Search

This dissertation is composed of three empirical studies. The first study examines the long-run relationship between the real world price of maize, soybeans and sugar with the real world price of crude oil and a series of macroeconomic variables using a cointegration analysis from January 1982 until December 2012. The main empirical results support a strong causal relationship between maize and soybeans with crude oil, the real interest rate and the real U.S. exchange rate. It is shown that real world crude oil prices are cointegrated with real world prices of maize and soybeans for the entire sample period and that real oil prices have a one-to-one relationship with these commodities. In other words, a one-percent increase in the price of real crude oil is associated with a one-percent increase in the price of maize and soybeans. Moreover, we find that permanent shocks to crude oil prices are transmitted to both maize and soybeans by a factor of 0.67 in both cases. In addition, our results show that despite the instability associated with the period between 2007-2008, the cointegrating relationship between crude oil and these agricultural commodities has remained stable during the entire sample period. Finally, our results also suggest that although the real interest rate and the U.S. exchange rate are cointegrated with these commodities, it is only permanent shocks to real crude oil prices that have a permanent effect on the commodities price behavior. In the second study I examine the degree of interdependence between three agricultural commodity prices, crude oil price returns, macroeconomic variables and the S&P GSCI commodity returns index. I apply Aielli [2013] cDCC model using monthly data from 1982 to 2012 to estimate the dynamic correlations of the returns series and endogenously detect any structural instability of the dynamic correlations. The results indicate that crude oil price returns present statistically significant dynamic correlations with all the macroeconomic variables in addition to the GSCI index. Additionally, we detect structural changes in these dynamic correlations mainly associated with the financial crisis of 2008. On the other hand, our results show that there exists no degree of interdependence between maize, soybeans and sugar with crude oil price returns and most of the macroeconomic variables. The exceptions are between soybeans with the U.S. exchange rate and sugar with global economic activity. Nevertheless, only the GSCI index presents significant dynamic correlations with these commodity price returns. In the third study I apply an asset pricing theory model in order to evaluate the extent to which investment in futures commodity price indexes influence spot price return in a portfolio of commodity and energy prices. Particularly, we are interested in measuring the common risk factors of six agricultural commodities (e.g. maize, soybean, sugar, wheat, barley and sorghum) and global crude oil. Here, I aim to estimate their relationship with equities, the U.S. dollar, interest rates, a series of variables measuring the global macroeconomic performance and commodity futures price indexes. In contrast to the literature, instead of using a principal components analysis (PCA), which is concerned with explaining the return variances of a portfolio, I use a reduced rank approach in order to capture the canonical correlations in an effort to measure those factors which explain the risk to the commodity portfolio. This approach help us to understand the cross-section dependency of commodity markets with the global macroeconomic cycles as well as to capture the extent to which increasing portfolio investment by institutional investors, which has given rise to the so called “financialization of commodities”, are motivated by diversification strategies or by speculative behavior in these markets. Our findings, indicate that even though macroeconomic factors, market specific and commodity futures indexes are captured among the risk factors in this commodity portfolio. Nevertheless, the factors associated with the market specific and the commodity futures indexes offer a hedge for the risks provided by the common macroeconomic factor. Therefore, using this approach, I conclude that commodity futures index investment appears to offer the diversification effect which has been the main driver of the so called commodity financialization within the past fifteen years.