Can Automatic Government Spending Be Procyclical?

Working Paper: NBER ID: w28521

Authors: Luciana Galeano; Alejandro Izquierdo; Jorge P. Puig; Carlos A. Vegh; Guillermo Vuletin

Abstract: It is well-known by now that government spending has typically been countercyclical in industrial countries and procyclical in developing economies. Most of this literature has focused on analyzing aggregate government spending or discretionary spending categories such as government consumption and government investment. Little is known, however, about the cyclical behavior of automatic government spending, which comprises unemployment insurance, family programs, and social security transfers. Automatic government spending follows from laws, or even constitutional clauses, that benefit individuals who meet certain eligibility criteria. In principle, the main categories of automatic government spending are expected to be either countercyclical (especially unemployment insurance and other shock absorber programs) or acyclical (particularly social security and other structural programs). \nWe find that while automatic government spending is, as expected, countercyclical in industrial countries, it is, surprisingly, procyclical in the developing world. We track the source of this puzzling procyclical behavior to (i) the effective lack of automatic stabilizers like unemployment insurance and (ii) more intriguingly, the existence of perverse automatic de-stabilizing mechanisms in social security spending (in particular in the absence of indexation mechanisms). We also show that the presence and nature of these two social programs are crucial new determinants of aggregate government spending cyclicality as well as macroeconomic volatility, even after controlling for other well-known determinants and addressing potential endogeneity concerns.

Keywords: government spending; procyclical; automatic stabilizers; developing countries; social security

JEL Codes: E02; E32; E62; H53; H55


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
automatic government spending (E62)macroeconomic volatility (E39)
lack of automatic stabilizers (E65)procyclical behavior of automatic spending (E62)
destabilizing nature of social security spending (H55)procyclical behavior of automatic spending (E62)
procyclical behavior of automatic spending (E62)increased macroeconomic volatility (E32)
economic growth (O49)automatic government spending (E62)
lack of countercyclical spending policies (E62)increased vulnerability of disadvantaged populations (I14)

Back to index