ESSAYS ON MACROECONOMIC POLICY WITH HETEROGENEOUS AGENTS

Abstract

What are the consequences of microeconomic heterogeneity for the transmission and design of macroeconomic policies? This dissertation collects together 3 essays that contribute to our understanding of this important question.

In chapter one, I study the transmission of aggregate shocks in a New Keynesian model in which households’ incomes are heterogeneously exposed to changes in aggregate income, and borrowing frictions limit opportunities for aggregate risk sharing. I analytically show that shock transmission is asymmetric: output responds more to contractionary shocks than to expansionary shocks of equal magnitude. Estimating key model parameters using the micro evidence on heterogeneous consumption exposures to changes in output generates asymmetric responses of output to monetary policy shocks that can explain at least 60% of the empirical asymmetry.

In chapter two, co-authored with David Arnold, we study the macroeconomic effects of Employment Protection Legislation (EPL) in Brazil. The fact that Brazilian EPL only affects jobs with tenures greater than three months causes a spike in the job termination hazard at three months, the size of which identifies the behavioral response of firms to the imposition of EPL. We estimate a structural model that maps this response into macroeconomic outcomes of interest. The imposition of EPL causes a 1.3% drop in GDP, which is driven by a 0.9 percentage point increase in the unemployment rate. Intuitively, the reduction in vacancy creation by firms caused by EPL widens the gap between the efficient level of vacancy creation and the observed equilibrium level.

Finally, in chapter three, I study optimal taxation in an economy in which households make consumption, labor supply, and fertility choices. I derive sufficient statistics for the sign and shape of optimal wedges on child quantity, goods investment and time investment, and provide intuition for the main economic forces at play. Distorting fertility choices relaxes incentive constraints, which facilitates redistribution, but also may discourage households to earn income, thus hampering the redistributive strength of the income tax. A quantitative exercise demonstrates that the bulk of welfare gains available from subsidizing investments in children can be obtained using feasible linear subsidies on child investment goods.