Data-Driven Detection and Characterization of Shocks in the Global Trade Network
Restricted (Penn State Only)
- Author:
- Jurado Guzman, Amanda Cristina
- Graduate Program:
- Civil Engineering
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- June 15, 2023
- Committee Members:
- Patrick Fox, Program Head/Chair
Alfonso Ignacio Mejia, Thesis Advisor/Co-Advisor
Soundar Kumara, Committee Member
Christine Kirchhoff, Committee Member - Keywords:
- Supply chain
Resilience
Global trade
Trade shocks - Abstract:
- The accurate and rigorous characterization of global import shocks – a sudden decrease in the inflows of a product to a country – is crucial for managing environmental, health, and economic crises that can propagate along global supply chains connections, such as the recent COVID-19 pandemic and the invasion of Ukraine. Here, we perform a rigorous characterization of the frequency and severity of import shocks in the global trade network for over 1200 products at an annual timescale for 1995-2018. Shocks are identified using a statistical model that detects breaks in the temporal autocorrelation of countries’ imports time series. We use the shock severity and frequency as well as resilience analysis to characterize shocks along different analysis units (products, sectors, and countries). For the resilience analysis, we estimate measures of resistance, resilience, and recovery time of countries to import shocks. Overall, the shocks' frequency is rising globally over time. Countries in Africa and Central America show a higher shock frequency across shocks of increasing magnitude. Shocks also reveal patterns of variation across products and sectors. For instance, mining and transportation products have the highest shock frequency. On average, European countries have the highest resilience and resistance to import shocks. In contrast, countries in Oceania, sub-Saharan Africa, and Southeast Asia have the lowest resistance, and Central America and African countries have the lowest resilience. Interestingly, in agreement with ecological resilience theory, our evaluation of the resilience-resistance tradeoff hypothesis – a relationship often observed in ecological systems – indicates that resilient countries tend to exhibit a lower resilience to shocks. However, we also observe that resistant countries tend to recover faster, suggesting the possibility that human systems may be able to overcome the resilience-resistant tradeoff of ecological systems. Our results indicate that international trade connections can be structured and managed to favor or undermine country-level resilience.