The Macroeconomics Effects of Government Spending Under Fiscal Foresight

• Author: Ruthira Naraidoo,Charl Jooste
• Published date: 01 March 2017
• DOI: http://doi.org/10.1111/saje.12151
• Date: 01 March 2017

THE MACROECONOMICS EFFECTS OF GOVERNMENT

SPENDING UNDER FISCAL FORESIGHT

CHARL JOOSTE

†

AND RUTHIRA NARAIDOO*

†

Abstract

Consumption and output responses to fiscal shocks are studied in a model with fiscal foresight.

Fiscal foresight reduces both output multipliers and consumption. However, key features such as

sticky wages, credit constrained households and elastic labour supply, are able to generate both

sizeable output multipliers and positive consumption, in effect preserving key Keynesian effects.

This model fits a developing economy like South Africa well since it is able to capture transpar-

ent communication of government as well as control for credit constrained consumption and

sticky wages.

JEL Classification: E12, E32, E62, H30, H31

Keywords: General equilibrium, fiscal spending, foresight

1. INTRODUCTION

This paper disentangles the effects of fiscal shocks when agents have foresight. As high-

lighted by Ramey (2011) and Leeper et al. (2012) models that do not explicitly account

for foresight is misspecified and may produce faulty output and consumption responses

to fiscal shocks. To address the issue of foresight we build on the familiar models of Gali

et al. (2007) and Furlanetto (2011). We investigate whether certain conditions (sticky

wages, share of rule-of-thumb consumption and elastic labour supply) are able to pre-

serve the results of Gali et al. (2007) and Furlanetto (2011) under foresight, or whether

they will inevitably complement the results obtained by Ramey (2011). The Gali et al.

(2007) model requires an exogenous wage mark-up shock along with a large share of

rule-of-thumb consumers to generate positive aggregate household consumption

responses given a fiscal shock. Furlanetto’s (2011) sticky wage model still generate posi-

tive consumption responses despite a fall in household wages due to the beneficial effect

of lower interest rate response. The two models are closely related and are appropriate

for our study in trying to analyse whether foresight eliminates positive consumption.

These models are also well-suited to developing economies like South Africa and capture

essential features such as union bargaining on wages, credit constrained households,

sticky prices reflecting the monopolistic competition in the market (Fedderke and Schal-

ing, 2005), investment adjustment costs and transparent communication by government

on future spending and taxes.

The paper has two contributions. The first shows that fiscal foresight eliminates the

results obtained in Gali et al. (2007), i.e. consumption does not increase with respect to

* Corresponding author: Faculty of Economics and Management Sciences, Department of

Economics, University of Pretoria, Pretoria 0002, South Africa.

E-mail: ruthira.naraidoo@up.ac.za

†

University of Pretoria

V

C2017 Economic Society of South Africa. doi: 10.1111/saje.12151

68

South African Journal of Economics Vol. 85:1 March 2017

South African Journal

of Economics

an increase in fiscal spending and consequently that a large share of rule-of-thumb con-

sumption is not enough to preserve their original results. The second contribution shows

that sticky wages, as in the model of Furlanetto (2011), minimises the negative impact of

fiscal foresight. Under certain calibrations we show that with sticky wages one can still

generate positive consumption responses and produce sizeable output multipliers.

It is important from the outset to highlight why sticky wages produce such different

results. The main difference follows from labour supply and in particular how wages

react. Since the model contains lump sum taxes, any increase in fiscal spending today

will have to be financed through higher taxes in the future. The lump sum tax, although

not distortionary, shifts the consumers’ budget constraint to the left. Consequently, opti-

mising consumers reduce their consumption immediately. This negative wealth effect

induces consumers to supply more labour, which results in a decrease in the real wage

(see Ramey 2011). Here sticky wages ensure that the fall in real wages is muted. A sec-

ond channel is also at play – an increase in fiscal spending increases prices through an

increase in aggregate demand, which increases the monetary policy rate. Higher interest

rates reduce investment and decreases consumption even more. Under sticky prices and

wages, prices and wages are less prone to a government shock and hence interest rates are

muted and have less of an effect on consumption. This result also holds for news shocks.

We add two types of foresight, viz., medium and high to fiscal shocks, which comple-

ments the works by both Gali et al. (2007) and Leeper et al. (2012) in investigating fiscal

expansions. Finally, we provide sensitivity analysis as to how the model behaves under

different labour supply elasticities and different shares of rule-of-thumb consumption.

Given that the response of labour is crucial within the context of consumption

response to government spending, we argue that elastic labour supply along with sticky

wages counters the negative effects of foresight on consumption even more. With a large

share of rule-of-thumb consumption and elastic labour typical of most developing coun-

tries, some of the Gali et al. (2007) results, sign not size, are preserved.

2. LITERATURE REVIEW

The effects of foresight have become increasingly important in trying to explain differen-

ces in empirical and theoretical work on the macroeconomic consequences of fiscal

shocks. Blanchard and Perotti’s (2002) paper serves as reference point for the majority of

empirical work and has been highly contested by Ramey (2011). While there seems to

be some agreement on the effects of government spending on output, there exists a major

disagreement regarding the effects of spending shocks on consumption and wages. Some

authors find strong evidence of Keynesian effects – both consumption and wages increase

in response to spending (Rotemberg and Woodford, 1992; Blanchard and Perotti, 2002;

Gali et al., 2007), while others show that household consumption decreases when gov-

ernment spending increases (Baxter and King, 1983; Ramey and Shapiro, 1998; Hall,

2009; Cogan et al., 2009, Farmer and Plotnikov, 2012).

Ramey and Shapiro (1998), Ramey (2011) and Leeper et al. (2012) argue that many

shocks identified from the standard VAR are anticipated changes in government spend-

ing and are not pure government spending shocks. Some agents have information about

future spending and tax policy changes before they are implemented, especially when

governments communicate transparently. This is called fiscal foresight. This often arises

due to pre-announced policy changes, legislative lags or simple policy implementation

69South African Journal of Economics Vol. 85:1 March 2017

V

C2017 Economic Society of South Africa.