Essays in the Application of Linear and Non-Linear Bayesian VAR Models to the Macroeconomic Impacts of Energy Price Shocks

This thesis is a collection of five self contained empirical macroeconomic papers on the asymmetric effects of energy price shocks on various economies. Chapter 1 formally determines the number of regime changes in the US natural gas market by employing a MS-VAR model. Estimated using Bayesian methods, three regimes are identified for the period 1980 - 2016, namely, before the Decontrol Act, after the Decontrol Act and the Recession. The results show that the natural gas market tends to be much more sensitive to market fundamental shocks occurring in a Recession regime than in the other regimes. Augmenting the model by incorporating the price of crude oil, the results reveal that the impacts of oil price shocks on natural gas prices are relatively small. Chapter 2 provides new empirical evidence on the asymmetric reactions of the U.S. natural gas market and the U.S. economy to its market fundamental shocks in different phases of the business cycle. To this end, we employ a ST-VAR model to capture the asymmetric responses depending on economic conditions. Our results indicate that in contrast to the prediction made by a linear VAR model, the STVAR model provides a plausible explanation to the behavior of the U.S. natural gas market, which asymmetrically reacts in bad times and good times. Chapter 3 examines the relationship between China's economic growth and global oil market fluctuations between 1992Q1 and 2015Q3. We find that: (1) the time varying parameter VAR with stochastic volatility provides a better fit as compared to it's constant counterparts; (2) the impacts of intertemporal global oil price shocks on China's output are often small and temporary in nature; (3) oil supply and specific oil demand shocks generally produce negative movements in China's GDP growth whilst oil demand shocks tend to have positive effects; (4) domestic output shocks have no significant impact on price or quantity movements within the global oil market. Chapter 4 examines the effects of world energy price shocks on China's macroeconomy. We propose a new index of primary commodity energy prices which accurately reflects both the structure of China's energy expenditure shares, as well as intertemporal fluctuations in international energy prices. The index is then in employed a sufficiently rich set of time varying BVARs, identified by a new set of agnostic sign restrictions. Uniformly sized positive energy price shocks are shown to consistently generate economic stagflation over the past two decades. Chapter 5 compares the macroeconomic effects of global oil and iron ore price shocks on the Australian economy. The main results suggest that, over the period 1990Q1 to 2014Q4, the oil shock has a relative larger impact than that of the iron ore shock on output and inflation while the iron ore shock is the dominant source of interest and exchange rate movements. The effects crucially depend on the underlying sources of oil or iron ore price shifts