Essays in heterogeneous agent macroeconomics.

Abstract

This dissertation consists of two essays. The first essay revisits a fundamen- tal puzzle for productivity driven business cycle theories. A challenge for these models of business cycles is that labor productivity and hours are neg- atively correlated in the United States, whereas the theory predicts a strong positive correlation. I refine the empirical puzzle and provide a theory to account for it. I emphasize that hours per worker and individual hours relative to usual hours move counter to productivity. In the cross-section, I find that almost all workers increase their hours when productivity is low, except for low (residual) wage earners. Furthermore, low wage earners suffer larger wage losses when productivity falls. Based on these findings, I hypothesize that non-neutral movements in productivity exacerbate fric- tions due to adverse selection. I use recent advances in competitive search theory to imbed Akerlof's Rat Race into a Neo-Classical growth model with search frictions. Firms bundle high earnings with long hours in order to separate more able and willing workers from the less productive. When low productivity workers fall (relatively) farther behind, the firm requires longer hours from everyone else; perversely, many workers work harder even as the market value of their time falls. A calibrated version of the model is used to measure the aggregate volatility generated by productivity shocks when labor markets are burdened by adverse selection. In the second essay, my coauthor Jacob Short and I study the behavior of entrepreneurs in order to understand the sources and magnitude of risks that they face. Entrepreneurship is risky; entrepreneurs forgo wages and invest their time and resources into a business with large potential gains, but uninsurable risks. It is vital to know the extent of these risks, and the insurance available against them, in order to assess corporate tax and personal bankruptcy reforms. We document that incorporated entrepreneurs operate larger businesses, accumulate more wealth, and are on average more productive than unincorporated entrepreneurs. We embed the U.S. bankruptcy and incorporation legal systems in a quantitative macroeconomic theory of occupation, incorporation, and default choices that accounts for the cross-sectional facts. In the model, as in the U.S., incorporation provides insurance via limited liability beyond personal bankruptcy exemptions, at the expense of administrative burdens and an endogenous interest rate premium. Our model suggests that capital embodied shocks are important entrepreneurial risks. A calibrated economy in which each unit of installed capital entails a small probability of a catastrophic shock (full de-struction of capital) is able to account for the data along multiple untargeted dimensions. We find the welfare gains for entrepreneurs from eliminating investment risk are huge (5.9% increase in annual consumption). And, the welfare loss from removing limited liability appears to be large.