Resumen de Essays on information frictions in macroeconomics
As modern economies can be represented by a network of interdependent agents, many settings feature a coordination motive: an agent’s decisions depend not only on the state of the economy or their beliefs thereof, but also on their expectations of how others will respond. Workhorse macroeconomic models typically feature representative agents and can not incorporate the possibility that such agents disagree on the current state and future trajectory of the economy.
In the first chapter, I study the effect of monetary policy in the presence of particular information frictions which impede the ability of agents to coordinate. The first friction is strategic uncertainty, which refers to the uncertainty that agents face about the behavior of others and the resulting macroeconomic outcomes. However, agents do not make decisions in vacuums; they condition their responses on information, which is endogenous in most situations of interest. Endogenous signals capture the role of market research, prices, and macroeconomic indicators in coordinating actions and beliefs. Taken together, these frictions allow for sentiments, or beliefs about aggregate demand, to be self-fulfilling. The perfect information benchmark in the New Keynesian framework turns out to be non-trivial, and relaxing it in this manner allows for an alternate channel in which monetary policy can affect outcomes. Through its effect on aggregate variables, the stance of monetary policy determines the precision of endogenous signals that firms receive, and consequently, the degree of coordination in firms’ production (pricing) decision. As a result, the distribution of non-fundamental shocks is no longer independent of policy, introducing a novel tradeoff between stabilizing output and inflation. Strong inflation targeting increases the variance of non-fundamental fluctuations.
The second chapter analyzes welfare properties of the framework introduced in chapter one. I consider an appropriate efficiency benchmark, one which represents the best allocation among those that respect resource feasibility and the decentralization of information. The endogenous information structure in the decentralized equilibrium is shown to result in non-fundamental fluctuations, which characterizes an inefficiency in both the use and aggregation of information. The Taylor rule is no longer sufficient to rule out such indeterminacy. Instead, a simple interest rate rule that mitigates the degree to which it targets inflation implements the efficient allocation, by eliminating non-fundamental fluctuations and precluding the output-inflation trade-off.
Motivated by theories of aggregate fluctuations arising from shocks to agents’ expectations in incomplete information settings, the third chapter considers the relationship between sentiment and uncertainty shocks. Both shocks are related to information and the formation of beliefs, and each are typically identified as short-run sources of co-movement in macroeconomic aggregates. Sentiments, defined as a change to expectations about economic activity and orthogonal to news about future technology, can be interpreted as rational optimism or pessimism. As such, they may affect confidence, or uncertainty. A maximum forecast error variance approach is used to identify sentiment and uncertainty shocks in a structural vector autoregressive model. Sentiment shocks are shown to account for more variation than news. However, they explain less of the variation in GDP and hours than previous studies have shown, as uncertainty is also an important source of short-run fluctuations. This chapter also provides evidence that sentiments and uncertainty shocks as identified in the literature are correlated and may not be truly structural.
