.
• Outcome (SEIID): Composite index combining innovation input (R&D personnel; R&D investment / operating income) and innovation output (annual patent applications; increase in intangible assets / beginning total assets).
• Core explanatory (FiDI): Novel region-level FiDI index constructed by:
  • Extracting 50 keywords (digitalisation + intelligence terms) from nine authoritative industry reports (2020–2024).
  • Crawling Baidu News search result volumes for those keywords, aggregating at region-year level, log-transforming to reduce skew.
  • Integrating counts of regional fintech companies (details reported in paper).
  • Authors contrast this with the Peking University Digital Inclusive Finance index (used for robustness).
• Controls: firm size, leverage, Tobin’s Q, ROA, age, largest shareholder share, technology/finance policy proxies, etc.
• Empirical strategy:
  • Panel regressions with industry and year fixed effects: SEIID_it = β0 + β1 FiDI_it + controls + u_i + θ_t + ε_it.
  • Mediator tests: additional regressions including (1) information disclosure quality (KV), (2) transaction cost proxy (SER), and (3) regional IUR cooperation (LNIUR) to test indirect paths.
  • Robustness: alternative FiDI measure (PKU index) and other checks described.
• Descriptive stats: SEIID mean = 0.074 (SD 0.101); FiDI mean ≈ 14.07 (SD 0.821).

Implications for AI Economics

• Measurement innovation: The FiDI index operationalizes AI- and data-driven financial modernization (via keyword search volumes + fintech counts). This approach can be adapted in AI economics to track regional AI-finance adoption where administrative statistics are limited.
• Mechanisms highlight AI’s economic role:
  • Reducing information asymmetry and search/monitoring costs shows how AI in finance can reallocate capital toward longer‑horizon, riskier R&D — a key channel for productivity growth in high‑tech sectors.
  • Enabling better cross‑institutional coordination (IUR) points to AI‑enabled complementarities between finance, human capital, and research infrastructure.
• Policy relevance:
  • Promoting FiDI (e.g., data infrastructure, AI capabilities in banks, smart contracting, reliable data governance) can be a lever to accelerate SEI innovation, but benefits are uneven geographically and across industry types — targeted regional and sectoral policies are needed.
  • Regulators should balance innovation support with safeguards: the paper reiterates prior concerns that rapid digital finance growth can be a “double‑edged sword” if it leads to financialization that neglects the real economy or increases systemic risk.
• Research priorities for AI economics:
  • Extend analysis to unlisted firms, startups and university spin‑offs to capture the full innovation ecosystem.
  • Strengthen causal identification (e.g., quasi‑experiments, instrumental variables tied to exogenous rollout of AI/FiDI infrastructure).
  • Explore distributional effects (which firms/regions capture benefits) and potential adverse effects (excess leverage, misallocation).
  • Apply and refine FiDI-style measures across countries to compare how AI-enabled finance shapes industrial upgrading internationally.

Reference (paper summarized): Wei Xue, Ziyan Li, Qianyi Lei (2026). "Financial Digital Intelligence and Innovative Development of Strategic Emerging Industries: Impact Effects and Mechanism Tests." Economic Computation and Economic Cybernetics Studies and Research, Vol. 60, Issue 1/2026. DOI: 10.24818/18423264/60.1.26.08.