Uncertainty in a Disaggregate Model: A Data Rich Approach Using Google Search Queries

Abstract

This paper estimates the impact of uncertainty shocks in a disaggregate model featuring state-level unemployment and uncertainty, which is measured using Google search data. We show that the disaggregate model captures important spillover effects which a model using aggregate data would overlook resulting in significantly different peak responses and time dynamic effects. We find the effect of uncertainty shocks on state-level unemployment is recessionary and heterogeneous. The importance of national factors in propagating the effect of uncertainty is also heterogeneous across states, and overall less relevant than state-level factors. These heterogeneous effects are found to be related to state-specific industry compositions and the fiscal position.