UC Riverside

This dissertation covers several topics in macroeconomics. Chapter one provides an overview for this dissertation. Chapter two explores the role of demand shocks, as an alternative to productivity shocks, in driving both domestic and international business cycles. In addition to those well-documented domestic and international business cycle properties, this paper focuses on two additional stylized facts in the industrialized countries: procyclical trade openness (GDP fraction of trade volume) and countercyclical government size (GDP fraction of government spending). Using a parsimonious dynamic stochastic general equilibrium model, I show that the model's predictions under productivity shocks are not consistent with these facts. Instead, a demand-shock-driven model replicates the above facts while matching other business cycle properties.

Chapter three examines the long-run relationship between trade openness and government size in a two-country dynamic general equilibrium model. I analytically show that, if the non-tradable sector is more capital intensive, higher government expenditures drive up the relative capital stock in the tradable sector in steady state. This gives rise to a relatively higher output and a relatively lower price in the tradable sector. As a result, when trade openness increases, a benevolent government would expand public expenditures to push up (down) the relative output (price) of tradables so as to achieve agents' \textit{desired} consumption plan with more consumption of tradables. Therefore, a positive relationship between trade openness and government size is observed. On the contrary, if the tradable sector is more capital intensive, a negative correlation follows.

In chapter four, I present new estimates of the factor substitution elasticity and biased factor-augmenting technical progress using the supply-side system for the aggregate U.S. economy during the period 1948-2012. On the basis of recursive scheme estimations, I first show that significant variations of estimated model parameters arise from different sample periods. I next incorporate labor market friction into the supply-side system and show that the augmented model fits the data better. With labor market friction, the estimated elasticity of substitution between capital and labor does not statistically significantly differ from unity. The long-run technical progress tends to be purely labor-augmenting although non-negligible variations arise during some sample periods.