The commodity future investment with the impact of Chinese specific factors

Abstract

This thesis explores the commodity futures investment strategy with the impact of the Chinese specific factors. First, I study the so called Chinese specific factors. To do so, I investigate how commodity future price interacts with domestic macroeconomic variables and overseas futures prices respectively. Specifically, Chapter 2 emphasizes the interaction between domestic commodity futures prices and domestic macroeconomic variables such as interest rates, monetary growth, exchange rates and industrial growth. Among these variables, monetary growth should receive deeper attention because it is widely regarded as the main channel of monetary policy transmission. Subsequently, Chapter 3 focuses on the interaction between domestic commodity futures prices and overseas commodity futures prices. Having gained a clear understanding of the Chinese specific factors, a dynamic timing strategy is accordingly proposed in chapter 4.

Chapter 2 is primarily focused on the interaction between domestic macroeconomic variables and domestic commodity future price movement. Specifically, I try to explore whether low (high) interest rates, loose (tight) money supplies, low (high) foreign exchange rates (Renminbi / US Dollar rate) and high (low) economic growth will lead to high (low) commodity prices and whether commodity prices present overshooting behaviour in response to the interest rate, money supply or changes in the foreign exchange rate. It has been argued by Frankel (1986, 2006) that commodity prices tend to overshoot in response to interest rates as well as to changes in the exchange rates based on Dornbusch’s (1976) model. Evidences from the SVAR models show that part of the theory regarding the relationship between macroeconomic variables and commodity price movement can be supported. The empirical also results suggest that the commodity price shock itself make the largest contribution to commodity price shocks in general. An interest rate shock barely contributes while an M1 growth shock contributes substantially in metals. Foreign exchange rate shocks contribute approximately 40 percent to some commodities, while industrial output shocks comprise approximately 20 to 30 percent to some metals.

In chapter 3, the thesis tries to explore the impacts between China’s futures market and overseas futures markets in chapter 3. Research from this angle could help reveal which side has stronger pricing power. Specifically, I aim to study the information spillover effect between the domestic spot and futures market as well as the information and risk spillover effects between the domestic metal futures market and the overseas metal futures market. Moreover, to check whether China has gained pricing power in the global commodities market, I also study the risk spillover effect between the domestic metal futures market and other overseas financial markets. From the empirical evidences in Chapter 3, it could be seen that asymmetry factors are significant in the futures market, no matter in the Chinese market or oversea market. The empirical results of Granger causality test in Chapter 3 show that movement in the SHFE market could directly guide movement in the LME market, indicating a rise in China’s pricing power in the global commodity market. However, such pricing power is limited and should not be wildly exaggerated.

Chapter 4 forms an effective dynamic timing strategy in China’s commodity market with full consideration of the Chinese specific factors. I adopt Vrugt, Bauer and Molenaar’s (2004) dynamic modeling approach to predict the sign of monthly returns for the three metal futures listed on the Shanghai Futures Exchange: copper, aluminum and zinc. Following Vrugt, Bauer and Molenaar (2004), the base set of explanatory variables is classified into three categories: 1) business cycle indicators; 2) monetary environment indicators; 3) indicators of market sentiment. The empirical results in Chapter 4 show that the timing strategy can offer better returns, a lower standard deviation and, as a consequence, a higher information ratio for all three metal futures.