Book Description

This dissertation analyzes business cycles and international asset pricing under disaster risk. In the first chapter, I use annual consumption and financial data for 31 countries over 140 years and I document that developing countries exhibit a more volatile consumption and a significantly larger equity premium. By employing a Bayesian Markov Chain Monte Carlo approach, I estimate an empirical model of macroeconomic disasters - low-probability events with disastrous consequences such as the Great Depression - in developing and high-income countries. I find that developing countries have a higher overall probability of entering a disaster and that they are also much more likely to enter an individual disaster such as a sovereign debt crisis. Disasters in high-income countries are shown to be shorter, on average, but more severe and uncertain. Group heterogeneity in disaster parameters allows me to generate a substantial equity premium for both groups of countries. Disaster contagion plays a vital role in explaining the equity premium puzzle for high-income countries. The model-simulated correlations of equity premium within each group of countries are qualitatively in line with data. The second chapter provides evidence that the U.S. stock market returns not only exhibit large negative skewness, but that they also provide poor payoffs during deep consumption recessions. Using out-of-the-money S&P 500 index options, I obtain a hedged risk premium and show that the hedged risk premium captures the equity risk premium during normal times. I isolate the disaster risk premium as the difference between the total equity risk premium and the hedged risk premium. In addition, I illustrate that the risk premium due to disasters explains about eighty percent of the total equity risk premium. In the cross-section of stock returns, I find that stocks that are more negatively related to the disaster risk premium yield considerably higher subsequent returns. However, this finding is not robust to adjusting for Fama-French price factors. I also find a little predictive power of the disaster risk premium with respect to the aggregate stock market returns due to the lack of autocorrelation in the disaster risk premium. The third chapter recognizes the importance of a large informal economy for business cycles in emerging countries. I show that a two-sector real business cycle model of a small open economy with a poorly measured informal sector, Cobb-Douglas utility function, and country spread fluctuations accounts for the low volatility of hours worked and large relative volatility of consumption to output in emerging countries. Due to the non-separability between consumption and labor supply, the model cannot explain the countercyclical real interest rates and trade balance that prevail in developing countries. The results suggest that GHH preferences are necessary to generate countercyclical real interest rates and trade balance in a neoclassical setting with working capital constraint and exogenous movements in real interest rates.