Three essays on oil shocks on macroeconomic activities and financial market

Oil prices began to fluctuate from 1973 when the control of oil price passed from the U.S. to OPEC. During the previous 20 years we have experienced large oil price fluctuations and these oil fluctuations are often viewed as an important driver to the macroeconomic activity and financial market. According to Hamilton (2011), 10 out of 11 post world II economic recessions have followed by an oil price increase. So what are the effects of oil price fluctuation to the economy and are these effects symmetric? These are the research questions this dissertation answers.;The general consensus held in the literature is that oil price increases have more effects on economic activities than oil price decreases do. And several studies offer important insights about the origination of such asymmetry. Hamilton (1988) illustrate that adjustment cost to changing oil prices could be the source of the asymmetry. Bernanke (1997) point out the monetary policy could be another possibility of asymmetry. Balke (2002) use nonlinear dynamic relations, confirm that monetary policy and adjustment cost could account for the asymmetry. Although asymmetry is well accepted, a few studies recently have raised some concerns with regards to the robustness of the conclusion, Kilian and Vigfusson (2012) demonstrate that the commonly used nonlinear combination of oil price variable is not helpful for the out of sample forecasting.;Chapter 1 empirically estimates the effect of oil price shocks on the real stock returns for both U.S. and China. The Sadorsky's VAR model is extended by incorporating world oil production and world economic activity index to check the main reason for oil price movements. From empirical analysis, I find that global demand shock has a statistically significant positive effect on real stock returns for both U.S. and China, supply shock has a negative effect on real stock returns for China, but not significant for the U.S. and specific demand shock has a negative impact on real stock returns for the U.S. but not significant for China. In addition, the results from variance decomposition analysis show that aggregate oil shocks account for a statistically significant 19% of the volatility in real stock returns for the U.S. and 24% for China.;Chapter 2 re-evaluates the asymmetric relationship between oil price shocks and a wide range of U.S. macroeconomic variables based on a nonlinear FAVAR model. I modify the FAVAR model in this way: first, observable oil price and Hamilton net oil price variables enter observation equations simultaneously. Second, to avoid the inconsistent and misleading OLS estimator by Kilian and Vigfusson (2011), the Hamilton oil price does not enter the VAR system as an endogenous variable but as an independent exogenous variable. In addition, two common factors extracted from a large number of macroeconomic dataset are included in a VAR framework. Since the model is nonlinear, I use simulation methods to calculate the impulse response functions. To preview the results, this essay has the following findings. First, the asymmetric effects of oil price shock on macroeconomic activity are significant evident, against the conclusion obtained from Kilian and Vigfusson (2011). Rising oil price has a negative effect on output, gross saving, employee's payrolls, housing price, consumer expectation and etc, while has a positive effect on Fed Funds rate, etc. Second, the asymmetry depends on the size of the shock: the larger the shocks are, the more evident the asymmetry is. Third, the degree of asymmetry is larger in the short run than in the long run. Fourth, the impulse responses of "real activity factor" and "monetary policy factor" to both positive and negative oil price shocks are asymmetric.;Chapter 3 investigates the effects of different oil shocks on U.S. asset returns. Unobserved common factors are extracted from financial dataset and included into a VAR which contains observed variables: a measure of economic activity, oil production, inventories, real oil price and future/spot oil price spread. By examining the correlation coefficients between common factors and financial dataset and interpreting the median of posterior distribution, I can have an indication of what these common factors represent. Furthermore, I identify oil supply, aggregate demand, specific demand, and speculation shocks by imposing restrictions on the signs of impulse responses for a small subset of variables. I find the responses of U.S. asset returns to oil price shocks differ greatly depending on the underlying causes of the oil price increase. In particular, speculation shocks in the oil market lower asset returns. In my industry-specific analysis, I find that supply shocks and inventory demand shocks are relatively important in the financial market in the short run while global demand shocks have had more persistent effects on asset returns. (Abstract shortened by UMI.).