Working Papers

This paper proposes the use of synthetic training data generated by large language models
to improve machine learning SDG classifiers. It shows that supplementing existing training data with
synthetic data produced by the ChatGPT tool improves the performance of the SDGClassy classifier.
This addition of synthetic data is especially useful in building SDG classifiers given the limited availability
of properly labeled data and the complex, interconnected nature of the SDGs. Synthetic data thus enables
more effective machine-learning applications in this context.

This paper assesses the effects of combining fiscal austerity with flexibilization policies aimed at reducing labour costs and increasing competitiveness. Core to our analysis is a global perspective where the aggregation problem is fully taken into account. We derive a stylized macroeconomic framework of distributive and demand dynamics. We show that even in export-led regimes, after considering global feedbacks, flexibilization policies do not lead to higher income and employment. Rather, the end result is contractionary. Over time, the world economy is essentially wage-led and responds positively to coordinated Keynesian stimuli.