Automation, uncertainty, and global value chains
Faber et al. examine how uncertainty in developing countries and automation technologies have affected reshoring of production from developing to developed countries. They use a shift-share approach to measure country-industry pairs’ exposure to uncertainty shocks in developing countries between 2000-2014. The authors exploit variation in pre-existing trade relationships in 2000 to generate plausibly exogenous variation in exposure to uncertainty and examine how this uncertainty interacts with an industry-level measure of “robot replaceability.” Finally, they use data on input sourcing from the World Input-Output Database for 18 developed countries, 17 developing countries, and 19 industries.
Main Findings:
- Higher uncertainty in developing countries leads to significant reshoring, but only in highly robotized industries.
- A one standard deviation increase in exposure to uncertainty reduces demand for inputs from developing countries by about 10% in robotized industries.
- The effect is stronger for inputs from Eastern Europe, Latin America and China, and increases substantially after the 2008 financial crisis.
- Uncertainty shocks increase reliance on inputs from other developed countries in non-robotized industries, but reduce it in robotized ones.
- Results are robust to various threats to identification and alternative specifications.
The authors conclude that rising uncertainty, combined with advances in automation technologies, has likely contributed to the slowdown in globalization observed since the global financial crisis. The ability to automate production appears to be a key factor enabling firms to reshore in response to higher uncertainty in developing countries.
Global Value Chains in a World of Uncertainty and Automation
Authors: Marius Faber, Kemal Kilic, Gleb Kozliakov, Dalia Marin
From: University of Basel, Technical University of Munich, University of California, Davis
Automation, reshoring, and developing countries
In this paper, Nii-Aponsah et al. analyze the potential effects of automation technologies on the global reorganization of production activities between advanced and developing economies, with a focus on reshoring. The authors extend Duchin’s World Trade Model (WTM) to include non-tradable sectors, creating the World Trade Model with Non-Tradables (WTMNT). They use input-output data from the World Input-Output Database (WIOD) for 2014, covering 56 sectors and 44 countries/regions and construct automation scenarios by estimating sector-level automation risks using data from the International Assessment of Adult Competencies (PIAAC). Finally, the model is solved under three scenarios: baseline (actual data), automation in developed countries only, and global automation.
Main Findings:
- Automation in developed countries leads to significant reshoring of production activities from developing to developed economies:
– Lower-income Asia is most adversely impacted, especially in manufacturing.
– Advanced regions gain a significant share of global GDP. - When automation is global:
– The “Rest of the World” category emerges as a main winner in terms of GDP share.
– Lower-income Asia still loses the most in GDP share, particularly in manufacturing. - In the developed world automation scenario:
– The European Union gains the most in GDP share (6.5 percentage points).
– Lower-income Asia loses the most (7 percentage points). - In the global automation scenario:
– The Rest of the World group is the big winner (+6 percentage points of global GDP share).
– Lower-income Asia is the largest loser (-4.4 percentage points). - Sectoral impacts vary:
– Manufacturing sees significant shifts from lower-income Asia to developed regions.
– Services sectors show mixed results depending on the automation scenario. - The study finds that advanced economies adopting automation technologies would likely benefit via growth in incomes, including positive factor earnings.
Overall, the paper concludes that automation-induced reshoring could significantly impact the global distribution of production and income, with developing economies, particularly in lower-income Asia, facing potential challenges.
Automation-Induced Reshoring and Potential Implications for Developing Economies
Authors: Hubert Nii-Aponsah, Bart Verspagen, Pierre Mohnen
From: Maastricht University
