Listed agricultural firms in China that adopt AI register sizable productivity gains driven by labor reallocation and cross‑firm resource sharing rather than immediate boosts to university–industry R&D; benefits are concentrated among growing firms and in regions exposed to greater natural risk.
Artificial intelligence and the sustainable development of agricultural enterprises: a total factor productivity perspective
AI adoption among Chinese listed agricultural firms raises firm-level total factor productivity, chiefly by optimizing labor structures and enabling inter-firm resource sharing rather than by boosting university–industry collaborative R&D.
From the perspective of sustainable agricultural development, the adoption of artificial intelligence (AI) not only improves factor allocation efficiency but also constitutes a critical economic foundation for efficiency-driven sustainable growth in agriculture by optimizing resource utilization and strengthening risk-management capacity. Using panel data on agricultural firms listed on the Shanghai and Shenzhen A-share markets from 2007 to 2023, this study applies a multidimensional fixed-effects model to estimate the impact of AI on firms’ total factor productivity (TFP). The empirical results demonstrate that AI significantly enhances TFP. However, mechanism analysis reveals a structural divergence in transmission pathways: while AI fosters productivity growth mainly by optimizing labor structures and facilitating inter-firm resource sharing, it has yet to significantly promote university-industry collaborative R&D capabilities. Heterogeneity analysis further indicates that these productivity gains are more pronounced among firms in their growth stage and in regions facing higher natural risks. Overall, the expanding use of AI is reshaping agricultural production systems and has emerged as a key driver of high-quality development in the sector. Within an efficiency-driven sustainability framework, continued advances in AI are expected to play a pivotal role in achieving a dynamic alignment among the objectives of efficiency, environmental performance, and long-term sustainability in agriculture.
Summary
Main Finding
AI adoption by listed agricultural firms in China significantly increases total factor productivity (TFP). The productivity gains operate mainly through optimizing labor structures and enabling inter-firm resource sharing, rather than via enhanced university–industry collaborative R&D. Gains are larger for firms in the growth stage and for firms operating in regions with higher natural risk.
Key Points
- AI adoption → meaningful, positive effect on firm-level TFP in agriculture.
- Mechanisms:
- Primary channels: labor-structure optimization (e.g., more productive allocation of human capital) and improved inter-firm resource sharing.
- Limited channel: no significant increase in university–industry collaborative R&D capability detected.
- Heterogeneity:
- Stronger TFP improvements for firms in the growth stage versus mature firms.
- Firms in regions with higher natural risk exhibit larger productivity gains from AI.
- Interpretation: AI is reshaping agricultural production systems and acting as a driver of efficiency-led, high-quality agricultural development, especially where robustness to natural risk is valuable.
Data & Methods
- Sample: Panel data on agricultural firms listed on the Shanghai and Shenzhen A-share markets, covering 2007–2023.
- Estimation: Multidimensional fixed-effects model (controls for multiple unobserved heterogeneities) to identify the impact of AI adoption on firm TFP.
- Empirical strategy: Direct estimation of AI’s effect on TFP plus mediation (mechanism) and heterogeneity analyses to probe channels and differential returns across firm stages and regional risk environments.
Implications for AI Economics
- AI as a productivity-enhancing general-purpose technology in agriculture:
- Raises TFP primarily via reallocation and coordination channels (labor mix, resource sharing) rather than immediate R&D ecosystem strengthening.
- Policy and modeling implications:
- Policies should prioritize complementary investments that amplify labor-structure gains (training, human capital re-skilling) and reduce frictions to inter-firm resource sharing (data platforms, supply-chain linkages).
- To realize longer-run innovation spillovers, targeted support is needed to strengthen university–industry linkages and collaborative R&D around AI in agriculture.
- Heterogeneous returns imply that diffusion models and cost–benefit analyses must account for firm life-cycle stage and regional exposure to natural risks when predicting welfare and productivity impacts.
- Research implications:
- Further work should quantify the size of spillovers from inter-firm sharing, trace longer-run effects on innovation capacity, and explore complementarities between AI, environmental outcomes, and sustainable resource use.
- Sustainability framing:
- Within an efficiency-driven sustainability framework, AI can help align productivity, environmental performance, and resilience, but realizing this alignment requires complementary policies (training, R&D networks, data/infrastructure, and risk-mitigation tools).
Assessment
Paper Typequasi_experimental
Evidence Strengthmedium — Long panel of firm-level data and multidimensional fixed effects strengthen causal claims by removing many time-invariant and coarse time-varying confounders, and mediation/heterogeneity analyses are consistent with proposed channels; however, adoption is plausibly endogenous (more productive or better-managed firms may self-select into AI), no clear exogenous instrument or natural experiment is described, and measurement of 'AI adoption' and TFP may introduce additional bias.
Methods Rigormedium — Methods use advanced panel techniques (multiple fixed effects) and explicit mechanism testing, indicating careful empirical work; but the absence of an explicit exogenous source of variation (IV, randomized adoption, or event-study around plausibly exogenous shocks) and limited detail on TFP estimation and AI-adoption measurement reduce the methodological rigor relative to a high-identification causal study.
SampleFirm-year panel of agricultural firms listed on the Shanghai and Shenzhen A-share exchanges, covering 2007–2023; firm-level TFP is the outcome, with a binary/continuous measure of AI adoption and firm/regional covariates used for mediation and heterogeneity analyses.
Themesproductivity adoption
IdentificationPanel multidimensional fixed-effects regression leveraging within-firm variation in AI adoption over 2007–2023 (controls for firm, time, industry/sector, and regional effects) plus mediation analysis to test channels (labor-structure reallocation and inter-firm resource sharing) and heterogeneity tests by firm life-cycle stage and regional natural risk.
GeneralizabilityRestricted to listed (public) agricultural firms in China — excludes smallholders, cooperatives, and informal producers, Sector-specific (agriculture) — findings may not transfer to manufacturing or services, China-specific institutional, regulatory, and market context may limit transferability to other countries, Listed firms are likely larger and more capitalized than average agricultural producers, biasing external validity, AI-adoption definition (likely based on disclosures/investment) and TFP measurement choices may not generalize to alternative operationalizations, Time window up to 2023 — effects could change as AI matures or diffusion widens
Claims (10)
| Claim | Direction | Confidence | Outcome | Details |
|---|---|---|---|---|
| The adoption of AI improves factor allocation efficiency and constitutes a critical economic foundation for efficiency-driven sustainable growth in agriculture by optimizing resource utilization and strengthening risk-management capacity. Firm Productivity | positive | medium | factor allocation efficiency / total factor productivity (TFP); risk-management capacity (inferred from heterogeneous effects by natural-risk regions) |
0.29
|
| This study uses panel data on agricultural firms listed on the Shanghai and Shenzhen A-share markets from 2007 to 2023 and applies a multidimensional fixed-effects model to estimate the impact of AI on firms’ total factor productivity (TFP). Firm Productivity | null_result | high | study design / estimation of AI impact on total factor productivity (TFP) |
0.48
|
| AI significantly enhances firms' total factor productivity (TFP). Firm Productivity | positive | high | total factor productivity (TFP) |
statistically significant
0.48
|
| AI fosters productivity growth mainly by optimizing labor structures. Firm Productivity | positive | medium | TFP (via labor structure / labor composition) |
0.29
|
| AI fosters productivity growth by facilitating inter-firm resource sharing. Firm Productivity | positive | medium | TFP (via inter-firm resource sharing) |
0.29
|
| AI has not yet significantly promoted university–industry collaborative R&D capabilities. Innovation Output | null_result | medium | university–industry collaborative R&D capability (and its contribution to TFP) |
no statistically significant effect reported
0.29
|
| Productivity gains from AI are more pronounced among firms in their growth stage. Firm Productivity | positive | medium | total factor productivity (TFP) by firm life stage (growth stage) |
heterogeneous effect (larger for growth-stage firms)
0.29
|
| Productivity gains from AI are more pronounced in regions facing higher natural risks. Firm Productivity | positive | medium | total factor productivity (TFP) by regional natural-risk level |
heterogeneous effect (larger in higher natural-risk regions)
0.29
|
| The expanding use of AI is reshaping agricultural production systems and has emerged as a key driver of high-quality development in the sector. Firm Productivity | positive | medium | sectoral development quality / high-quality development in agriculture |
0.29
|
| Within an efficiency-driven sustainability framework, continued advances in AI are expected to play a pivotal role in achieving a dynamic alignment among efficiency, environmental performance, and long-term sustainability in agriculture. Firm Productivity | positive | speculative | alignment of efficiency, environmental performance, and long-term sustainability (projected / speculative outcome) |
0.05
|