Essays on commodity prices and their impact on economic growth and financial markets

Abstract

This thesis empirically investigates the impact of commodity prices on the macroeconomy and financial markets by drawing explicitly upon their forecasting power for economic growth and stock market returns. Typically, primary commodity trade generates a significant proportion of national income in resource-rich countries and, therefore, any short-run movements in primary commodity prices may have important consequences for economic growth and national financial markets. Supported by a thorough review of the existing literature, the analysis is carried out in three empirical chapters. Chapter 2 provides an advancement of the index number theory by developing improved index measures of national commodity export prices for a wide range of countries and territories, 217 in total, over the period of January 1980 to April 2017. It proposes a new approach for data collection, which builds upon the past studies by accommodating more precise and accurate data sets. This study demonstrates empirically that the constructed index series outperform those created in past studies. Chapter 3 looks at the forecasting power of commodity prices for economic growth for a set of 33 commodity-dependent countries between January 1980 and December 2016. Using a mixed-frequency time-varying approach, the empirical results reveal evidence of in-sample causality from commodity prices to economic growth in the case of 31 out of 33 countries. This inference becomes weaker when the estimation horizon becomes longer. Moreover, the commodity-based predictive regressions outperform the benchmark models in 79% of the countries. The substantial evidence found in support of a link between commodity prices and economic growth indicates the long-standing requirement for trade diversification in countries that remain heavily dependent on commodities. Chapter 4 investigates the relationship between global commodities and national financial markets for 63 countries and territories between January 1951 and March 2018. The study considers five measures of global commodities that are defined as global shocks: world oil prices, world oil demand, world oil supply, world commodity prices (all items) and world metal prices. Using a mixed-frequency time-varying approach, this study provides evidence that commodity prices can predict stock market returns. In the best-case scenario, the world economic activity, denoted as world oil demand, has forecasting power on stock market returns for 54 out of 63 countries. Whereas, in the worst-case scenario, the world oil (metal) prices predict stock market returns for 42 out of 63 countries. This study demonstrates that world commodity prices (all items) exert more influence on stock market returns than oil prices