0 cumulative citations
View corpus contextChinese listed firms under intense institutional pressure invest more in capital‑intensive decarbonisation but pull back on voluntary cultural and policy measures; firms with stronger AI capabilities respond with even greater hard investments and are less likely to abandon soft practices.
The Impact of Institutional Pressures on Firms' Low‐Carbon Behaviors: A Configuration Approach
openalex
correlational
medium evidence
7/10 relevance
Summary only
summary available; pdf_status=paywall
DOI
Source
PDF
Among Chinese listed firms (2007–2022), stronger institutional pressures raise adoption of capital/technology ('hard') low‑carbon measures but are associated with fewer organizational/'soft' measures, while higher firm AI capability amplifies the pressure→hard investment link and cushions the pressure→soft practice decline.
Citation observations
Cumulative provider counts captured on specific dates; providers are never combined.
ABSTRACT Although the institutional environment is recognized as crucial for firms' low‐carbon development, how configurations of institutional pressures collectively shape firms' low‐carbon behaviors remains underexplored. Drawing on institutional theory, configuration theory, and resource‐based view, this study investigated the impact of institutional pressure configurations on firms' low‐carbon behaviors and the moderating role of artificial intelligence (AI) capability. The study used cluster analysis, analysis of variance (ANOVA), and OLS regressions with panel data from Chinese A‐share listed firms (2007–2022) to verify our research propositions. The results reveal three distinct firm clusters based on institutional pressure profiles. These configurations exert varying effects on both hard and soft low‐carbon behaviors. Generally, more intense pressures promote greater hard low‐carbon behaviors; conversely, high‐pressure firms are associated with significantly fewer soft low‐carbon behaviors compared to low‐pressure firms. Furthermore, a firm's AI capability enhances the positive influence of more intense institutional pressure on the adoption of hard low‐carbon behaviors and mitigates the negative influence of high institutional pressure on the adoption of soft low‐carbon behaviors. This study extends institutional perspectives on firms' environmental behaviors and provides actionable guidance for low‐carbon management.
Summary
Main Finding
Firms sort into three distinct clusters by institutional pressure profiles, and these configurations differentially affect low‑carbon behaviors. Stronger institutional pressures generally increase firms' adoption of "hard" low‑carbon behaviors (capital/technology investments), but high‑pressure firms adopt fewer "soft" low‑carbon behaviors (policies, culture, training, voluntary practices) than low‑pressure firms. A firm's AI capability strengthens the positive effect of intense institutional pressure on hard behaviors and mitigates the negative effect of high pressure on soft behaviors.
Key Points
- The study integrates institutional theory, configuration theory, and the resource‑based view to examine how combinations of institutional pressures shape firm-level low‑carbon actions.
- Cluster analysis identifies three firm groups with distinct institutional‑pressure profiles (e.g., low, high, and more intense/mixed pressure clusters).
- More intense institutional pressures → greater uptake of hard low‑carbon measures.
- High institutional pressure → significantly fewer soft low‑carbon measures compared with low‑pressure firms.
- AI capability is a moderating resource: it amplifies the positive pressure→hard‑behavior link and cushions the negative pressure→soft‑behavior link.
Data & Methods
- Data: Panel data on Chinese A‑share listed firms, 2007–2022.
- Methods:
- Cluster analysis to classify firms by institutional pressure profiles.
- Analysis of variance (ANOVA) to compare cluster differences.
- OLS panel regressions testing effects of pressure configurations on low‑carbon behaviors and interactions with firm AI capability.
- Theoretical framing: institutional theory (pressures), configuration theory (combinations of pressures), resource‑based view (AI as a firm capability/resource).
Implications for AI Economics
- AI as an enabling resource: AI capabilities increase firms' capacity to respond to regulatory and societal pressures by improving monitoring, process optimization, and technology adoption—raising the returns to regulatory pressure for hard investments.
- Policy design: Regulators aiming to accelerate hard measures (e.g., emissions‑reducing capital investments) may achieve larger effects when combining pressure with programs that build firms' AI capabilities (training, subsidies for AI adoption). For soft practices, support for AI can prevent pressure from crowding out voluntary/organizational initiatives.
- Strategic firm behavior and competition: Firms with AI capabilities gain a strategic complement to environmental regulation, potentially widening productivity and compliance gaps between AI‑enabled and non‑enabled firms. This affects market dynamics and transition costs.
- Empirical/quantitative modeling recommendations:
- When modeling policy impacts on firm decarbonization, include interaction terms between institutional intensity and firm AI capability.
- Account for heterogeneity in institutional configurations rather than treating pressure as unidimensional.
- Evaluate welfare and adoption cost implications of combining regulation with AI diffusion policies.
- Research directions: identify causal effects (IV, diff‑in‑diff natural experiments), test generalizability beyond Chinese listed firms, decompose which AI functions (analytics, automation, forecasting) drive the moderating effects, and assess long‑run productivity and emissions outcomes.
Assessment
Paper Typecorrelational
Evidence Strengthmedium — Uses a long panel of Chinese A‑share listed firms (2007–2022) and multi-method analysis (clustering + panel OLS with interactions), which supports credible associations and heterogeneity analysis; however, the lack of a clear causal identification strategy (no IV/DID/exogenous shock) leaves results vulnerable to endogeneity, reverse causality, omitted variables, and measurement/proxy concerns.
Methods Rigormedium — Appropriate exploratory approach (configuration/clustering) and standard panel regressions for hypothesis testing, including interaction tests of AI capability; but rigor depends on unspecified details (choice and validation of clustering algorithm, cluster stability, inclusion of fixed effects, control variables, robustness checks, and measurement validity of 'AI capability' and 'soft/hard' behaviors), and key causal threats are not addressed.
SamplePanel of Chinese A‑share listed firms observed 2007–2022; firms are classified into three clusters based on institutional pressure profiles; outcomes are firm‑level low‑carbon behaviors disaggregated into 'hard' (capital/technology investments) and 'soft' (policies, culture, training, voluntary practices); firm AI capability is included as a moderating resource (specific proxies for AI capability and sample size are not provided in the supplied text).
Themesadoption innovation governance
IdentificationObservational associations using cluster analysis to classify firms by institutional-pressure profiles and OLS panel regressions (with interaction terms) to estimate relationships between pressure configurations, AI capability, and low‑carbon behaviors; no exogenous variation (IV), natural experiment, or difference‑in‑differences strategy reported.
GeneralizabilitySample limited to Chinese A‑share listed firms (may not generalize to private, small, or non‑listed firms), Context‑specific institutional and regulatory environment in China may limit transferability to other countries, Results may differ across industries and firm sizes (sectoral heterogeneity not detailed), Findings rely on the validity of chosen proxies for 'AI capability' and for soft vs hard low‑carbon behaviors, Observational design limits causal generalization beyond detected associations
Claims (6)
| Claim | Direction | Outcome | Confidence & Evidence | Details |
|---|---|---|---|---|
| Firms sort into three distinct clusters based on their institutional-pressure profiles. Market Structure | mixed | Institutional-pressure configuration and firm-group classification |
Reading fidelity
high
Study strength
medium
|
not reported
|
| More intense institutional pressures are associated with greater adoption of hard low-carbon behaviors, such as capital and technology investments. Adoption Rate | positive | Adoption of hard low-carbon measures, including capital and technology investments |
Reading fidelity
high
Study strength
medium
|
not reported
|
| High institutional pressure is associated with significantly fewer soft low-carbon behaviors than low institutional pressure. Adoption Rate | negative | Adoption of soft low-carbon practices, including policies, organizational culture, training, and voluntary practices |
Reading fidelity
high
Study strength
medium
|
not reported
|
| Firm AI capability strengthens the positive association between intense institutional pressure and hard low-carbon behaviors. Adoption Rate | positive | Adoption of hard low-carbon behaviors under institutional pressure |
Reading fidelity
high
Study strength
medium
|
not reported
|
| Firm AI capability mitigates the negative association between high institutional pressure and soft low-carbon behaviors. Adoption Rate | positive | Adoption of soft low-carbon behaviors under high institutional pressure |
Reading fidelity
high
Study strength
medium
|
not reported
|
| Institutional pressure should be modeled as a multidimensional configuration rather than as a single unidimensional construct when analyzing firm decarbonization. Task Allocation | mixed | Differential adoption of hard and soft low-carbon behaviors across institutional-pressure configurations |
Reading fidelity
high
Study strength
medium
|
not reported
|