On the Difference Between Tax and Spending Policies in Models with Finite Horizons

Working Paper: NBER ID: w2557

Authors: William H. Branson; Giampaolo Galli

Abstract: This paper uses the Blanchard (1985) finite horizon model to study how taxes and government spending can be managed to stabilize aggregate demand. It is shown that tax policy cannot stabilize demand in less time than it stabilizes the public debt, but that, if government spending is the instrument of policy, demand can be stabilized independently of the dynamics of the debt. These results imply that if the objective is to stabilize the debt while maintaining demand as close as possible to a pre-determined target path, and taxes are the instrument, taxes would have to be changed temporarily as much as feasible. On the other hand, if the instrument is government spending, it can be changed gradually to achieve the objectives. The dynamic effects of taxes are a straightforward implication of the intertemporal budget constraint, when it is assumed that agents cannot be surprised by government policies. More traditional dynamics can be obtained if it is assumed that the government succeeds in announcing a policy and implementing a different one. If however the announcement is no credible, discretion is inferior to a predetermined tax rule.

Keywords: tax policy; government spending; aggregate demand; public debt; finite horizons

JEL Codes: E62; H20


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
Government Spending (H59)Public Debt (H63)
Public Debt (H63)Consumption (E21)
Government Spending (H59)Consumption (E21)
Taxes (H29)Consumption (E21)
Government Spending (H59)Stabilization of Aggregate Demand (E00)
Taxes (H29)Stabilization of Demand (J23)

Back to index