The Commonplace
Home Papers Evidence Explore Trends Syntheses Digests References Docs 🎲 Workforce Futures
← Papers
Direction, evidence grade, and study type are AI-generated labels (gpt-5-mini), not human-verified. Syntheses are LLM-written. "Tensions" are machine-detected candidates, not confirmed contradictions. A research-acceleration tool, not peer review. How this is built →

Chinese cities jointly designated as big-data and smart-city pilots register meaningful gains in green productivity, with spillovers; the dual-policy package lifts green total factor productivity mainly by improving energy efficiency, spurring digital innovation and concentrating green-oriented industries, and sequencing digital before intelligent implementation performs best.

Digital and intelligent policy instrument portfolio and green transformation of cities: evidence from China
Xinyi Wang, Kang Wang, Xianbing Cao · August 03, 2026 · Humanities and Social Sciences Communications
openalex quasi_experimental medium evidence 7/10 relevance Full text usable extracted full text DOI Source PDF

Structured author observations

Linked only from stored provider relations; the raw author line above is never matched by name.

OpenAlex

Latest observation:

  1. Xinyi Wang provider ID
  2. Kang Wang provider ID
  3. Xianbing Cao provider ID

Semantic Scholar

Latest observation:

  1. Xinyi Wang provider ID
  2. Kang Wang provider ID
  3. Xianbing Cao provider ID
Using a staggered DID on 251 Chinese cities (2006–2022), the paper finds that cities concurrently designated as big-data and smart-city pilots experience higher green total factor productivity, with positive spatial spillovers and mechanisms operating through reduced energy intensity, enhanced digital technological innovation, and industrial agglomeration.

Citation observations

Cumulative provider counts captured on specific dates; providers are never combined.

In the digital-economy era, can policies represented by the national big data comprehensive pilot zones and the smart city pilot program work in tandem to effectively foster urban green development? Using panel data for 251 Chinese prefecture-level cities from 2006-2022 and a staggered DID model, this paper evaluates the multidimensional impacts of “digital-intelligent” policy collaboration on regional green transition. The findings are as follows: first, the dual-pilot policy significantly raises green total factor productivity in pilot cities, promotes local green transition, and generates marked green spatial spillovers; Second, the green co-empowerment effect of digital-intelligent policies is closely tied to regional characteristics and is further shaped by talent agglomeration and the strength of intellectual property protection; Third, mechanism analysis shows that the collaboration advances urban green development by improving energy consumption levels, enhancing digital technological innovation capacity, and strengthening industrial agglomeration; Fourth, heterogeneity analysis indicates that the empowering effect is more pronounced in regions with higher digital-economy development, stronger new-energy transitions, greater fiscal capacity, and service-oriented industrial structures; Fifth, further evidence shows that the empowering effect is strongest when green technological innovation surpasses a certain level and environmental regulation is of moderate intensity. In addition, compared with single pilots, dual pilots better promote urban green transition, and a sequencing strategy that implements “digital” before “intelligent” achieves superior outcomes. These results offer important guidance for local governments to coordinate multi-policy portfolios, harness synergistic policy effects, and accelerate economy-wide green transition.

Summary

Main Finding

The joint rollout of digital (national big data comprehensive pilot zones, BDPZ) and intelligent (national smart city, SC) policy pilots in China significantly and causally increases urban green total factor productivity (GTFP). Dual-pilot cities show stronger green transformation than single-pilot cities; effects include positive spatial spillovers to neighboring cities. The green-empowerment operates mainly via improved energy-use efficiency, boosted digital technological innovation, and industry agglomeration. Effects are conditional on regional characteristics (talent agglomeration, IP protection) and are strongest when green technological innovation is above a threshold and environmental regulation is of moderate intensity. Sequencing “digital” before “intelligent” yields better outcomes.

Key Points

  • Causal identification: The paper uses staggered difference-in-differences (DID) exploiting exogenous, time-varying adoption of BDPZ and SC pilots across 251 prefecture-level Chinese cities (2006–2022).
  • Primary outcome: Green total factor productivity (GTFP) — the study reports significant positive effects of dual pilots on GTFP and urban green transition.
  • Mechanisms:
    • Energy efficiency: Digital-intelligent integration reduces energy intensity via better sensing, scheduling, and automated/optimized operations.
    • Digital technological innovation: The policy bundle stimulates demand for and supply of digital R&D and commercialization, accelerating green-capable tech.
    • Industrial agglomeration: Lower transaction/search costs and improved coordination concentrate green/digital firms, speeding innovation diffusion.
  • Heterogeneity and moderators:
    • Larger effects where talent pools and stronger IP protection exist.
    • Larger effects in regions with higher baseline digital-economy development, more active new-energy transition, greater fiscal capacity, and service-oriented industrial structures.
    • Nonlinearities: the dual-pilot effect intensifies only after green technological innovation crosses a threshold; environmental regulation that is too weak or too strict reduces effectiveness — moderate regulation is optimal.
  • Spatial effects: Positive spillovers to neighboring cities (spatial lag/SAR analysis).
  • Policy design notes: Dual pilots outperform single pilots; sequencing matters — implementing data infrastructure/policy first (BDPZ) then intelligent applications (SC) is most effective.

Data & Methods

  • Sample: Panel of 251 Chinese prefecture-level cities, 2006–2022.
  • Identification strategy:
    • Staggered DID to estimate causal impact of policy adoption timing across cities.
    • Staggered DID panel threshold models to detect nonlinear (threshold) moderation by green technological innovation and environmental regulation.
    • Staggered DID spatial autoregressive (SAR) models to capture spatial spillovers.
  • Robustness: The paper contrasts single vs dual pilot effects, explores mechanism variables (energy intensity, measures of digital innovation, industrial agglomeration), and conducts heterogeneity analyses by regional characteristics (talent, IP protection, fiscal strength, industrial structure).
  • Outcomes and mediators: GTFP as main dependent variable; mediators include energy consumption measures, counts/indicators of digital technological innovation, and industry concentration/agglomeration indices.

Implications for AI Economics

  • Complementarity of data infrastructure and AI/application-layer policy:
    • The strongest green gains arise when data governance and infrastructure (BDPZ) precede and enable AI-driven intelligent systems (SC). For AI economists, this underscores that investments in data platforms and legal regimes (data sharing, standards) can be binding constraints for large-scale AI benefits.
  • Demand-induced AI innovation:
    • Policy-driven market demand (smart-city deployments, public-sector AI use-cases) stimulates private R&D and commercialization of AI and related green technologies. Models of innovation in AI economics should incorporate demand-pull from public-sector platforms and pilot programs.
  • Talent and IP as critical complements:
    • Returns to AI and data policies are amplified by talent agglomeration and credible IP protection. Policy evaluations should account for human capital and IP institutions when projecting AI-driven growth or green benefits.
  • Energy and rebound considerations:
    • While AI-enabled systems can reduce energy intensity at the urban/system level (optimization, scheduling), AI itself is energy- and compute-intensive. AI economics models must track net energy effects (savings from system optimization vs. energy cost of compute) and consider thresholds where AI yields net green gains.
  • Spatial and agglomeration dynamics:
    • AI policy impacts propagate spatially—successful pilots create spillovers to neighboring regions via firm relocation, knowledge diffusion, and platform effects. Spatial equilibrium models and regional policy analyses are important for assessing distributional consequences of AI deployment.
  • Policy sequencing and portfolio design:
    • Sequencing matters: prioritize data availability and governance before scaling AI applications. For policymakers aiming to harness AI for sustainability, coordinated multi-policy portfolios (data + application + regulation + talent/IP) outperform isolated interventions.
  • Measurement and evaluation:
    • The paper’s use of GTFP, staggered DID, threshold, and SAR models is a useful empirical toolkit for AI economics research evaluating policy packages. Future AI-economics work should adopt similarly careful causal designs and explore nonlinearity and spillovers.
  • Transferability and constraints:
    • Findings provide a template for developing countries, but outcomes will depend on local capacity (digital readiness, regulatory environment, workforce). AI economics should integrate institutional and capacity heterogeneity when generalizing results.

Caveat: This summary is based on an unedited manuscript in press (accepted July 24, 2026). Results are compelling but should be interpreted with usual caution pending final publication and peer-review artifacts.

Assessment

Paper Typequasi_experimental Evidence Strengthmedium — The paper leverages a long panel and staggered-DID variation with additional spatial and threshold specifications and mechanism tests, which strengthens causal claims versus simple correlations; however, assignment to pilot programs is plausibly endogenous (non-random political/economic selection), pre-trend and treatment heterogeneity issues inherent to staggered DID may remain, and results rest on observational administrative policy rollout rather than experimental variation. Methods Rigormedium — Appropriate modern quasi-experimental tools are used (staggered DID, spatial lag, threshold analysis) and multiple channels are examined, but the write-up (as supplied) does not show detailed diagnostics for parallel trends, placebo tests, convincing addressing of selection into pilot status, alternative identification strategies (e.g., instrumenting or regression-discontinuity), or robustness to recent concerns about staggered-DID estimators. SamplePanel data of 251 Chinese prefecture-level cities observed from 2006–2022; treatment is city-level adoption of national big data comprehensive pilot zone (BDPZ) and/or national smart-city (SC) pilot designation (single vs. dual pilots, and sequencing); main outcome is green total factor productivity (GTFP); additional city-level controls and moderators include measures of talent agglomeration, intellectual property protection, digital-economy development, energy transition, fiscal capacity, and industrial structure; spatial neighborhood matrices used for SAR models. Themesproductivity innovation governance IdentificationStaggered difference-in-differences (DID) exploiting cross-city and over-time variation in designation as national big data comprehensive pilot zones (BDPZ) and national smart-city (SC) pilot programs across 251 Chinese prefecture-level cities (2006–2022); supplemented by staggered DID panel threshold models to test nonlinear/moderating effects and a spatial autoregressive (SAR) model to capture spatial spillovers; heterogeneity, mechanism (mediation) tests, and robustness checks reported. GeneralizabilityFindings are China-specific: institutional, regulatory and policy-design details (central pilot designations) may not transfer to other countries., Prefecture-level cities only — may not generalize to rural areas, smaller towns, or national-level effects., Pilot designation is a policy instrument with political selection; effects of equivalent private-sector AI adoption could differ., Outcome is green productivity; results do not directly speak to labor-market displacement, wages, or firm-level profitability., Time period 2006–2022 includes major policy and economic shifts in China; effects may differ in later periods or with different technology maturity.

Claims (11)

ClaimDirectionOutcomeConfidence & EvidenceDetails
The combined national big data comprehensive pilot zone and smart city pilot policies significantly increase green total factor productivity in Chinese pilot cities. Firm Productivity positive Green total factor productivity
Reading fidelity high
Study strength medium
n=251
0.48
The dual-pilot policy promotes the green transformation of cities. Firm Productivity positive Urban green transformation
Reading fidelity high
Study strength medium
n=251
0.48
The positive effect of the dual-pilot policy generates significant green spatial spillovers beyond the directly treated cities. Firm Productivity positive Green development or green transition in neighboring cities
Reading fidelity high
Study strength medium
n=251
0.48
The green co-empowerment effect of digital-intelligent policies varies with regional characteristics and is shaped by talent agglomeration and intellectual property protection. Firm Productivity mixed Green total factor productivity or urban green transformation
Reading fidelity high
Study strength medium
n=251
0.48
The dual-pilot policy advances urban green development partly by improving energy consumption levels. Organizational Efficiency positive Energy-use conditions or energy-use efficiency
Reading fidelity high
Study strength medium
n=251
0.48
The dual-pilot policy advances urban green development by enhancing digital technological innovation capacity. Innovation Output positive Digital technological innovation capacity
Reading fidelity high
Study strength medium
n=251
0.48
The dual-pilot policy advances urban green development by strengthening industrial agglomeration. Market Structure positive Industrial agglomeration
Reading fidelity high
Study strength medium
n=251
0.48
The positive effect of digital-intelligent policy collaboration is stronger in regions with more developed digital economies, stronger new-energy transitions, greater fiscal capacity, and more service-oriented industrial structures. Firm Productivity positive Urban green transformation or green total factor productivity
Reading fidelity high
Study strength medium
n=251
0.48
The effect of digital-intelligent policy collaboration is strongest when green technological innovation exceeds a threshold and environmental regulation is of moderate intensity. Firm Productivity mixed Urban green transformation or green total factor productivity
Reading fidelity high
Study strength medium
n=251
0.48
Dual-pilot cities achieve better urban green-transition outcomes than cities exposed to only a single digital or smart-city pilot. Firm Productivity positive Urban green transition
Reading fidelity high
Study strength medium
n=251
0.48
Implementing the digital pilot before the intelligent or smart-city pilot produces better green-transition outcomes than the reverse sequencing strategy. Firm Productivity positive Urban green transition
Reading fidelity high
Study strength medium
n=251
0.48

Notes