Chinese publicly listed firms that emphasize big-data use see higher firm value, and that payoff becomes stronger after China's landmark data-protection law; the result suggests robust privacy rules can amplify corporate gains from data-driven operations.
How Big Data Enhances Firm Value Under Data Privacy Regulation
Using firm-level panel data from Chinese A-share companies, the study finds that greater reported big-data usage is associated with higher firm value, and this positive relationship is strengthened after the implementation of China's Personal Information Protection Law.
The rapid adoption of big data and AI is transforming economies, but also raises ethical concerns including data privacy breaches and algorithmic bias. As a central pillar of digital ethics, data ethics emphasizes the responsible use and protection of personal information. This study analyzes panel data covering Chinese A-share listed companies (2007–2021) and uses a fixed-effects regression approach to measure the impact of big data application on firm value. Big data usage is proxied by keyword frequency in annual reports. Results show that big data significantly improves firm value by improving operational efficiency and reducing costs. Importantly, this positive effect is further strengthened following the implementation of China's Personal Information Protection Law (PIPL), indicating that robust data privacy regulations positively moderates the relationship between big data and firm performance. Robust sensitivity tests confirm these findings, suggesting that a well-established legal framework for data privacy enhances the benefits of big data in corporate performance.
Summary
Main Finding
Big data adoption measurably increases firm value for Chinese A‑share companies (2007–2021), primarily by improving operational efficiency and lowering costs; this positive effect is strengthened after China’s Personal Information Protection Law (PIPL), indicating that robust data‑privacy regulation amplifies the corporate benefits of big data.
Key Points
- Big data usage is positively and significantly associated with firm value.
- Mechanisms identified: improved operational efficiency and reduced costs mediate the effect on firm performance.
- The positive big‑data → firm‑value relationship is stronger following implementation of the PIPL, implying regulatory protection of personal data enhances returns to data investments.
- Results are robust to multiple sensitivity checks reported by the authors.
- Interpretation caveats: big‑data usage is proxied by keyword frequency in annual reports (an imperfect measure), and while fixed effects and robustness tests support the association, causal claims are supported but still subject to typical observational limits.
Data & Methods
- Sample: panel data of Chinese A‑share listed firms spanning 2007–2021.
- Big‑data measure: frequency of relevant keywords in firms’ annual reports (textual proxy for adoption/usage).
- Estimation: firm and time fixed‑effects regressions to control for unobserved heterogeneity and common shocks.
- Mechanism tests: mediation analyses linking big‑data proxy to operational efficiency and cost measures.
- Policy moderation: interaction analysis or sub‑sample comparison around the PIPL to assess regulatory moderating effect.
- Robustness: multiple sensitivity checks (unspecified here) to validate stability of results.
Implications for AI Economics
- Policy: Well‑designed data‑protection laws (like PIPL) can increase the private returns to firms’ data and AI investments, suggesting regulation and innovation are complementary rather than strictly trade‑offs.
- Firm strategy: Investing in data and AI is more valuable under clear privacy protections; firms should pair technical adoption with compliance and governance to maximize value.
- Market outcomes: Stronger privacy regimes may raise aggregate productivity by increasing effective use of personal data while mitigating risks that deter firms or users from participating in data ecosystems.
- Research directions: further work should (a) refine measurement of firm‑level AI/big‑data activity beyond keyword proxies, (b) exploit causal designs (e.g., quasi‑experimental variation in regulation enforcement) to strengthen causal inference, and (c) assess distributional and ethical trade‑offs (consumer welfare, privacy harms, and algorithmic bias) alongside firm‑level gains.
Assessment
Paper Typequasi_experimental
Evidence Strengthmedium — The study uses a long panel with firm and year fixed effects and exploits a clear policy change (PIPL) to strengthen causal claims, which improves credibility; however, the main treatment (big-data usage) is proxied by keyword frequency in reports and may reflect reverse causality or time-varying omitted factors (e.g., managerial quality, firm strategy) that are not fully addressed, so causal inference is plausible but not airtight.
Methods Rigormedium — Appropriate use of panel fixed effects, policy interaction, and robustness checks indicates reasonable econometric practice; limitations include reliance on textual keyword intensity as a noisy proxy for actual big-data adoption, potential measurement error, limited discussion of parallel-trends or heterogeneous pre-trends for the PIPL moderation, and no clear instrumental strategy to address remaining endogeneity.
SampleBalanced/unbalanced panel of Chinese A-share listed firms from 2007–2021; big-data usage proxied by frequency of relevant keywords in firms' annual reports; firm value measured using market-based and/or accounting measures; standard firm-level controls included; analysis restricted to publicly listed Chinese firms.
Themesproductivity governance
IdentificationPanel fixed-effects regressions using within-firm variation in big-data keyword frequency from annual reports (2007–2021), with year and firm fixed effects; causal leverage partly comes from interacting big-data usage with a national regulatory shock (implementation of China's Personal Information Protection Law, PIPL) to test moderation; robustness checks (placebo tests and sensitivity analyses) are reported but no exogenous instrument or randomized assignment is used.
GeneralizabilityFindings apply to Chinese publicly listed (A-share) firms and may not generalize to private firms or non-Chinese institutional contexts., Big-data usage is measured via keyword frequency in annual reports, which may not accurately reflect real investment or implementation—limiting external validity., The moderating effect of PIPL depends on China's specific legal, institutional, and enforcement context and may not map to other countries' data-protection regimes., Time period ends in 2021; rapid advances in AI/ML and later regulatory changes could alter the relationship in subsequent years., Sectoral heterogeneity: effects may differ across industries (e.g., finance, manufacturing, services) and results aggregated at firm-level may mask this.
Claims (10)
| Claim | Direction | Confidence | Outcome | Details |
|---|---|---|---|---|
| The study analyzes panel data covering Chinese A-share listed companies from 2007 to 2021. Other | null_result | high | not applicable |
panel covering 2007-2021 (Chinese A-share listed firms)
0.48
|
| The empirical analysis uses a fixed-effects regression approach to measure the impact of big data application on firm value. Firm Productivity | null_result | high | firm value |
fixed-effects regression used to measure impact on firm value
0.48
|
| Big data usage is proxied by keyword frequency in firms' annual reports. Other | null_result | high | big data usage (proxy) |
big data usage proxied by keyword frequency in annual reports
0.48
|
| Big data application significantly improves firm value. Firm Productivity | positive | medium | firm value |
big data application significantly improves firm value (positive coefficient reported)
0.29
|
| The positive effect of big data on firm value operates through improving operational efficiency and reducing costs. Firm Productivity | positive | medium | operational efficiency; operating costs; firm value |
0.29
|
| The positive impact of big data on firm performance is strengthened following the implementation of China's Personal Information Protection Law (PIPL). Firm Productivity | positive | medium | firm value / firm performance |
0.29
|
| Robust sensitivity tests confirm the main findings, indicating that the results are not driven by model specification or sample selection. Firm Productivity | positive | medium | firm value |
0.29
|
| A well-established legal framework for data privacy (e.g., PIPL) enhances the benefits of big data for corporate performance. Firm Productivity | positive | medium | firm performance / firm value |
0.29
|
| The rapid adoption of big data and AI is transforming economies and raises ethical concerns such as data privacy breaches and algorithmic bias. Ai Safety And Ethics | mixed | medium | economic transformation; ethical risk indicators (conceptual) |
0.29
|
| Data ethics, as a central pillar of digital ethics, emphasizes the responsible use and protection of personal information. Ai Safety And Ethics | null_result | high | not applicable |
0.48
|