Robots, Tasks and Trade

Working Paper: CEPR ID: DP14487

Authors: Paulo Bastos; Erhan Artuc; Bob Rijkers

Abstract: This paper examines the effects of robotization on trade patterns, wages and welfare. It develops a Ricardian model with two-stage production and trade in intermediate and final goods in which robots can take over some tasks previously performed by humans in a subset of industries. An increase in robot adoption in the North reduces the cost of production and thereby impacts trade in final and intermediate goods with the South. The empirical analysis uses ordinary least squares and instrumental-variable regressions exploiting variation in exposure to robots across countries and sectors. Both reveal that greater robot intensity in own production leads to: (i) a rise in imports sourced from less developed countries in the same industry; and (ii) an even stronger increase in exports to those countries. Counterfactual simulations indicate that Northern robotization raises domestic welfare, but initially depresses wages. However, this adverse effect is likely to be reversed by further reductions in robot prices. Northern robotization may lead to higher wages and welfare in the South.

Keywords: automation; robots; tasks; jobs; wages; trade; intermediate inputs; global value chains; gains from trade

JEL Codes: F1; J23; J24; O3; O4


Causal Claims Network Graph

Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.


Causal Claims

CauseEffect
Increase in robot density in northern countries (O52)Increase in imports from less developed countries (F63)
Increase in robot density in northern countries (O52)Increase in exports to less developed countries (O19)
Increase in robot density in northern countries (O52)Changes in production costs and demand for intermediate goods (D24)
Increase in robot density in northern countries (O52)Enhanced competitiveness of northern producers (L11)
Increase in robot density in northern countries (O52)Increase in demand for southern inputs (R22)
Northern robotization (L63)Initial depression of wages (J39)
Northern robotization (L63)Long-term increase in domestic welfare (D69)
Decrease in robot prices (D49)Higher wages and welfare in both the north and the south (J39)

Back to index