Enterprise AI Analysis

Spatial Durbin Model & Hybrid Weight Matrix

Our empirical analysis utilizes panel data from 279 Chinese prefecture-level cities over 2011-2020. To accurately capture the spatial correlation inherent in enterprise distribution, we employ the Spatial Durbin Model (SDM), which accounts for spatial lag in both the dependent and independent variables.

The core explanatory variable, digital economy (digi), is a comprehensive index derived using the entropy weight method from five indicators including telecom business income and digital financial inclusion. Manufacturing agglomeration (agg) is measured by the ratio of industrial enterprises in each region to the overall number. Control variables include economic development, fixed asset scale, openness, education, government intervention, and financial development.

A hybrid economic distance weight matrix is constructed to reflect both geographical proximity and economic disparity between cities, ensuring robust and unbiased estimations.

Positive Agglomeration & Strong Spatial Spillover

The Moran's I test confirms significant positive spatial autocorrelation for manufacturing agglomeration levels across Chinese cities, validating the use of spatial econometric models. The SDM with fixed effects is confirmed as the most suitable model.

Our results show that the digital economy has a statistically significant positive direct effect (0.0990) on local manufacturing agglomeration. Crucially, it also exerts a substantial positive indirect (spillover) effect (6.196) on neighboring regions. This indicates that digital transformation not only concentrates enterprises locally but also profoundly influences the spatial patterns of adjacent areas, challenging the "death of distance" hypothesis.

Scale, Regional, and High-Tech Zone Dynamics

Enterprise Scale: The digital economy's agglomeration effect is significantly greater for large-scale manufacturing enterprises compared to small-scale ones. Large firms leverage digital dividends, scale economies, and network effects more effectively.

Regional Differences: The direct agglomeration effect is largest in the western region, while the indirect (spillover) agglomeration effect is greatest in the central region. This reflects policy impacts (e.g., "East Calculation, West Calculation") and central regions acting as amplifiers for digital economy spillovers.

High-Tech Zones: Cities with established high-tech zones experience a greater agglomeration effect from the digital economy, highlighting these zones as critical carriers for digital development and innovation.

Mechanisms: Market Potential, Resource Allocation & Economies of Scale

Our mechanism analysis confirms three key pathways:

1. Market Potential: The digital economy enhances market transparency, activates long-tail markets, and expands consumer reach, thereby increasing the market potential that drives enterprise agglomeration.

2. Resource Allocation: Digital technologies reduce search costs, improve information flow, and guide efficient resource allocation (labor, capital, technology) to productive industries, fostering agglomeration locally, although a "siphon effect" can reduce efficiency in neighboring areas.

3. Economies of Scale: With increased users and data, the digital economy significantly lowers unit costs for manufacturing enterprises, attracting further agglomeration. This effect is strong locally but can lead to a "siphon effect" for neighboring regions lacking core digital resources.

Strategic Policy & Enterprise Guidance

The digital economy significantly enhances the spatial agglomeration of manufacturing enterprises, both directly and through powerful spatial spillovers, validating our hypotheses and calling for strategic responses.

Policy Implications:

• Strengthen New Infrastructure: Invest in broadband, 6G, data centers to foster clustering.
• Differentiated Investment: Prioritize central/western regions for infrastructure and R&D funds to balance development.
• Empower SMEs: Develop sharing platforms and low-code tools to address data access and cost challenges for small enterprises.

These strategies will enable governments and enterprises to navigate the spatial reconstruction of manufacturing in the digital age.