General budget support as an aid instrument—impact on economic growth

• Date: 01 February 2019
• Author: Susanna Sandström,Kaisa Alavuotunki
• DOI: http://doi.org/10.1111/rode.12535
• Published date: 01 February 2019

REGULAR ARTICLE

General budget support as an aid instrument—

impact on economic growth

Kaisa Alavuotunki

1

|

Susanna Sandström

2

1

Aalto University, Helsinki, Finland

2

Åbo Akademi University, Turku,

Finland

Correspondence

Kaisa Alavuotunki, Finnish Fund for

Industrial Cooperation Ltd,

Uudenmaankatu 16B, P.O. Box 391,

00121 Helsinki, Finland.

Email: kaisa.alavuotunki@finnfund.fi

Funding information

Financial support from Bröderna Lars och

Ernst Krogius forskningsfond and the

Yrjö Jahnsson Foundation is gratefully

acknowledged.

Abstract

Since the change of the new millennium, general budget

support (GBS) has become a prominent, yet controversial

and heavily debated, modality for delivering aid. We

study GBS as an aid instrument from a cross‐country per-

spective. We examine if any growth impacts can be iden-

tified as a result of the use of GBS. We use data

covering nine 4‐year intervals from 1976 to 2011. We

modify two supply‐side IV strategies from previous aid–

growth literature. In our main approach, we employ an

interaction of an exogenous supply‐side variable (donor

government fractionalization) and an endogenous variable

(probability of receiving GBS) as an instrument for GBS,

and in the alternative approach we construct an instru-

ment following a supply‐side approach. Our results sug-

gest that GBS receiving countries have grown faster than

countries receiving other types of aid. Selection bias does

not explain this result. The growth effect is not only

attributed to lagged GBS but also to contemporaneous

GBS flows.

1

|

INTRODUCTION

There are numerous ways in which aid can have an impact on the recipient country. One of the

single most debated topics in the field of aid impact studies is the relationship between aid and

growth. Since Mosley (1987) suggested the existence of micro–macro paradox,

1

economists have

been keen to find evidence as to whether anything can be said about the growth impact of “total

aid”—and most attention has been put into the existence (or nonexistence) of such a relationship.

However, as a paper by Clemens, Radelet, Bhavani, and Bazzi (2012) shows, the impact of aid on

growth varies depending on the concepts used. Some portions of aid, such as humanitarian

DOI: 10.1111/rode.12535

Rev Dev Econ. 2019;23:231–255. wileyonlinelibrary.com/journal/rode © 2018 John Wiley & Sons Ltd

|

231

assistance, are not given with the intention to affect economic growth and may not necessarily do

so, at least not in the short run. This motivates further research on different forms of aid. Disaggre-

gating aid can contribute to a deeper understanding of how foreign aid works.

In this paper, we study the impact of general budget support (GBS) on economic growth from

a cross‐country perspective. The introduction of this aid modality has its background in a paradigm

shift in development cooperation. This shift has its origins in the 1990s, to a large extent growing

out of a perception that inappropriate aid modalities had become part of the development problem

in aid‐dependent countries (IDD and Associates, 2006). Efforts to bypass weaknesses in govern-

ment systems had weakened them further and led to a fractionalization of both national decision‐

making and aid‐giving. In the quest to improve aid effectiveness, GBS was introduced to be used

in countries committed to reform, channeling aid directly to partner governments using their own

allocation, procurement, and accounting systems, without being linked to specific project activities.

The aim of the partnership approach underlying the GBS agenda was to move away from tradi-

tional conditionality towards a country‐led development process and a more long‐term accountabil-

ity framework.

2

The results chain envisaged is shown in Figure 1. While the long‐term goal is

poverty reduction in a broad sense, there are also intermediate outputs and outcomes, such as more

predictable resource flows, and improved economic growth that are of importance.

After a more than decade‐long effort to promote GBS, it is of interest to study if any growth

impacts can be identified as a result of the use of GBS. Whether one should expect to see any

growth effects of GBS within a decade is debatable. If GBS funds largely are used on investments

in social sectors, the growth effects might come with a very long time lag. These results should

therefore be interpreted as some early impacts of GBS.

The rest of the paper is organized as follows. Section 2 discusses previous literature on GBS

and growth. Section 3 presents the data and delivers basic descriptive statistics of GBS. In Sec-

tion 4, we present the approach used in studying the impact of GBS on growth. Section 5 displays

the results and Section 6 concludes.

FIGURE 1 Diagrammatic overview of aggregate impact of GBS

Source. Adapted from Lawson et al. (2013, p. ix) [Color figure can be viewed at wileyonlinelibrary.com]

232

|

ALAVUOTUNKI AND SANDSTRÖM