Workers’ remittances and government size

• DOI: http://doi.org/10.1111/rode.12401
• Author: Kevin Williams
• Date: 01 August 2018
• Published date: 01 August 2018

REGULAR ARTICLE

Workers’remittances and government size

Kevin Williams

The University of the West Indies, St

Augustine, Trinidad and Tobago

Correspondence

Kevin Williams, Department of

Economics, The University of the

West Indies, St. Augustine, Trinidad

and Tobago.

Email: Kevin.Williams@sta.uwi.edu

Abstract

In this paper I use a panel of 133 developing countries

for the period 1972 to 2012 to estimate the effect of

workers’remittances on the size of government. Control-

ling for country and period fixed effects that jointly affect

workers’remittances and government size, the paper

finds that workers’remittances are associated with larger

government. The paper further examines whether demo-

cratic institutions mediate the relationship between work-

ers’remittances and government size. Democratic

institutions do not significantly mediate the effect that

workers’remittances have on the size of government.

1

|

INTRODUCTION

This paper investigates the effect that workers’remittances have on government spending

1

in

developing countries. Estimates from the Word Bank (2015) indicate that international migrants

sent home U.S.$418 billion in 2013 and U.S.$436 billion in 2014. What effects do these large

sums of monies have on government spending in recipient developing countries? For many devel-

oping countries, workers’remittances are the main source of foreign exchange (Chami et al.,

2008; Lartey, 2017; Ratha, 2007).

Thus, empirical evidence of the impact that workers’remittances have on government spending

in developing countries is important for several reasons. From a policy perspective, evidence of

the role of workers’remittances on government spending will inform fiscal-policy debates in recip-

ient developing countries and thus help governments to better manage their fiscal performance .

During periods of economic downturns, for example, governments can exploit the steady inflow s

of workers’remittances (by selling bonds to migrants) to smooth spending on public goods, such

as public health and security, two important components of government spending. Given these pos-

itive effects of workers’remittances on government spending, governments can therefore develo p

appropriate incentives to encourage their citizens who are living abroad to send more money to

their country of origin.

Further, the evidence here will provide insights for international organizations such as the

World Bank and the IMF and other credit-rating agencies that regularly assess the economic status

DOI: 10.1111/rode.12401

Rev Dev Econ. 2018;22:e115–e134. wileyonlinelibrary.com/journal/rode ©2018 John Wiley & Sons Ltd

|

e115

of developing countries, to now consider the importance of workers’remittances when assessing

the fiscal position of recipient developing countries. Additionally, researchers studying the effects

of the volatility of short-term capital flows in developing countries will have an improved under-

standing about a more stable form of international capital flows. Finally, as reflected in Section 2,

previous evidence of the relationship between workers’and government spending is lacking. This

study therefore fills this gap in the economic literature.

Although a larger government size enables developing countries to satisfy their public-good

obligations and be more resilient to external (global financial crises) and environmental shocks

arising from natural disasters, it nonetheless increases the potential for corruption. For example, it

is possible that a larger government size can enable political elites in remittance-receiving develop-

ing countries to reallocate expenditures away from public goods and services and toward political

patronage through increased public sector employment (Ahmed, 2012). Remittances make this

strategy more likely by increasing households’income, which allow recipients to purchase public

goods and services in private markets and hence removes the democratic pressure for government

to provide them, which in turn increases incentives for government to use public spending to buy

political support in order to consolidate their political power. Thus, because remittances can

weaken the link between recipient households’welfare and government performance, political

elites face fewer constraints to engage in patronage spending. This view generates the prediction

that in countries with weak democratic institutions remittances should expand the size of govern-

ment.

I further examine whether and how democratic institutions mediate the relations hip between

government spending and remittances in the panel of developing countries. Countries with strong

democratic institutions have more checks and balances and governments are therefore more

accountable to voters. In these countries, increased remittances are unlikely to cause political elites

to increase public spending in order to gain political support, because the probability of being

voted out of office is greater. Thus, remittances are expected to lead to lower government spending

in countries with strong democratic institutions.

Using a panel of 133 developing countries within a dynamic estimation framework that controls

for the persistence in government spending as well as country and period fixed effects, the empiri-

cal results show that workers’remittances are associated with larger government size. This finding

is robust to alternative model specifications that control for a range of covariates that are expected

to predict government spending. Weak democratic institutions do not significantly mediate the rela-

tionship between workers’remittances and government size.

The remainder of the paper is structured as follows. Section 2 discusses the literature on gov-

ernment size and channels through which workers’remittances affect government size. Section 3

discusses the methodology used to estimate the effect of workers’remittances on the size of gov-

ernment. Section 4 presents the results. Section 5 explores one potential mechanism that mediates

the effect of workers’remittances on the size of government. Section 6 concludes. The

Appendix contains the list of countries used in the analysis and fixed effect OLS estimates

(Tables A1 and A2, respectively).

2

|

RELATED LITERATURE AND CHANNELS LINKING

WORKERS’REMITTANCES TO GOVERNMENT SIZE

The paper contributes to the debate on whether globalization increases government spending. The

seminal paper in this debate is Rodrik (1998). The author shows that countries engaged in

e116

|

WILLIAMS