← Papers

Global semiconductor trade has grown more clustered and polarized since 2010, forming a loose core–subcore–periphery map in which the US, China and key advanced economies remain central while several developing producers rise; diplomatic ties significantly shape trade links, but their influence fades where technological capabilities diverge and strengthens where economic-complexity gaps are larger.

Restructuring of the Global Chip Trade Network: Characteristic Evolution and Driving Factors

Lei Fu, Xiangyi Ding · January 30, 2026 · Systems

openalex correlational medium evidence 7/10 relevance Summary only summary available; pdf_status=paywall 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: August 24, 2026

Lei Fu provider ID
Xiangyi Ding provider ID

Semantic Scholar

Latest observation: August 24, 2026

Lei Fu provider ID
Xiangyi Ding provider ID

From 2010–2023 the global chip trade network became more clustered and polarized with a loose core–periphery (and emergent core–subcore–periphery) structure, and political relations are strongly associated with bilateral chip trade—an effect weakened by greater technological distance but strengthened by larger economic-complexity distance.

Citation observations

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

1 cumulative citations

OpenAlex · Observed August 24, 2026

View corpus context

1 cumulative citations

Semantic Scholar · Observed August 24, 2026

View corpus context

As the “brain” of the information industry and modern manufacturing, chips have emerged as a focal point in global competition over critical technologies. Based on global chip trade data from 2010 to 2023, this study employs social network analysis to investigate the structural evolution of the chip trade network and applies the quadratic assignment procedure (QAP) to examine the driving mechanisms of network reconstruction. The findings are as follows: First, the global chip trade network exhibits a loosely connected core-periphery structure, characterized by clustering and polarization, with a pronounced short-term deglobalization trend. Second, China, the United States, Germany, France, South Korea, and Singapore have long dominated central positions in competitive dynamics, while developing economies such as Mexico, Malaysia, and the Philippines have significantly risen in prominence in recent years. Third, the network takes on a core–subcore–periphery configuration with clearly delineated trade communities, reflecting a community-based, multi-centric, and hierarchical pattern. Fourth, political relations serve as a key driver of network restructuring, with their promotional effect on chip trade being negatively moderated by technological distance yet positively moderated by economic-complexity distance.

Summary

Main Finding

The global chip-trade network (2010–2023) has evolved into a loosely connected, community-based core–subcore–periphery structure with clustering and polarization and a short-term deglobalization trend. Geopolitical ties strongly reshape this network: political relations are a key driver of trade links, but their effect is weakened by technological distance and strengthened by economic-complexity distance.

Key Points

Network structure
- Loosely connected core–periphery overall; refined as a core–subcore–periphery configuration.
- Clear clustering and polarization; distinct trade communities (multi-centric and hierarchical).
- Evidence of short-term deglobalization (reduced overall connectivity/trade integration in recent years).
Central actors and mobility
- Longstanding central countries: China, United States, Germany, France, South Korea, Singapore.
- Rising prominence of several developing economies in recent years: Mexico, Malaysia, Philippines.
Drivers of network reconstruction (dyadic level)
- Political relations (diplomatic/political ties) are a primary positive driver of bilateral chip trade.
- Moderation effects:
  - Technological distance negatively moderates the positive effect of political relations (i.e., political ties matter less when partners are far apart technologically).
  - Economic-complexity distance positively moderates the political effect (i.e., political ties boost trade more when partners differ in economic complexity, suggesting complementarity).

Data & Methods

Data: bilateral global chip trade flows, 2010–2023.
Methods:
- Social network analysis to characterize macro and meso network structure (core–periphery tests, community detection, centrality measures, clustering/polarization metrics).
- Quadratic Assignment Procedure (QAP) regression to analyze dyadic determinants of trade ties while accounting for network dependence (permutation-based inference appropriate for relational data).
- Moderation analysis within QAP to test how technological distance and economic-complexity distance alter the political-relations → trade relationship.

Implications for AI Economics

Strategic importance of chips: chip-trade topology directly affects AI hardware access, costs, and deployment speed. Centralized cores and politically aligned communities create asymmetric access to compute resources.
Geopolitics matters: political relationships can be as important as market fundamentals for securing chip supplies; sanctions, alliances, and diplomatic shifts will materially influence AI capability diffusion.
Technology gaps constrain policy leverage: diplomatic ties alone may not overcome large technological gaps—recipient countries with low technological capability may still face barriers despite favorable politics.
Complementarity opportunity: political engagement is especially effective between partners with different economic complexity, suggesting targeted partnerships (e.g., between advanced-chip producers and complex-but-complementary economies) can expand supply chains.
Supply-chain resilience and industrial policy:
- Diversify sources beyond the core and build regional trade communities to mitigate short-term deglobalization risks.
- Invest in domestic tech capabilities to reduce negative moderation from technological distance and to move up the economic-complexity ladder.
- Consider policies that combine diplomatic outreach with technology transfer, capacity building, and industrial upgrading to secure more reliable access to chips for AI development.

(Methods note: QAP-based findings reflect relational associations, not strict causal proofs; results are robust to network dependence but subject to data and model specification limits.)

Assessment

Paper Typecorrelational Evidence Strengthmedium — The paper leverages a long panel of bilateral trade networks and appropriate network regression (QAP) to produce robust associations and to test moderation effects, which gives reasonably persuasive correlational evidence about drivers of network change; however, lack of exogenous variation, potential reverse causality (trade affecting political ties), and possible omitted variables limit causal claims. Methods Rigormedium — Use of social network analysis and QAP is methodologically appropriate for dependent network data and the long time span (2010–2023) strengthens temporal coverage; nonetheless, the analysis appears to rely on aggregate country-level measures, may not fully address endogeneity or measurement error in political/technology-distance variables, and likely lacks robustness checks using alternative identification strategies (e.g., instrumental variables, event-based designs). SampleAnnual, country-level bilateral trade data for semiconductors/chips from 2010 to 2023 forming weighted trade networks (nodes = countries/economies; edges = trade value or intensity of chip trade), with additional country-pair covariates including measures of political relations, technological distance, and economic complexity distance; covers both advanced economies and developing countries (exact country list not specified here). Themesinnovation governance IdentificationObservational analysis of annual bilateral chip trade networks (2010–2023) using social network analysis and quadratic assignment procedure (QAP) regressions; QAP permutations account for network dependence when estimating associations between trade ties and covariates (political relations, technological distance, economic-complexity distance), but no exogenous shocks or instruments are used to establish causality. GeneralizabilityFindings apply to international semiconductor trade at the country level and may not generalize to firm- or plant-level supply-chain dynamics., Results are specific to chips/semiconductors and should not be directly extrapolated to other sectors or downstream AI services without caution., Political-relation measures and ‘‘distance’’ metrics may be coarse and vary in validity across contexts, reducing external validity., Temporal trends up to 2023 may not capture very recent policy changes, export controls, or rapid technological shifts after 2023., Aggregation obscures heterogeneity within countries (e.g., role of multinational firms, regional production hubs).

Claims (6)

Claim	Direction	Outcome	Confidence & Evidence	Details
Based on global chip trade data from 2010 to 2023, this study employs social network analysis to investigate the structural evolution of the chip trade network and applies the quadratic assignment procedure (QAP) to examine the driving mechanisms of network reconstruction. Other	null_result	methodological approach / analytical design used to study chip trade network	Reading fidelity high Study strength medium	not reported 0.3
The global chip trade network exhibits a loosely connected core–periphery structure, characterized by clustering and polarization, with a pronounced short-term deglobalization trend. Market Structure	negative	network connectivity / core–periphery structure and short-term deglobalization trend	Reading fidelity high Study strength medium	not reported 0.3
China, the United States, Germany, France, South Korea, and Singapore have long dominated central positions in competitive dynamics of the global chip trade network. Market Structure	mixed	network centrality (competitive central positions) in chip trade	Reading fidelity high Study strength medium	not reported 0.3
Developing economies such as Mexico, Malaysia, and the Philippines have significantly risen in prominence in recent years within the global chip trade network. Market Structure	positive	rise in network prominence / centrality for specified developing economies	Reading fidelity high Study strength medium	not reported 0.3
The network takes on a core–subcore–periphery configuration with clearly delineated trade communities, reflecting a community-based, multi-centric, and hierarchical pattern. Market Structure	mixed	network community structure and tiered configuration	Reading fidelity high Study strength medium	not reported 0.3
Political relations serve as a key driver of network restructuring, with their promotional effect on chip trade being negatively moderated by technological distance yet positively moderated by economic-complexity distance. Market Structure	positive	dyadic chip trade ties (trade intensity) as influenced by political relations and moderated by technological/economic-complexity distance	Reading fidelity high Study strength medium	not reported 0.3