TL;DR
This study maps the evolving spatial distribution of China's New Energy Vehicle (NEV) industry from 2008-2021, revealing a distinct "8"-shaped industrial axis dominated by coastal economic hubs. It identifies key drivers for NEV company location, highlighting the critical roles of government policy, innovation capacity, and tertiary industry development, while also noting the negative influence of production costs.
What Problem Does This Paper Address?
China's New Energy Vehicle (NEV) industry is a strategic pillar for its manufacturing upgrade and energy transition. However, its rapid growth has led to dramatic shifts in its spatial distribution, creating a need to understand these patterns and the factors driving them. Traditional automotive industry hubs don't necessarily translate to NEV success, suggesting a unique set of locational requirements for this emerging sector. This paper aims to fill this knowledge gap by analyzing the spatial evolution of the NEV industry and identifying the key determinants of where NEV-related companies choose to locate, providing crucial insights for regional development and industrial planning.
Methods
The researchers employed a multi-faceted approach to analyze the spatial patterns and influencing factors of China's NEV industry.
- Spatial Statistical Analysis:
- Global Spatial Autocorrelation (Moran's I): Used to determine if the distribution of NEV listed companies is random, clustered, or dispersed across China. They tested this using both Queen adjacency weights and inverse distance weights to understand the influence of neighboring regions versus simple spatial proximity.
- Kernel Density Analysis: Applied to visualize the intensity of NEV company agglomeration at a finer geographical scale, showing "hotspots" of concentration over time.
- Location Quotient (LQ): Calculated to measure the degree of specialization and regional industrial agglomeration of the NEV industry.
- Regression Analysis:
- Stepwise Regression: A backward stepwise regression method was used to identify the most significant factors influencing the number of NEV listed companies in each province. This method is particularly useful for selecting a parsimonious set of predictors from a larger pool of potential variables, helping to avoid multicollinearity.
Datasets
The primary dataset for this study consists of information on 459 A-share NEV concept-related listed companies in mainland China, registered and listed on the Shanghai and Shenzhen Stock Exchanges. Data was collected for the years 2008, 2013, 2018, and 2021.
In addition to company location data, the study utilized provincial-level statistical yearbook panel data for influencing factors. These data were primarily sourced from the China Statistical Yearbook and the China Automotive Industry Yearbook. For the regression analysis, data from 25 province-level administrative regions were used, excluding those with minimal or no automobile production due to data availability and reliability issues.
Key Insights and Ideas
- The "8" Shape Emerges: The NEV industry's spatial pattern is not a simple replication of traditional auto manufacturing. Instead, it's coalescing into a distinct "8"-shaped axis, driven by the Yangtze River Delta, Pearl River Delta, and Beijing-Tianjin-Hebei regions. This suggests a shift towards innovation and market-centric locations rather than solely relying on historical industrial bases.
- Traditional Hubs Lagging: Surprisingly, regions historically dominant in traditional auto manufacturing did not show a competitive advantage in the NEV sector. This implies that NEVs have a different "DNA" and require different locational endowments than their internal combustion engine predecessors.
- Policy is King (and Innovation Queen): Government policy strength and regional innovation capacity (measured by patents) are the most significant positive drivers for NEV company location. This underscores the crucial role of supportive government initiatives and a strong R&D ecosystem in attracting and fostering NEV development.
- Tertiary Sector Matters: A higher share of the tertiary industry in a province's economy also positively influences NEV company presence. This hints at the service-oriented and high-tech aspects of the NEV value chain, moving beyond traditional manufacturing paradigms.
- Cost Sensitivity is High: Per capita wage, a proxy for production cost, emerged as a significant negative factor. This indicates that NEV companies, despite their advanced nature, are still sensitive to labor costs, suggesting a blend of high-tech and labor-intensive characteristics.
- Roads Aren't Always a Boon: Counterintuitively, road density showed a negative correlation. The authors suggest this might be due to path dependency in traditional industries, where high road density could reinforce existing transportation modes, making the transition to NEVs more challenging.
Key Results
- Spatial Pattern: By 2021, NEV listed companies were heavily concentrated in the Yangtze River Delta (YRD), Pearl River Delta (PRD), and Beijing-Tianjin-Hebei (BTH) regions, accounting for over 72% of the national total. The overall spatial pattern formed an "8" shape, with the YRD and PRD as core agglomeration zones and BTH as a secondary cluster.
- Spatial Correlation: Global Moran's I analysis indicated a significant spatial correlation (agglomeration) of NEV listed companies, particularly when considering neighboring provinces (Queen adjacency weights). This correlation strengthened over time, peaking in 2021.
- Influencing Factors: Stepwise regression identified five key factors influencing the number of NEV listed companies:
- Positive Impact: Policy strength (X13), Patents (X14), Tertiary industry share (X4).
- Negative Impact: Per-capita wage (X11), Road density (X7).
- Model Fit: The regression model explained a substantial 88.9% of the variation in the number of NEV listed companies (R² = 0.889), indicating a strong predictive capability.
Limitations and Future Work
The authors acknowledge two main limitations:
- Data Scope: The study relied on data from listed companies, which may not fully capture the spatial distribution of the entire NEV industry. Many smaller, high-quality companies and production bases located in lower-tier cities were not included. Future work should incorporate a broader range of enterprise data and production base information for a more comprehensive view.
- Factor Exploration: While the study selected influential factors based on existing literature, it's possible that less common or industry-specific indicators were overlooked. Further research could explore additional data sources (e.g., regional or city-level data) and emerging indicators to refine the understanding of influencing factors.
The paper concludes by emphasizing that the NEV industry's development logic differs from traditional manufacturing. It suggests that governments should prioritize stimulating market demand, strengthening local innovation capacity, and fostering collaborative innovation, rather than solely focusing on physical resources.
Summary updated