Informational efficiency and hedging effectiveness in Malaysian crude palm oil markets / Go You How

Crude palm oil (CPO) is one of the important commodities of Malaysian economy. According to the Malaysian Palm Oil Board, Malaysia is the second largest producer of CPO with 39 per cent of world production and 44 per cent of world exports in 2014. Nonetheless, as a commodity, it suffers from price fluctuation due to factors such as climate change and flood that affect the supply and demand of palm oil. Hence, there lies the issue of volatility in CPO prices which has a negative impact on food security. There are three objectives in this research. The first objective is to examine price-volume relation in CPO futures market during the pre-crisis, crisis and post-crisis periods, respectively. Based on daily data from 2000 to 2012, cross-correlation function shows volatility spillovers between return and trading volume in precrisis and post-crisis produce different degree of correlations in different time spans, supporting the “heterogeneity of traders” hypothesis. The inconsistent time span observed in cross-correlation function also supports the noise traders’ hypothesis. It is observed that in the post-crisis period, market participants have become more risk averse. As a result, there has been an increase in volatility persistence which has reduced the level of informational efficiency. The second objective is to examine the hypothesis of Tilton et al. (2011) that asserts investor demand affects commodity prices when spot and futures prices are closely correlated during strong contango. After taking into account market efficiency as measured by variance ratio, our result based on the period of 2000-2014 indicates that higher degrees of efficiency are linked to the high correlations between spot and futures returns during weak contango period and vice versa. During backwardation, bi-directional causality-in-mean happens between spot and futures returns. Notably, it takes a longer period for the former to Granger cause the latter, indicating a change in the spot return is a long-lived phenomenon. In contrast, it takes a shorter period for futures return to Granger cause spot return, indicating it is a shortlived phenomenon. During weak contango, futures volatility Granger causes spot volatility. During strong contango period, there is no causality-in-mean and variance. Hence, we extend the hypothesis that the preference of holding a long position in the futures market is due to the anticipation of insufficient supply of inventories in the short run for CPO as it is susceptible to seasonality and climate change. The third objective is to evaluate the effectiveness of eight hedging models with different mean and variance-covariance specifications for the period of 1986-2013. From the perspective of economic modeling, incorporating the basis term in modeling the joint dynamics of spot and futures returns during the crises provide better results. High dynamic hedge ratios during the Asian financial crisis contribute to the support for CCC-GARCH model. During the global financial crisis, BEKK-GARCH model appears to provide more risk reduction as compared to others. Overall, these findings add to the stylized fact on the dynamic relationship between CPO spot and futures markets under different market conditions