On the Existence of a Behavioral Component to the Business Cycle

Abstract

This dissertation consists of two essays which address the origins of the business cycle. In particular, it asks: to what extent do behavioral or psychological effects, famously termed "animal spirits" by John Maynard Keynes, contribute to the amplification of business cycle fluctuations. The first essay, titled "The Labor Market Effects of Bad Economic News", examines the effects of economically pessimistic newspaper articles on hiring and employment patterns. Combining information on newspaper subscriptions with automated content analysis of newspaper articles, the paper reconstructs the flow of pessimistic news across the United States during the past recession on a county-by-county, quarter-by-quarter basis. This high resolution map of pessimistic news delivery is then used to estimate the causal impact of media pessimism on labor market outcomes. Exposure to negative news is found to suppress hiring and total employment during the early stages of the recession by up to 40% compared to pre-recession levels; overall, media pessimism can account for some 7% of jobs lost between 2007 and 2010. Further analysis of Google search data suggests that this contractionary effect is mediated by changes in public attitude caused by exposure to pessimistic stories in the media. Importantly, this study considers only articles which report negative news about the state of the national economy, rather than stories which focus on local events. It argues that the prevalence of such news stories affects local labor market conditions, but is unlikely to be affected by such conditions. This approach helps to address the simultaneity issues which have dogged previous research on the topic. The second essay, titled "Uncertainty and Risk Averse Firms in DSGE" a develops theoretical framework to rationalize the previous paper's empirical results. This paper solves a simple general equilibrium model in which firms are risk averse over future profits in a manner analogous to household risk aversion. It shows that response to increased economic uncertainty - particularly uncertainty with regards to future consumer demand, economies with risk averse firms are likely to undergo a business cycle contraction. This result also addresses a long standing problem in the RBC literature; namely, how to generate a contraction with a Keynesian demand side shock. In most models with risk averse utility-maximizing households, a reduction in aggregate demand due to consumer-side changes is expansionary. The paper argues that by introducing firm-side risk aversion into the model, this counter-intuitive behavior can be corrected in a realistic and parsimonious manner.