Does energy‐related aid affect emissions? Evidence from a global dataset

• Author: Andy McKay,Maurizio Intartaglia,Sambit Bhattacharyya
• Published date: 01 August 2018
• DOI: http://doi.org/10.1111/rode.12409
• Date: 01 August 2018

REGULAR ARTICLE

Does energy-related aid affect emissions? Evidence

from a global dataset

Sambit Bhattacharyya

|

Maurizio Intartaglia

|

Andy McKay

Department of Economics, University of

Sussex

Correspondence

Sambit Bhattacharyya,

Department of Economics, University of

Sussex, Brighton, UK.

Email: S.bhattacharyya@sussex.ac.uk

Funding information

Directorate-General for Research and

Innovation, Grant/Award Number:

308601

Abstract

Donor countries have been using international aid in the

field of energy for at least three decades. The stated

objective of this policy is to reduce emissions and pro-

mote sustainable development in the global South. In

spite of the widespread use of this policy tool, very little

is known about its effect on emissions. In this paper we

perform an empirical audit of the effectiveness of energy-

related aid in tackling CO

2

and SO

2

emissions. Using a

global panel dataset covering 128 countries over the per-

iod 1971–2011 and estimating a parsimonious model

using the Anderson and Hsiao estimator, we do not find

any evidence of a systematic effect of energy-related aid

on emissions. We also find that the non-effect is not con-

ditional on institutional quality or level of income. Coun-

tries located in Europe and Central Asia do better than

others in utilizing this aid to reduce CO

2

emissions. Our

results are robust after controlling for the environmental

Kuznets curve, country fixed effects, country-specific

trends, and time-varying common shocks.

1

|

INTRODUCTION

Modern industrial society runs on fossil fuel. Burning fossil fuel releases thermal energy which is

then transformed into electricity. Electricity is a key input in the production of goods and services

destined for mass consumption. Consumers derive satisfaction from the consumption of these

mass-produced goods. In modern society, sustained improvement in the average level of consump-

tion is a key indicator of material well-being and improved living standards. The use of fossil fuel

not only generates thermal energy but also releases greenhouse gases (carbon dioxide, sulfur diox-

ide, methane, and others) into the atmosphere, causing global warming and climate change. Until

DOI: 10.1111/rode.12409

1166

|

©2018 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/rode Rev Dev Econ. 2018;22:1166–1194.

recently the environmental consequences of industrialization were largely ignored. The global

threat of a catastrophic climate change has helped raise awareness and brought countr ies together

in favor of a coordinated policy response.

In a globalized world of free trade and migration (to a lesser extent), global governance of

emissions mitigation is challenging. It is relatively inexpensive for industrial production to cross

borders and move to cheaper locations. Indeed, starting from the 1980s, the world has noticed a

significant dislocation of industries from the industrialized nations to the emerging markets, signifi-

cantly increasing the latter’s share of greenhouse gas emissions. Coupled with the global challenge

of reducing greenhouse gas emissions, this migration of polluting industries brings in a key ques-

tion of distributive justice in a Rawlsian sense.

1

To what extent should the emissions of emerging

market economies be restricted so that the objectives of sustainable development and reducing glo-

bal greenhouse gas emissions can be achieved?

At the operational level, states around the world have aimed to address these challenges by

making use of both bilateral and multilateral institutional mechanisms. In particular, countries have

used the mechanism of international transfers, especially in the field of energy, to achieve the twin

objectives of emissions reduction and sustainable development. Policy-makers have been using

these policy tools for at least three decades, yet the effects are not very well known. To the best of

our knowledge, there is hardly any systematic quantitative research on the effect of energy-related

aid on emissions in the aid recipient countries. In this paper, we seek to explore this very question:

do we notice a perceptible difference in the level of emissions in the aid recipient countries as a

result of energy-related aid going back to the 1970s?

A cursory look at the global aggregates reveals that both foreign aid commitment and disburse-

ment for the energy sector (especially electricity generation) have exploded over the last decade.

For example, per capita aid disbursement for power generation over the 2000s grew by 4 percent

per year on average, whereas the annualized growth rate of aid commitment in power generation

for the same period was approximately 5 percent. Carbon dioxide (CO

2

) emissions, however, have

increased at an annualized rate of 2.5 percent over the same period. Emissions of sulfur dioxide

(SO

2

) have declined since the mid-1990s, largely due to the introduction and subsequent adoption

of unleaded fuels for transport. Figures 1–4 present these data.

Even though there has been some degree of co-movement between emissions and energy-

related aid, it is problematic to interpret this association as causal. What we plot are global trends

which ignore variations within and across countries. A third latent factor could also be responsible

for the co-movement. Furthermore, there is no obvious theoretical prior when it comes to the effect

of energy-related aid on emissions. On the one hand, policy-makers in donor countries woul d

expect results in terms of reduced emissions through better targeting of the energy infrastructure in

the recipient countries. On the other hand, this aid could very well be off target and spent on pro-

jects that have little discernible impact on emissions. Therefore, the lack of a strong prior either

way makes this policy design a prime candidate for empirical audit. More detailed and systematic

modeling is necessary to understand the co-movement in the raw time series data.

In this paper we aim to systematically explore the effect of energy-related aid on CO

2

and SO

2

emissions. In particular, we analyze the effect of an energy-related aid shock on emissions using a

panel data model. We exploit a global panel dataset covering 128 countries over the period 1971–

2011. Note that our aid data are sourced from AidData.org. This dataset is an improvement over

the Creditor Reporting System (CRS) maintained by the OECD’s Development Assistance Com-

mittee (DAC) and offers far wider country coverage. Furthermore, our dataset allows us to distin-

guish between renewable and non-renewable sources of power generation, and energy supply

infrastructure. We estimate a parsimonious model using fixed effects, Arellano–Bond, and

BHATTACHARYYA ET AL.

|

1167