UC Berkeley

This dissertation is a compilation of three essays on the role of information frictions in macroeconomics.

The first essay contributes to the literature on the impact of uncertainty on the business cycle. The cross-sectional dispersion of firm-level outcomes, such as sales growth or stock returns, is markedly countercyclical. Recent papers have framed this fact as evidence that exogenous “uncertainty shocks” are important drivers of business cycles. This paper provides empirical evidence that the co-movement of various dispersion measures with the business cycle is better understood as the economy’s endogenous response to traditional first moment shocks – dispersion is the effect, not the cause. It then develops a theoretical model that links the cross-sectional dispersion of micro-level outcomes to the aggregate state of the economy. The mechanism is based on time-varying rational inattention. In bad times, firms pay more attention to idiosyncratic shocks hitting their business environment. More precise micro- level information about the underlying heterogeneity leads to higher dispersion in realized outcomes. In line with the empirical findings, the model generates countercyclical dispersion without relying on exogenous second moment (uncertainty) shocks.

The second essay uses survey expectations to assess the microfoundations of an important class of macroeconomic models. Many theoretical macro models try to explain the pervasive nominal and real stickiness in the data by assuming rational decision-making under imperfect information. The behavior of consensus (average) forecasts is consistent with the predictions of these models, which can be seen as supportive empirical evidence for the models’ microfoundations (Coibion and Gorodnichenko, 2012). This paper demonstrates, however, that the individual-level data underlying the consensus forecasts are at odds with this interpretation. In particular, I document that individual expectations in the Survey of Professional Forecasters do not pass a very weak test of rational expectations: current forecast revisions are strong predictors of subsequent forecast errors. Information frictions alone cannot explain this pattern.

I go on to propose a simple modification of the noisy information framework that allows for a particular form of non-rational expectations: agents may incorrectly weight new information against their prior. I show that this parsimonious model can match the survey data along several dimensions. Using the structure of the model, I estimate the direction and size of inefficiencies in the expectations formation process. I find that in most cases agents put too much weight on their private information, which can be interpreted as overconfidence in the precision of private information. I also show that there is substantial heterogeneity across agents in the deviation from rational expectations, and I relate these differences to observable characteristics. Finally, I discuss potential interpretations of my empirical results and their implications for macroeconomic theory.

The third essay explores the potential trade-off between competition and systemic stability in financial intermediation. Why do banks feel compelled to operate with such high leverage despite the risks this poses? Using a simple model, I argue that the degree of competition goes a long way in explaining capital structure decisions. On the one hand, information frictions (adverse selection) render debt a cheaper form of financing than equity. On the other hand, more reliance on debt increases the probability of bankruptcy, which results in the loss of the bank’s charter value. The degree of competition affects charter values, and hence changes the way banks balance between these two forces. A panel analysis of European banks’ capital structure around the introduction of the euro reveals statistically and economically significant effects consistent with this hypothesis. Banks, in particular smaller banks, decreased their equity ratios after entering the currency area. Complementary evidence suggests that this effect can be attributed to increased competitive pressures boosted by the euro.