Do Leaders Affect Government Spending Priorities?

Working Paper: NBER ID: w15368

Authors: Adi Brender; Allan Drazen

Abstract: Since a key function of competitive elections is to allow voters to express their policy preferences, one might take it for granted that when leadership changes, policy change follows. Using a dataset we created on the composition of central government expenditures in a panel of 71 democracies over 1972-2003, we test whether changes in leadership induce significant changes in spending composition, as well as looking at the effect of other political and economic variables. We find that the replacement of a leader tends to have no significant effect on expenditure composition in the short-run. This remains true after controlling for a host of political and economic variables. However, over the medium-term leadership changes are associated with larger changes in expenditure composition, mostly in developed countries. We also find that in established democracies, election years are associated with larger changes in expenditure composition while new democracies, which were found by Brender and Drazen (2005) to raise their overall level of expenditures in election years, tend not to have such changes.

Keywords: government spending; leadership; political budget cycles

JEL Codes: D72; D78


Causal Claims Network Graph

Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.


Causal Claims

CauseEffect
Replacement of a leader (D79)Expenditure composition (E20)
Election years (K16)Expenditure composition (E20)
New democracies (P39)Expenditure composition (E20)

Back to index