The predictive power of commodity prices for future economic growth: evaluating the role of economic development

Abstract

This paper investigates the extent to which commodity prices can predict economic growth, the growth exposure of the countries to specific commodity group, and whether the country stage of development matters for growth dependency on commodities. We use a novel time-varying mixed-frequency vector autoregressive model to jointly estimate causality between monthly commodity prices and quarterly economic growth for the period from January 1980 to March 2020. Our findings suggest that growth dependency on commodities varies over time and across different synthetic measures, but in general solid evidence of predictability is determined. The overall commodity index shows better performance in predicting economic growth than the segregated proxies of either fuel or non-fuel commodities in developed economies. However, the overall commodity index is as effective as the fuel index in predicting economic growth for their developing counterparts. We develop a new index of global commodity growth connectivity (GCGCI) to determine the economic growth dependency on commodities over time. Our results show that economic growth dependency on commodities has increased at least three times with financialization of commodity markets. The GCGCI for all commodities reveals a higher growth dependency on commodities in developing countries than developed ones in the post-financialization period. Interestingly, we find that growth decreases its reliance on fuel commodities in both developed and developing countries after financialization of commodity markets. Our findings bear implications for policymakers to design stabilization policies in an attempt to protect economies vulnerable to commodity price shocks and lay the foundation for sustainable economic growth.

More information

Identifiers

ORCID for Martin Enilov: orcid.org/0000-0002-7671-6975

Catalogue record

Export record

Altmetrics