Industry policies and technological innovation in artificial intelligence clusters: are central positions superior? – Nature

 

Report on AI Innovation, Policy, and Sustainable Development

1.0 Executive Summary

This report analyzes the relationship between government industrial policies and technological innovation within China’s Artificial Intelligence (AI) clusters. The analysis is framed within the context of the United Nations Sustainable Development Goals (SDGs), providing critical insights for achieving SDG 9 (Industry, Innovation, and Infrastructure), SDG 8 (Decent Work and Economic Growth), SDG 11 (Sustainable Cities and Communities), and SDG 17 (Partnerships for the Goals). Using patent data from 2011-2020, this study employs location quotient, social network analysis, and dynamic panel modeling to evaluate policy effectiveness.

Key Findings:

• Cluster Identification: Twenty-nine distinct AI innovation clusters were identified across China, forming the backbone of the nation’s AI development and contributing to the economic vitality of urban centers (SDG 11).
• Policy Effectiveness: Government industrial policies are found to be effective in promoting technological innovation within these AI clusters, directly supporting the targets of SDG 9.
• Role of Partnerships: Interregional cooperation is the primary mode of collaboration for these clusters, highlighting the importance of partnerships (SDG 17) in driving advanced industrial development.
• Centrality’s Paradoxical Effect: High network centrality, while indicative of a strong position in partnership networks, diminishes the positive impact of industrial policies. This suggests that over-centralization can lead to resource imbalances, posing a challenge to equitable and sustainable innovation growth (SDG 10).

2.0 Introduction: AI Innovation and the Global Goals

Artificial Intelligence (AI) is a core driving force of the new industrial revolution and a critical engine for achieving sustainable economic growth and enhancing innovation efficiency, aligning with the ambitions of SDG 8 (Decent Work and Economic Growth) and SDG 9 (Industry, Innovation, and Infrastructure). The development of AI innovation exhibits pronounced clustering characteristics, typically within major urban centers, making these clusters vital for building SDG 11 (Sustainable Cities and Communities). This report investigates the effectiveness of government industrial policies in fostering technological innovation in these AI clusters. Furthermore, it examines how a cluster’s position within innovation networks—a reflection of SDG 17 (Partnerships for the Goals)—moderates the impact of these policies. The case of China is selected to provide a scalable reference for other nations pursuing technology-driven sustainable development.

3.0 Methodology

3.1 Identification of AI Clusters for SDG Analysis

To identify the key geographical areas driving AI innovation, a three-step methodology was employed, combining spatial agglomeration metrics with network analysis to reflect the multi-faceted nature of industrial ecosystems.

1. Delimiting Spatial Agglomeration Areas: The Location Quotient (LQ) of innovation output (AI patent applications) was calculated for all prefecture-level administrative regions in China to identify areas with a specialized concentration in the AI industry.
2. Identifying Potential Clusters: A specialization threshold (the 80th percentile of patent applications) was applied to filter out regions with low innovation scale, ensuring that identified clusters represent significant hubs of activity relevant to achieving SDG 9.
3. Measuring Innovation Linkages (SDG 17): Using Social Network Analysis (SNA) on cooperative patents, the intraregional and interregional innovation linkages of potential clusters were measured. Regions demonstrating both internal and external collaboration were confirmed as AI clusters, underscoring the importance of partnerships.

3.2 Analytical Framework

A dynamic panel system generalised method of moments (System-GMM) model was used to investigate the relationship between policy, network position, and innovation, mitigating potential endogeneity issues.

• Dependent Variable (Technological Innovation): Measured by the number of invention patent applications, a key indicator of progress toward SDG 9.
• Independent Variable (Industrial Policy): Measured by the number of AI-related incentive policies enacted by local governments.
• Moderating Variable (Network Centrality): Measured using closeness centrality to reflect a cluster’s access to knowledge and resources within the innovation network, a tangible metric for evaluating the structure of partnerships under SDG 17.
• Control Variables: Economic Development (per capita GDP), Infrastructure Construction (per capita road area), and Innovation Environment (fiscal expenditures on science) were included to account for local conditions influencing SDG 8 and SDG 9.

4.0 Key Findings

4.1 Profile of China’s AI Innovation Clusters

The analysis identified 29 AI clusters, including Beijing, Shenzhen, Shanghai, and Hangzhou. These clusters are pivotal to national innovation strategies and serve as engines for urban economic development, directly contributing to SDG 11. A crucial finding is that the volume of interregional collaborations significantly exceeds intraregional collaborations for every cluster. This underscores that the modern innovation landscape, even for geographically defined clusters, relies heavily on broad partnership networks, reinforcing the principles of SDG 17.

4.2 Industrial Policy as a Catalyst for SDG 9

The regression analysis confirms a significant and positive relationship between industrial policies and technological innovation. This finding demonstrates that targeted government intervention, through financial and talent incentives, effectively stimulates the innovation required to build resilient infrastructure and foster inclusive and sustainable industrialization, as mandated by SDG 9. The policies help overcome market failures and de-risk the high-cost, long-term process of frontier technology development.

4.3 The Moderating Role of Network Centrality: A Challenge for Inclusive Growth

The report reveals a significant negative moderating effect of network centrality on the policy-innovation relationship. This means that for clusters with very high centrality, the positive impact of industrial policies is weakened. This phenomenon suggests a potential “curse of centrality,” where maintaining a central network position requires substantial resources, potentially crowding out local R&D investment and creating resource drains. This finding has direct implications for:

• SDG 10 (Reduced Inequalities): Over-concentration of policy benefits in highly central clusters could exacerbate regional disparities between central and peripheral innovation hubs.
• SDG 17 (Partnerships for the Goals): While partnerships are beneficial, their structure matters. Unbalanced networks can lead to inefficiencies that undermine the overall goal of shared progress.

5.0 Implications for Sustainable Development

5.1 Policy Recommendations for SDG 9 (Industry, Innovation, and Infrastructure)

To maximize progress on SDG 9, governments should continue to implement supportive industrial policies for high-tech sectors. However, these policies must be designed to foster a balanced innovation ecosystem. This includes:

• Establishing government-led funds for strategic R&D in core technologies.
• Implementing talent attraction and development programs to build the human capital necessary for innovation, contributing to SDG 4 (Quality Education) and SDG 8 (Decent Work).
• Ensuring policies are effectively implemented and reach the intended enterprises, as confirmed by the study’s policy penetration analysis.

5.2 Balancing Growth and Equity (SDG 8, 10, 11)

The finding that high network centrality can weaken policy effectiveness calls for a more nuanced approach to regional development. To ensure that the benefits of AI-driven growth are shared broadly, policymakers should:

• Avoid over-concentrating resources and policy focus on a few “superstar” clusters.
• Promote balanced development by strengthening the internal innovation capacity of all clusters, not just their external linkages.
• Design policies that foster sustainable economic structures within cities (SDG 11) and mitigate the risk of increasing regional inequality (SDG 10).

5.3 Fostering Effective Partnerships (SDG 17)

Given that interregional collaboration is dominant but can create resource imbalances, the focus should be on fostering effective and equitable partnerships. Governments can:

• Provide special subsidies for joint R&D projects that ensure mutual benefit and knowledge sharing.
• Establish structured platforms for industry-university-research cooperation.
• Dynamically adjust policies based on regular analysis of intra- and interregional collaboration patterns to ensure a healthy, balanced, and resilient national innovation network.

6.0 Conclusion

This report concludes that while government industrial policies are a potent tool for advancing technological innovation in AI clusters, their effectiveness is moderated by the structure of innovation networks. The identification of 29 AI clusters in China, the confirmation of a positive policy-innovation link, and the discovery of a negative moderating effect of network centrality provide critical, evidence-based insights. By framing these findings within the Sustainable Development Goals, this report offers actionable guidance for policymakers worldwide who aim to leverage technological advancement to build innovative, inclusive, and sustainable economies in line with the 2030 Agenda.

SDGs Addressed in the Article

1. SDG 8: Decent Work and Economic Growth

   • The article connects the development of Artificial Intelligence (AI) to economic growth, stating that AI clusters are “important drivers of… regional economic growth” and that AI is a “new engine for enhancing innovation efficiency.” It also quantifies the potential economic impact, noting the “market scale of the AI industry in China will reach 1.73 trillion yuan ($240.4 billion) by 2035.”

2. SDG 9: Industry, Innovation and Infrastructure

   • This is the central theme of the article. It directly investigates how “industry policies” can “promote technological innovation in the artificial intelligence clusters.” The entire study is framed around fostering innovation, upgrading industrial capabilities through AI, and building resilient innovation infrastructure through clusters and networks. The research uses “patents as a critical representation of innovation.”

3. SDG 11: Sustainable Cities and Communities

   • The research focuses on “AI clusters” which are geographically located in “administrative regions at the prefecture level and above,” effectively analyzing cities as hubs for innovation. The study identifies “twenty-nine AI clusters” and analyzes the “innovation network of interregional cooperation” between them, which relates to strengthening regional development planning and creating positive economic links between urban areas.

4. SDG 17: Partnerships for the Goals

   • The article heavily emphasizes the role of partnerships in driving innovation. It analyzes “interregional cooperation” as the “main form of collaboration for AI clusters” and examines “cooperative patents” between “firms, research institutions, and universities.” The study also explores the public-private partnership dynamic by assessing how “government intervention” and “industry policies” influence innovation within private enterprise clusters.

Specific SDG Targets

1. SDG 8: Decent Work and Economic Growth

   • Target 8.2: Achieve higher levels of economic productivity through diversification, technological upgrading and innovation, including through a focus on high-value added and labour-intensive sectors.
     
     The article directly addresses this by focusing on the AI industry, a high-value-added sector, and exploring how “technological innovation is critical to the high-quality development” and can serve as a “new engine for enhancing innovation efficiency.”

2. SDG 9: Industry, Innovation and Infrastructure

   • Target 9.5: Enhance scientific research, upgrade the technological capabilities of industrial sectors in all countries… encouraging innovation and substantially increasing the number of research and development workers… and public and private research and development spending.
     
     The article’s core analysis on how “industry policies can effectively promote technological innovation” aligns perfectly with this target. It measures innovation through patents and considers factors like “fiscal expenditures on science” and “government subsidies” which are forms of R&D spending.
   • Target 9.b: Support domestic technology development, research and innovation in developing countries, including by ensuring a conducive policy environment for, inter alia, industrial diversification and value addition to commodities.
     
     The study’s investigation into the “influence of IP [Industry Policies] on TI [Technological Innovation]” and its use of the “number of incentive policies enacted by local governments” as a variable directly relates to creating a conducive policy environment for domestic innovation.

3. SDG 11: Sustainable Cities and Communities

   • Target 11.a: Support positive economic, social and environmental links between urban, peri-urban and rural areas by strengthening national and regional development planning.
     
     The paper’s use of “social network analysis (SNA) to identify AI clusters” and its finding that “interregional cooperation is the main form of collaboration” directly examines the innovation links between cities, providing a basis for strengthening regional development planning.

4. SDG 17: Partnerships for the Goals

   • Target 17.6: Enhance North-South, South-South and triangular regional and international cooperation on and access to science, technology and innovation and enhance knowledge sharing on mutually agreed terms…
     
     The analysis of “interregional cooperation” and “cooperative patents” among the 29 identified clusters is a direct example of regional cooperation on science, technology, and innovation within a country (a form of South-South cooperation).
   • Target 17.17: Encourage and promote effective public, public-private and civil society partnerships…
     
     The research model, which examines the effect of “industry policies” (public) on “technological innovation” in AI clusters (private sector), is an analysis of public-private partnerships for innovation. The mediation analysis using “government support (GS)” further solidifies this link.

Indicators for Measuring Progress

1. For SDG 8 (Target 8.2)

   • Per capita gross domestic product (GDP): Mentioned as a control variable to measure “economic development (ED).”
   • Market scale of the AI industry: The article cites projections for the industry’s market size as evidence of its economic importance.

2. For SDG 9 (Targets 9.5 & 9.b)

   • Number of invention patent applications: Used as the primary dependent variable to measure “technological innovation (TI).”
   • Number of incentive policies: Used as the independent variable to measure “industry policies (IP)” and the conduciveness of the policy environment.
   • Fiscal expenditures on science and technology: Used as a control variable to measure the “innovation environment (IE).”
   • Government subsidies: Analyzed as a measure of “government support (GS)” to enterprises, indicating public R&D investment.

3. For SDG 11 (Target 11.a)

   • Network centrality (NC): A key metric from Social Network Analysis used to measure a cluster’s (city’s) position and connectedness within the interregional innovation network.
   • Volume of interregional collaborations: Calculated from cooperative patents to measure the strength of links between clusters.
   • Location Quotient (LQ): Used to identify the spatial agglomeration of the AI industry in specific cities.

4. For SDG 17 (Targets 17.6 & 17.17)

   • Number of cooperative patents: Used to identify and quantify collaborations between different organizations and regions.
   • Total volume of intraregional and interregional collaborations: Calculated to measure the extent of partnerships within and between clusters.
   • Social Network Analysis (SNA): The methodology itself serves as a tool to map and analyze the structure of partnerships.

Summary Table of SDGs, Targets, and Indicators

SDGs	Targets	Indicators Identified in the Article
SDG 8: Decent Work and Economic Growth	8.2: Achieve higher levels of economic productivity through diversification, technological upgrading and innovation.	Per capita gross domestic product (GDP) Market scale of the AI industry
SDG 9: Industry, Innovation and Infrastructure	9.5: Enhance scientific research, upgrade the technological capabilities of industrial sectors, and encourage innovation.	Number of invention patent applications Fiscal expenditures on science and technology Government subsidies to enterprises
	9.b: Support domestic technology development, research and innovation through a conducive policy environment.	Number of incentive policies enacted by local governments
SDG 11: Sustainable Cities and Communities	11.a: Support positive economic links between urban areas by strengthening regional development planning.	Network centrality (NC) of clusters/cities Volume of interregional collaborations Location Quotient (LQ) of industry in cities
SDG 17: Partnerships for the Goals	17.6: Enhance regional cooperation on and access to science, technology and innovation.	Number of cooperative patents Social Network Analysis (SNA) of innovation networks
	17.17: Encourage and promote effective public-private partnerships.	Analysis of industry policies’ (public) effect on enterprise clusters (private) Government support (subsidies) as a measure of partnership

Source: nature.com