Abstract
Industrial development is the process by which economies learn how to produce new products and services. But how do economies learn? And who do they learn from? The literature on economic geography and economic development has emphasized two learning channels: inter-industry learning, which involves learning from related industries; and inter-regional learning, which involves learning from neighboring regions. Here we use 25 years of data describing the evolution of China's economy between 1990 and 2015—a period when China multiplied its GDP per capita by a factor of ten—to explore how Chinese provinces diversified their economies. First, we show that the probability that a province will develop a new industry increases with the number of related industries that are already present in that province, a fact that is suggestive of inter-industry learning. Also, we show that the probability that a province will develop an industry increases with the number of neighboring provinces that are developed in that industry, a fact suggestive of inter-regional learning. Moreover, we find that the combination of these two channels exhibit diminishing returns, meaning that the contribution of either of these learning channels is redundant when the other one is present. Finally, we address endogeneity concerns by using the introduction of high-speed rail as an instrument to isolate the effects of inter-regional learning. Our findings suggest that inter-regional and inter-industry learning played a role in China's great economic expansion.
Introduction
Economic development is essentially a learning process: how does a region "learn" to produce a new, sophisticated product it has never made before? During its historic 25-year growth spurt from 1990 to 2015, China provided a unique laboratory to study this phenomenon. As the nation multiplied its GDP per capita tenfold, its provinces underwent a massive diversification of industrial capabilities. This paper explores the mechanisms behind this diversification, identifying two primary "collective learning" channels—learning from related industries within the same province and learning from neighboring provinces.
The Mechanisms of Collective Learning
Economists generally point to two channels for knowledge transfer in regional development:
- Inter-industry learning: This occurs when a province leverages existing capabilities to move into related sectors. For example, a region with a strong textile industry (making shirts) finds it easier to develop the technical and supply-chain expertise required to manufacture pants.
- Inter-regional learning: This involves knowledge spilling over from a successful neighbor. A province may adopt new industrial practices by observing and interacting with adjacent regions that have already mastered those industries.
While both channels are recognized in economic literature, the challenge lies in proving causality. Do regions become similar because they learn from each other, or do they connect because they were already similar? To solve this "chicken and egg" problem, this study utilizes the expansion of China’s high-speed rail (HSR) network as a quasi-natural experiment to isolate the impact of physical connectivity on economic diversification.
Methodology and Data
The researchers employed a multi-stage analytical approach using firm-level data from the RESSET Financial Research Database (1990–2015) and macroeconomic indicators from the China Statistical Yearbook. They constructed an "industry space"—a bipartite network of provinces and industries—to map how frequently industries co-locate, which serves as a proxy for shared technical requirements.
Two key indicators were developed to measure learning potential:
- $\\omega$ (Density of active related industries): A measure of how many related industries already exist within a province.
- $\\Omega$ (Density of active neighboring provinces): A measure of how many neighbors are already successful in a specific industry.
To establish causality, a Difference-in-Differences (DID) analysis was applied to pairs of provinces connected by HSR. Because HSR routes were often determined by political goals rather than existing industrial similarity, they provided an ideal instrumental variable to see if increased connectivity forced industrial convergence.
Key Findings
Path-Dependency and the Substitution Effect
The research confirms that industrial entry is highly path-dependent. A province’s probability of entering a new industry is significantly higher when it already possesses "nearby" sectors in the industry space. However, the most striking discovery is the substitution effect between the two learning channels. Inter-industry and inter-regional learning exhibit diminishing returns; if a province already has a high density of related industries, the additional benefit of having a successful neighbor is reduced. These channels are essentially redundant—a province only needs one strong "teacher" (either internal or external) to acquire a new capability.
The Importance of Tacit Knowledge
The study found that the introduction of HSR—a passenger-focused service—significantly increased the industrial similarity of connected provinces. This suggests that the knowledge being transferred is tacit. Unlike codified blueprints that can be sent digitally, tacit knowledge requires face-to-face interaction between engineers, managers, and researchers to share uncodified expertise and nuances of production.
Productivity Gains
The impact of this learning is not just seen in the variety of industries, but in their performance. Industries in provinces connected by rail saw a significant increase in labor productivity (revenue per worker) when they were linked to other provinces where that specific industry was already established. The average productivity increase for these connected pairs was approximately CNY 98,713 (~USD 15k) per worker.
Conclusion
The evolution of China's economy demonstrates that regional diversification is driven by a combination of internal capabilities and external connectivity. While both inter-industry and inter-regional learning are powerful drivers, they act as substitutes for one another. As China continues to develop, the role of physical infrastructure like high-speed rail remains critical, not just for moving people, but for facilitating the face-to-face exchange of the tacit knowledge required for industrial innovation.
Future research may look toward more granular data—such as city or district-level interactions—to further decode the micro-mechanisms of these spillovers, such as labor mobility and supply chain integration.
Summary updated