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.

In Article 25 (1) of the Universal Declaration of Human Rights, the United Nations recognized in 1948 the basic human right to “security in the event of unemployment, sickness, disability, widowhood, old age or other lack of livelihood in circumstances beyond ... control.” This paper examines how economic insecurity is related to, yet different from, poverty and inequality, why it matters for human well-being and how it has been changing in different countries around the world in recent years. The paper concludes with discussion of how economic insecurity has been and will be affected by the Covid-19 pandemic/recession.

Economic diversification is very relevant for poorer developing countries to create jobs and foster economic development. That need has been recognized in key internationally agreed development goals. The empirical economic literature has identified several stylized facts about the pattern of diversification of economies, but the development of explanations for those patterns in general has been only loosely associated with economic theory on growth, trade, technology change and structural transformation. Making that connection is relevant because it could inform policymakers in developing countries in designing and implementing policies for promoting diversification. This paper presents…