Pablo Egaña Del Sol

The future of work in developing economies. What can we learn from the global south?

With Alejandro Micco (UChile)


In the last 10 years, concern has been growing about the effect that automation will have on the future of work. Numerous studies have investigated the impact that automation will have on labor markets but almost all have focused on developed nations, which generally have more service-oriented economies and a more skilled labor force. This is one of the first paper to study automation risk rates for developing nations. We examine the effect of automation on a group of developing countries in Latin America, Africa, Southeast Europe, and Asia. We also consider a group of 18 developed nations. To address the heterogeneity of occupations across countries, we apply a task-based approach and recalibrate the effect of automation on the labor market in order to analyze the task structure between and within countries. Modeling off previous studies, we follow an expectation-maximization algorithm to predict the risk of automation at the worker level. Highly automatable jobs are those with a 70\% or greater risk of automation. The results suggest that developing countries have higher levels of predicted automation risk compared to developed economies, regardless the considerable heterogeneity of the results. We also find that occupations containing relatively more routine tasks are more likely to be automated, and that workers with a higher level of education have a reduced risk of automation. Finally, we explore the relationship between automation risk with wages and employment, we regress differentiating the effect between developed and developing countries and by gender. Our results indicate that the risk of automation is negatively correlated with both wages and employment growth, and this negative correlation is larger in magnitude for developed countries and for women.