China’s ‘intelligent manufacturing’ pilot program measurably improved ESG performance and curtailed greenwashing among listed manufacturers — raising ESG scores by roughly 0.14 and strengthening credibility of sustainability reporting, especially for competitive, growing, and capital‑scarce firms.
Intelligent Manufacturing Demonstration Projects Driving Corporate ESG Ratings: An Analysis Based on Innovation Efficiency and Cost Management
Participation in China’s IMDPs raised listed manufacturers’ ESG ratings by about 0.14 points and reduced measures of greenwashing and ESG-rating uncertainty, with effects operating through higher innovation efficiency and improved cost management and strongest in competitive, growth-stage, and capital-scarce firms.
This study examines whether China’s Intelligent Manufacturing Demonstration Projects (IMDPs, 2015–2018) can improve firms’ environmental, social, and governance (ESG) performance and thereby strengthen the quality of green transformation in manufacturing. Exploiting the staggered rollout of IMDPs as a quasi-natural experiment, we combine propensity score matching with a multi-period difference-in-differences design using panel data on Chinese listed manufacturing firms from 2009 to 2022. We find that IMDP participation increases ESG ratings by approximately 0.14 rating levels relative to comparable non-participating firms. Mechanism analyses suggest that the effect operates through higher innovation efficiency and improved cost management, consistent with a channel of capability upgrading and resource reallocation toward sustainability-related activities. The effect is stronger for firms under intense competitive pressure, at the growth stage, and in capital-scarce industries, indicating that industrial policy can be particularly valuable where market frictions constrain green investment. Importantly, we go beyond ESG scores by constructing measures of greenwashing and ESG rating uncertainty, and show that IMDPs reduce greenwashing and lower ESG uncertainty. These results imply that intelligent manufacturing policies can generate economically meaningful benefits by improving firms’ sustainability performance and the credibility of ESG information, which are central to capital allocation and the effectiveness of green governance.
Summary
Main Finding
Participation in China’s Intelligent Manufacturing Demonstration Projects (IMDPs, 2015–2018) causally improves manufacturing firms’ ESG performance and the credibility of ESG information: IMDP firms increase ESG ratings by about 0.14 rating levels versus matched non-participants, and show reductions in both greenwashing and ESG rating uncertainty. Effects are driven by higher innovation efficiency and better cost management.
Key Points
- Causal design: Uses the staggered rollout of IMDPs as a quasi-natural experiment; combines propensity score matching (PSM) with a multi-period difference-in-differences (DiD) framework.
- Effect size: IMDP participation → ~+0.14 ESG rating levels relative to comparable non-participating firms.
- Mechanisms: Improvements operate through (i) higher innovation efficiency and (ii) improved cost management, consistent with capability upgrading and resource reallocation to sustainability activities.
- Information quality: IMDPs reduce measures of greenwashing and lower ESG rating uncertainty, improving the credibility of ESG signals.
- Heterogeneity: Larger effects for firms facing intense competitive pressure, firms at a growth stage, and firms in capital-scarce industries—suggesting industrial policy is most valuable where market frictions hinder green investment.
Data & Methods
- Sample: Panel of Chinese listed manufacturing firms, 2009–2022.
- Treatment: Participation in IMDPs (2015–2018 staggered rollout).
- Identification strategy: PSM to build comparable treatment/control groups, combined with multi-period DiD to exploit timing of adoption and control for time-varying confounders.
- Outcomes: ESG ratings (primary), constructed measures of greenwashing, ESG rating uncertainty; intermediate outcomes include innovation efficiency and cost management metrics.
- Robustness: Heterogeneity and mechanism analyses to support causal channels and external validity across firm types and industry conditions.
Implications for AI Economics
- Industrial policy can accelerate AI-driven manufacturing upgrades that yield positive environmental, social, and governance externalities. IMDPs—programs that promote intelligent manufacturing (digitalization, automation, AI integration)—not only boost operational efficiency but also reallocate resources toward sustainability.
- Information and market effects: By reducing greenwashing and ESG uncertainty, targeted AI/manufacturing policies improve the quality of ESG signals used in capital allocation, potentially lowering information frictions and improving the effectiveness of green finance and governance.
- Complementarity with market forces: Stronger effects where competition is high, firms are in growth stages, or capital is scarce imply that policies help overcome financing and incentive gaps that otherwise slow AI/green investment—important when private returns are uncertain or externalities are large.
- Policy design: Results support targeted interventions (e.g., grants, demonstrations, standards) that foster capability upgrading and diffusion of AI-enabled technologies in sectors with binding frictions, rather than purely blanket subsidies.
- Research directions: Quantifying how AI-specific components of intelligent manufacturing contribute separately to productivity vs. ESG outcomes, measuring long-run financial returns to AI-driven green investments, and studying spillovers across supply chains and financial markets would deepen understanding of AI’s role in green transitions.
Assessment
Paper Typequasi_experimental
Evidence Strengthmedium — Strengths include a plausibly exogenous, staggered policy rollout, long firm-level panel (2009–2022), PSM to improve comparability, multi-period DiD and mechanism tests, and additional outcomes (greenwashing, ESG uncertainty). Remaining concerns reduce strength: potential selection on unobservables into IMDPs, measurement error in ESG ratings, possible bias from heterogeneous timing if not using recent staggered-DiD estimators, and risk of spillovers or concurrent policies.
Methods Rigormedium — The paper uses appropriate and standard quasi-experimental tools (PSM + multi-period DiD) and explores mechanisms and heterogeneity, indicating careful empirical work; however, PSM cannot address unobserved confounders, the analysis hinges on the parallel-trends assumption, and the write-up should explicitly address recent methodological issues with staggered DiD estimators and robustness to alternative matching/estimation choices.
SampleFirm-level panel of Chinese listed manufacturing firms observed 2009–2022, with treatment defined as participation in IMDPs rolled out 2015–2018; treated firms are matched to non-participating firms via propensity scores; outcomes include third‑party ESG ratings and constructed measures of greenwashing and ESG rating uncertainty, plus firm controls and industry/time fixed effects.
Themesinnovation adoption governance
IdentificationExploits the staggered rollout of China’s IMDPs (2015–2018) as a quasi-natural experiment; combines propensity score matching to construct comparable treated and control firms with a multi-period difference-in-differences (DiD) panel design (firm and time fixed effects) to estimate treatment effects; includes mechanism tests (innovation efficiency, cost management) and heterogeneity analyses; relies on parallel trends and selection-on-observables assumptions for causal interpretation.
GeneralizabilityLimited to listed manufacturing firms in China — excludes small, private, and non-manufacturing firms, Results reflect a particular industrial-policy program (IMDP); may not generalize to market-driven AI adoption or different policy designs, Context-specific to China’s institutional and regulatory environment, ESG ratings and constructed greenwashing metrics may not be comparable across countries or rating providers, Study period ends in 2022 and may not capture more recent AI/Industry 4.0 developments
Claims (9)
| Claim | Direction | Confidence | Outcome | Details |
|---|---|---|---|---|
| IMDP participation increases ESG ratings by approximately 0.14 rating levels relative to comparable non-participating firms. Organizational Efficiency | positive | high | ESG ratings |
approximately 0.14 rating levels
0.48
|
| The effect of IMDP participation on ESG performance operates through higher innovation efficiency. Innovation Output | positive | high | innovation efficiency |
0.48
|
| The effect of IMDP participation on ESG performance operates through improved cost management, consistent with capability upgrading and resource reallocation toward sustainability-related activities. Organizational Efficiency | positive | high | cost management |
0.48
|
| The positive effect of IMDP participation on ESG performance is stronger for firms under intense competitive pressure. Organizational Efficiency | positive | high | ESG ratings |
0.48
|
| The positive effect of IMDP participation on ESG performance is stronger for firms at the growth stage. Organizational Efficiency | positive | high | ESG ratings |
0.48
|
| The positive effect of IMDP participation on ESG performance is stronger in capital-scarce industries. Organizational Efficiency | positive | high | ESG ratings |
0.48
|
| IMDPs reduce greenwashing. Decision Quality | positive | high | greenwashing |
0.48
|
| IMDPs lower ESG rating uncertainty. Decision Quality | positive | high | ESG rating uncertainty |
0.48
|
| Intelligent manufacturing policies can generate economically meaningful benefits by improving firms’ sustainability performance and the credibility of ESG information, which are central to capital allocation and the effectiveness of green governance. Organizational Efficiency | positive | medium | sustainability performance and credibility of ESG information |
0.29
|