Book Description

This dissertation examines the interaction between macroeconomic aggregates and financial markets in two different essays. The expansion of derivatives markets has prompted interest in estimating options-implied measures to analyze market participants’ beliefs about future movements in the prices of these derivatives’ underlying assets and the probability these participants assign to unlikely events (see Datta et al., 2014). In this spirit, analyzing oil market is important for two main reasons. First, among all commodities, crude oil futures and derivatives are the most traded and liquid asset in the whole commodity market. Second, the informational content of oil derivatives can be indicative of shifts in global economic expectations which may be of interests to producers, investors and policy makers. Because the risk neutral density (RND, hereafter) consists of information from various option series that have a wide range of strike prices and maturities, we can conjecture more detailed effects of news announcements on market sentiment by investigating the changes in the RND. Chapter 1 links the crude oil market to macroeconomic risk by studying the RND around the U.S. macroeconomic news announcements. I use a non-parametric method to recover the RND and conduct regression analysis using daily data. The analysis provides several noteworthy results. First, I find that the RND is systematically affected by certain macroeconomic news announcements. Second, after controlling for the content of the news, my results indicate that good news tend to make the distribution less negatively skewed, whereas bad news have an opposite effect. However, I do not find any systematic pattern between the content (bad/good) of the news and the implied volatility or kurtosis. Hence, my results show that better/worse-than-expected news in macroeconomic announcements may both increase and decrease implied volatility and kurtosis of the option implied distribution. Finally my estimates obtained from nonlinear regressions display that the magnitude of the surprise may play into this effect; for example worse-than-expected news in Housing Starts announcement decrease the implied volatility and increase the implied kurtosis only when the size of surprise is not too large. How should a central bank conduct monetary policy in the presence of financial shocks? In Chapter 2, I use different nonlinear policy rules and address this question. Most empirical work on monetary policy relies on simple linear policy rules, however it is not clear whether such a rule can be an adequate representation of a process as complex as that of monetary policy. I first estimate Markov Switching Taylor rules with constant transition probabilities to allow for state-contingent policy making during 1987.3-2008.4. As a proxy for financial stress, I use the Adjusted National Financial Conditions Index constructed by the Chicago Fed. Then, I allow transition probabilities driving the monetary policy stance to vary over time and be a function of economic and financial indicators. The paper provides clear-cut evidence that, during the Greenspan-Bernanke tenure, the U.S. monetary policy can be characterized falling into two distinct regimes; a conventional regime where the Fed puts a greater emphasis on targeting inflation while stabilizing the economic outlook and a distressed regime where the Fed responds aggressively to output gaps and is less concerned with inflation. The distressed regime is closely correlated with times of financial imbalances. The empirical results show that nonlinear models outperform the simple linear specification in terms of model fit and the ability to track the actual interest rate. Also, the economic and financial indicators are found to be informative in dating the evolution of the state of the monetary policy stance. The results have implications for nonlinear rules to be a useful guideline for forecasting and policy analysis.