Government consumption, government debt and economic growth

• Published date: 01 May 2020
• Author: Shahrzad Ghourchian,Hakan Yilmazkuday
• DOI: http://doi.org/10.1111/rode.12661
• Date: 01 May 2020

Rev Dev Econ. 2020;24:589–605. wileyonlinelibrary.com/journal/rode

|

589

© 2020 John Wiley & Sons Ltd

1

|

INTRODUCTION

The Great Recession of 2007–2009 has resulted in many governments bailing out their financial in-

stitutions and even providing finance for the real sector using government resources. Combined with

the necessity of an expansionary fiscal policy due to the restricted monetary policy at the zero lower

bound, many governments around the world started having problems regarding their budgets, and they

eventually employed austerity measures, potentially at the cost of their economic growth. Influential

Received: 6 January 2017

|

Revised: 8 February 2020

|

Accepted: 11 February 2020

DOI: 10.1111/rode.12661

REGULAR ARTICLE

Government consumption, government debt and

economic growth

ShahrzadGhourchian

|

HakanYilmazkuday

Oklahoma City University, Oklahoma City,

OK 73106, USA

Correspondence

Hakan Yilmazkuday, Department

of Economics, Florida International

University, Miami, FL 33199, USA.

Email: skuday@gmail.com

Abstract

This paper compares the effects of government consump-

tion and government debt on economic growth using data

from 83 countries, including both developed and develop-

ing markets, over the period from 1960 to 2014. Linear

regressions reveal that the negative effects of government

consumption are relatively higher than the negative effects

of government debt. A nonlinear investigation further sug-

gests that the restrictions on government expenditure to

prevent negative growth are more important for countries

with lower trade openness, lower inflation, or greater finan-

cial depth, whereas the restrictions on government debt are

shown to be more important for countries with higher trade

openness, lower inflation or greater financial depth.

KEYWORDS

cross-country analysis, economic growth, government consumption,

government debt, thresholds

JEL CLASSIFICATION

H50; H63; O23; O47

590

|

GHOURCHIAN ANd YILMAZKUdAY

studies such as by Rogoff and Reinhart (2010) have ignited the debate around such budget problems

and their impact on growth from a policy perspective by showing a negative correlation between gov-

ernment debt and growth for countries with debt above 90% of their gross domestic product (GDP)

for the post-World War II era.

Within this picture, though, the effects of government consumption/expenditure on growth have

not been investigated and compared enough with those of government debt. While government debt

can affect growth through reductions in public saving (e.g. as in Elmendorf and Mankiw, 1999), gov-

ernment consumption affects growth through factor accumulation or influences on technical progress

such as public research and development (as discussed in Gemmell, 2001), the reductions in company

profits and private investment (as in Alesina, Ardagna, Perotti, and Schiantarelli, 2002), or organized

interest groups attempting to gain benefits for themselves in the form of legislation or transfers (as

in Olson, 2008). Such a comparison between government consumption and government debt is also

important from the policy perspective. For example, in 2012, Carlo Cottarelli, former Director of the

Fiscal Affairs Department, of International Monetary Fund, wrote:1

Government debt remains very high in many advanced economies, and fiscal adjustment

to bring debt down over the medium term is essential. Nearly all advanced economies

plan to reduce their deficits this year. But if growth slows more than expected, some may

feel inclined to preserve their short-term plans through additional tightening, even if [it]

hurts growth more. My bottom line for them: unless you have to, you shouldn’t.

He also emphasized the importance of country-specific fiscal policies due to the economic character-

istics of the countries. Accordingly, the debate is not only about government debt itself but also about the

short- and medium-term adjustments of fiscal policies which can be measured by government consump-

tion/expenditure and/or tax revenues.

Based on the discussion so far, in this paper, we compare the effects of government consumption

and government debt on growth by using data from 83 countries over the period between 1960 and

2014, including both developed and developing markets.2

In order to connect our results to the ex-

isting studies, we first consider linear regressions that are supported by statistical tests regarding the

potential issue of endogeneity. Such a linear investigation results in government consumption hav-

ing a bigger reducing impact on growth compared to the negative effects of government debt. This

result contradicts the argument in studies such as by Elmendorf and Mankiw (1999) who show that

government consumption may boost aggregate demand due to the Keynesian view, but government

debt may lead to a reduction in output and income by reducing investment and the capital stock. We

further investigate this contradiction by considering nonlinear/threshold effects of government fiscal

policies on growth as advocated in studies such as by Kumar and Woo (2010), Rogoff and Reinhart

(2010), Cecchetti, Mohanty, and Zampolli (2011), Checherita-Westphal and Rother (2012), Panizza

and Presbitero (2014), Baum, Checherita-Westphal, and Rother (2013), and Pescatori, Sandri, and

Simon (2014). Such nonlinear analyses show that the effects of both government consumption and

government debt on growth are highly affected by the economic characteristics of the countries in-

vestigated. It is implied that certain countries should pay more attention to their government expendi-

ture, while certain others should pay more attention to their government debt, if they wish to prevent

negative economic growth.

Overall, this paper contributes to the literature by comparing the effects of government expenditure

and government debt, by using a rich data set with more countries and time coverage than existing

studies, and by considering nonlinearities in the relationship between growth and government expen-

diture/debt that are essential in the determination of country-specific policies. The rest of the paper is