Sovereign Risk and Economic Complexity: Machine Learning Insights on Causality and Prediction

• Autores: Jose E. Gomez Gonzalez, Jorge Mario Uribe Gil, Óscar M. Valencia
• Localización: Documents de Treball ( IREA ), Nº. 15, 2023
• Idioma: inglés
• Enlaces
  • Texto completo (pdf)
• Resumen

  • We investigate how a country’s economic complexity influences its sovereign yield spread with respect to the US. We analyze various maturities across 28 countries, consisting of 16 emerging and 12 advanced economies. Notably, a one-unit increase in the economic complexity index is associated to a reduction of about 87 basis points in the 10-year yield spread (p<0.01). However, this effect is largely non-significant for maturities under 3 years and, when significant (p<0.1), the reduction is around 54 bps. This suggests that economic complexity affects not only the level of the sovereign yield spreads but also the curve slope. Our first set of models utilizes advanced causal machine learning tools, allowing us to control for a large set of potential confounders. This is crucial given our relatively small dataset of countries and roughly 15 years of data, as well as the low frequency of annual variables. In the second part of our analysis, we shift our focus to economic complexity’s predictive power. Our findings reveal that economic complexity is a robust predictor of sovereign spreads at 5-year and 10-year maturities, ranking among the top three predictors, alongside inflation and institutional factors like the rule of law. We also discuss the potential mechanisms through which economic complexity reduces sovereign risk and emphasize its role as a long-run determinant of productivity, output and income stability, and the likelihood of fiscal crises.