The Effects of Government Spending: A Disaggregated Approach

Abstract

We disaggregate government spending into five macroeconomic-relevant components: average wage, employment, purchases of intermediate goods and services, investment and transfers. We set up a simple RBC model with only search and matching frictions in the labour market to show that these components have different, quantitative and sometimes qualitative, effects on output, private wages and employment, the unemployment rate and private consumption. Using simulated data we show that a VAR with aggregate government spending and output does not identify any type of fiscal shock. We then use the several identification strategies proposed in the literature to understand the effects of different components, for the United States. We find that both the average wage and employment have larger multipliers than purchases of intermediate goods, investment and transfers. They also have distinct effects on private wages and private consumption.