Essays on heterogeneous agent macroeconomics
MetadataShow full item record
This dissertation consists of three essays studying macroeconomic implications of economic agents' decision making subject to risk. In Chapter 1, I explain the decline in both consumption and investment after shocks that depress investment, a task that many macroeconomic models struggle to accomplish. I show that, when markets are incomplete and unemployment risk is countercyclical, shocks that reduce investment raise a precautionary savings motive and thus depress consumption. In particular, the calibrated incomplete markets model generates procyclical consumption and explains most of the consumption volatility relative to output at business cycle frequencies. Chapter 2 explores how the presence of incomplete markets and unemployment risk alters the sources of aggregate fluctuations. I incorporate structural shocks that are widely used in the business cycle literature in the setting presented in Chapter 1. I estimate the model using US macroeconomic data. I find that the presence of incomplete markets and unemployment risk significantly reduces the contribution of discount factor shocks to consumption fluctuations over the business cycle. This result is important because it makes business cycle fluctuations less dependent on shocks that are typically not grounded on solid microfoundations. In Chapter 3, I derive the optimal loan contract between risk-sensitive lenders and credit constrained borrowers in the financial accelerator model of Bernanke, Gertler, and Gilchrist (1999). I assume that lenders have Epstein and Zin recursive preferences and parameterize the risk aversion coefficient to match the equity premium. An important feature of the optimal contract is indexation to the lenders' marginal utility of consumption and the borrowers' marginal value of internal funds. I find that, under the optimal contract, the financial accelerator mechanism becomes very powerful, because lenders require high (low) interest rates on loans during downturns (booms). The result suggests that the extent to which financial frictions matter for aggregate fluctuations depends on the insurance incentive of lenders.