Essays on the impacts of commodity price fluctuations

• Ano de defesa: 2024
• Autor(a) principal: Melo, Alisson Curátola de
• Orientador(a): Guimarães, Bernardo de Vasconcellos
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Palavras-chave em Português: Preços de commodities; Fator global; Risco país; Identificação por heterocedasticidade
• Palavras-chave em Inglês: Commodity prices; Global factor; Country risk; Identification through heteroskedasticity; USDA; OPEC
• Link de acesso: https://hdl.handle.net/10438/35043

This doctoral dissertation investigates the macroeconomic effects of commodity price fluctuations through three articles organized into chapters. The first chapter examines the link between commodity prices and business cycles in resource-rich countries, focusing on Latin America. It estimates a two-sector dynamic stochastic general equilibrium (DSGE) model that distinguishes commodity price fluctuations into a global factor and an idiosyncratic component. The study finds that the global factor predominantly drives commodity prices, while the impact of these prices on economic cycles and interest rate premiums is substantially lower than suggested by previous studies. The second chapter employs high-frequency data and identification through heteroskedasticity (IH) to assess how grain and oil prices affect financial indicators. Notably, it explores a novel source of price heteroscedasticity, the Grain Stocks report from the United States Department of Agriculture (USDA). The results reveal that heteroscedasticity-based coefficients are significantly below ordinary least squares (OLS) estimates, with grain price shocks yielding positive economic outcomes and oil price hikes having adverse impacts. The final chapter discusses the IH method’s limitation in differentiating direct from indirect effects and proposes combining IH with theoretically based exclusion restrictions. Building on the second chapter’s findings, this third article indicates that the effects of commodity prices are not only markedly lower than OLS estimates but often occur through indirect channels.