Changes in the revenue-expenditure nexus: confronting evidence with fiscal policy in Brazil

This study examines causality shifts in the revenue-expenditure nexus in Brazil during the years 1996-2019. The literature usually assumed a time-invariant causality link between government revenues and expenditures. Such assumption becomes often implausible when the sample period covers significant economic, political and policy changes, like the case of Brazil in the last two decades. Based on time-varying Granger causality tests, we found three distinct causality episodes: fiscal synchronization in 2005-2008, spend-and-tax in 2010-2013, and a tax-and-spend in 2015-2018. These episodes are intercalated with periods of no causality that coincides with economic crisis, political turmoil and major shifts in fiscal policy. The findings suggest that the interplay between fiscal policy orientation and primary surplus targeting is an important factor in the revenue-expenditure nexus. The most recent causality episode suggests stronger effects running from revenues to expenditures in the Brazilian primary budget dynamics. In this context, according to the tax-and-spend hypothesis, cutting spending seems to be the right course of action to restore fiscal balance in Brazil. A tax increase should be considered with caution because it may only incentive the government to grow its expenditures.