The Dynamic Effects of Fiscal Stimulus in a Two‐Sector Open Economy

• DOI: http://doi.org/10.1111/rode.12054
• Date: 01 August 2013
• Author: Anthony Makin,Ross Guest
• Published date: 01 August 2013

The Dynamic Effects of Fiscal Stimulus in a

Two-Sector Open Economy

Ross Guest and Anthony Makin*

Abstract

In 2009/10 governments around the world implemented unprecedented fiscal stimulus in order to counter

the impact of the Global Financial Crisis of 2008/09. This paper analyzes the impact of fiscal stimulus using

a dynamic open economy, overlapping generations model that allows for feedback effects of fiscal stimulus

on private sector expenditure via changes in the tax rate and the interest rate. There are two types of

goods—traded (T) and nontraded (N) goods—which differ in their capital intensities. The main qualitative

result is that the dynamic output gains from fiscal stimulus depend on the productivity of the initial stimulus

spending, on the speed of repayment of debt, on the sensitivities of the interest rate to government debt

and of labor supply to the tax rate. Also, the overlapping generations framework allows an

intergenerational welfare analysis. Among the biggest winners from stimulus are those about to retire. The

biggest losers are those near the start of their working lives when the stimulus is implemented.

1. Introduction

The internationally coordinated fiscal response of the Group of 20 (G20) economies

to the Global Financial Crisis (GFC) of 2008/09 has restored fiscal activism to

the forefront of macroeconomic analysis. Concerned about the possibility of

another Great Depression, G20 economies implemented the largest ever worldwide

fiscal expansion to stimulate aggregate spending in order to prevent a rise in

unemployment.

Discretionary fiscal stimulus was strongly encouraged by the International Mon-

etary Fund (IMF) in response to the GFC, at least partly on the assumption that

households react to increases in current disposable income. The IMF specifically

advocated measures that were “timely, large and lasting” (Spillembergo et al., 2008),

despite the already weak pre-crisis budgetary positions of many advanced economies

experiencing high fiscal deficits and public debt levels. In response, G20 governments

boosted public spending and cut taxes to deliver stimulus packages valued at 2% of

world gross domestic product (GDP) in 2009 and 1.6% in 2010 (IMF, 2010). The bulk

of public spending in most advanced economies was directed to the nontraded sector.

Deficits and debt levels worsened considerably in major economies during the crisis,

notably in the USA, Japan and the UK, as a result of the effects of both automatic

stabilizers and discretionary measures.

In the long running debate on the merits and demerits of using discretionary fiscal

policy, the Keynesian perspective that government spending, taxes and transfers can

be readily deployed to minimize fluctuations in the business cycle, and hence employ-

ment levels, contrasts with alternative theoretical approaches that imply that fiscal

activism is ineffective. Given positive interest rates, various alternative approaches,

* Guest: Department of AFE, Gold Coast Campus, Griffith University, 4222 Australia. Tel: +61-7-5552-

8783; Fax: +61-7-5552-855; E-mail: r.guest@griffith.edu.au. Makin: Department of AFE, Gold Coast

Campus, Griffith University, 4222 Australia.

Review of Development Economics, 17(3), 609–626, 2013

DOI:10.1111/rode.12054

© 2013 John Wiley & Sons Ltd

such as the Mundell (1963)–Fleming (1962), Ricardian and traditional loanable analy-

sis, suggest fiscal multipliers are zero. See also Coleman (2010).

For instance, with perfect capital mobility and a floating exchange rate, the

Mundell–Fleming model proposes that fiscal stimulus fully crowds out net exports,

while Ricardian equivalence proposes that the higher public debt associated with

increased government borrowing crowds out private consumption. Moreover, at the

practical level, the implementation of fiscal stimulus is often beset by timing problems

that can make fiscal stimulus pro-cyclical (see Lane, 2003; Leigh and Stehn, 2009).

Empirical studies, using a range of methods and econometric techniques, provide

mixed results about whether discretionary fiscal policy affects aggregate private

expenditure, and estimates of fiscal multipliers vary widely across economies and

through time. For instance, vector autoregression (VAR) studies by authors including

Blanchard and Perotti (2002), Mountford and Uhlig (2009), and Gali et al. (2007)

tend to find positive short-run effects of fiscal stimulus on output and consumption,

with occasional negative effects on private investment. So-called narrative studies that

focus on the stimulatory effects of very high government spending on private con-

sumption during particular episodes by, among others, Ramey and Shapiro (1998),

and Edelberg et al. (1999), have found a weaker, sometimes negative, relationship.

Meanwhile, Denaux (2007) examined the effects of public spending and taxes on

long-run economic growth in an endogenous growth framework, whereas Park (2010)

and Bhattarai (2011) modeled the effects of fiscal policy on a range of macroeconomic

variables using general equilibrium approaches.

Overall, the extant literature on using fiscal policy to influence national income and

employment, as surveyed by Auerbach et al. (2010), mostly presumes that economies

are closed to international economic influences. In contrast, this paper assumes a

small open economy, and distinguishes between short- and long-run effects of

increased government spending.

The next section establishes the foundations of a two-sector open economy simula-

tion model based on the tradables–nontradables dichotomy of goods and services

markets. The purpose of the simulation model is to investigate the effects of fiscal

stimulus over time. Section 3 introduces the government sector and the fiscal multi-

plier; section 4 presents the key simulation results illustrating negative feedback

effects of fiscal stimulus over the longer run; and section 5 concludes the paper.

2. A Dynamic Open Economy Simulation Model

The following outlines the key relationships underpinning a small economy’s goods

and services’ markets before incorporating the firm sector, the household sector the

labor market and government spending.

Specification of Goods and Services’ Markets

In the tradition of the dependent economy model originally proposed by Salter (1959)

and Swan (1960), and adapted by, inter alia, Fischer and Frenkel (1972), Bruno

(1976), Obstfeld and Rogoff (1996), and Yano and Nugent (1999), the economy pro-

duces and consumes two distinct classes of goods and services—tradables (T) and

nontradables (N). The prices of nontradables are set by domestic demand and supply

of nontradables. The foreign currency prices of tradables are set in world markets,

and converted to domestic values via the prevailing nominal exchange rate, that is:

610 Ross Guest and Anthony Makin

© 2013 John Wiley & Sons Ltd