| Summary: | <p>In many markets, acquiring and processing the information needed to make optimal decisions costs agents time, effort, and/or money. This thesis is concerned with the macroeconomic implications of the way households behave in the face of these costs, which is known as rational inattention. The thesis consists of three self-contained chapters.</p>
<p>The first and second chapters both explore limited household attention to the choice between different savings products. Regulators (e.g. FCA, 2015) have found that households who gather more information when choosing savings products obtain higher interest rates, and the chapters explore two implications of this.</p>
<p>The first chapter studies how the attention decision varies over the business cycle, and the impact that then has on the cycle. I find in both theory and data that attention to savings product choice is countercyclical. This amplifies business cycle fluctuations, because it introduces a countercyclical wedge between policy rates and the interest rate faced by households. In an estimated DSGE model this amplification is substantial, increasing the variance of consumption by 17%. Policies that reduce the cost of information weaken this amplification and so reduce business cycle volatility.</p>
<p>The second chapter explores the implications of rational inattention to savings choices for inequality and fiscal policy. In a model related to that in chapter 1, I find that poorer households pay less attention to saving choices because they have less to gain from higher interest rates. They therefore face lower interest rates, so have less asset income and less desire to save. I show that this feedback between wealth and attention renders the representative agent steady state of the model unstable with respect to idiosyncratic household income shocks. The stable steady state has a two-agent structure, with poor households at the borrowing constraint paying no attention, while wealthy households pay attention and earn high returns. As the steady state inequality is driven by endogenous attention choices, it responds to policy. I find that aggregate consumption has a smaller and more persistent response to large fiscal expansions in this model than a comparable model where inequality is driven by exogenous discount factor heterogeneity.</p>
<p>The third chapter, co-authored with Martin Ellison, moves away from attention to savings choices, considering the information households choose to acquire about the future of the labour market. We find that introducing rational inattention into a model with uninsurable unemployment risk can generate multiple steady states, where with full information the model has a unique steady state. Large mutations in the distribution of household beliefs can shift the economy between steady states. Labour market beliefs in the model are heterogeneous and persistent, consistent with survey evidence.</p>
|
|---|