| One active area of recent research in macroeconomics is the business cycle dynamics of the economy when agents are heterogenous and markets are incomplete. The literature has shown that standard results of the representative agents/complete market paradigm could profoundly change when we adopt a heterogenous-agents framework and agents cannot fully insure against risks. This volume is an attempt to contribute to this research program from both quantitative and computational perspectives. In particular, I provide a quantitative analysis of some welfare issues, explore the role of monetary policy in the long run, and examine the robustness of a main idea proposed in the literature. Computationally, the currently-used numerical algorithm has been modified to accommodate these extensive applications resulting in a faster routine that can be applied in much richer setups.The first chapter analyses the welfare consequences of eliminating the economic crisis in an economy with substantial agent heterogeneity. The model is calibrated to match the U.S. wealth distribution, employment features, and the size of the investment drop during the U.S. Great Depression. I find a small gain from eliminating the depression-style economic crisis. There are, however, large differences across different groups: very poor and unemployed consumers gain a lot, while richer and employed individuals gain less and sometimes lose from eliminating the economic crisis.The second chapter examines the validity of Sidrauski's (1967) money superneutrality result in this class of models. I show that the superneutrality result no longer holds in this case. Tobin's (1965) effect and heterogeneity in inflation tax account for the non-superneutrality of money. However, the effects of higher money growth rate on real variables are rather small. The change in money growth rate mostly gets reflected in an increase in the aggregate price level.The third chapter explores the robustness of the "approximate aggregation" insight proposed by Krusell and Smith (1998) in a much richer setting. The conclusion shows that approximate aggregation no longer holds. In an economy featuring a crisis state and two assets capital and money, we can no longer adequately describe macroeconomic aggregates using only the first moment of the wealth distribution. |
