Foreign Direct Investment and CO<sub>2</sub> Emissions in Sub-Saharan Africa: A Heterogeneous Panel Causality Analysis

Mohamadou Oumarou; Mohammadou Nourou; Ibrahim; Votsoma Philemon

doi:doi:10.11648/j.ijeee.20240905.11

Research Article |

| Peer-Reviewed

Foreign Direct Investment and CO₂ Emissions in Sub-Saharan Africa: A Heterogeneous Panel Causality Analysis

Mohamadou Oumarou^*, Mohammadou Nourou

, Ibrahim, Votsoma Philemon

Published in International Journal of Economy, Energy and Environment (Volume 9, Issue 5)

Received: 20 August 2024 Accepted: 5 September 2024 Published: 23 September 2024

Views: Downloads:

Download PDF

Share This Article

Twitter
Linked In
Facebook

Abstract

Following the density of the literature and the consensus in empirical studies, the aim of this article is to examine the nature of the relationship between foreign direct investment (FDI) and carbon dioxide (CO₂) emissions in sub-Saharan Africa (SSA). To this end, the methodological strategy employed is based not only on a theoretically sound multivariate framework, but also on recent developments in panel data econometrics, namely fully modified ordinary least squares (FMOLS) estimators, dynamic ordinary least squares (DOLS) estimators and the vector error correction model. In addition, the stationarity properties of the panel variables are examined, and the panel cointegration technique is used to test cointegrating relationships in the series of variables. The panel is composed of 38 SSA countries over the period 2000-2022. The main results show that in SSA: the variables move together in the long term. A 1% increase in inward FDI increases CO₂ emissions by 0.210%. This result suggests that FDI has flowed to SSA because of its weak environmental regulations, thus verifying the pollution haven hypothesis. In the long term, there is a bidirectional relationship between inward FDI and CO₂ emissions. In all the models used, renewable energy consumption reduces CO₂emissions. Therefore, SSA needs to put in place effective environmental rules to better guide FDI; put in place strategies to harness and add value to its energy sector, implement policies and strategies that ensure FDI attractiveness without abandoning the environment.

Published in	International Journal of Economy, Energy and Environment (Volume 9, Issue 5)
DOI	10.11648/j.ijeee.20240905.11
Page(s)	105-118
Creative Commons	This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.
Copyright	Copyright © The Author(s), 2024. Published by Science Publishing Group

Keywords

Foreign Direct Investment, CO₂ Emissions, FMOLS, DOLS, Sub-Saharan Africa

1. Introduction

The economic and social status of SSA is still precarious and open to internal and external shocks

[3]

. The developing economies of Africa, and more specifically the countries of SSA, have deployed various practices to ensure the emancipation of their level of sustainable development. Some of these include the promotion of economic growth, industrialization, real agricultural development, financial development, renewable energy consumption and human capital sustainability

[19]

It appears that combining these determinants of growth in the quest for sustainable development has fundamental implications for environmental sustainability. The actual and potential impacts of climate change in SSA are enormous, affecting many aspects of people's daily lives. For example, rising temperatures, drying soils, increased pest and disease pressure, displacement of suitable areas for cultivation, increasing desertification in the Sahara region, flooding, deforestation, and erosion can all be indicators that climate change is beginning and represents one of the greatest environmental, social and economic threats facing SSA.

600,000 people die every year in the African region from causes related to emissions from wood and charcoal burning

[8]

. The situation is more serious for SSA due to poverty, low technological know-how and most importantly, more than half of the population depends on climate-driven enterprises such as small-scale farming, peasant farming, agriculture and hawking

[37]

. Consequently, the issue of global warming does not have the same resonance in developing countries, particularly in SSA and developed countries. Unlike in developed countries, in SSA the energy most consumed is that derived from biomass, which accounts for 50% on average of total energy consumption, of which 60-80% is wood fuel

[45]

Economic growth has slowed in the region over the past two decades due to wars and other reasons, such as bad weather and declining terms of trade

[46]

. Despite recent improvements in economic growth, the sub-continent continues to struggle to reduce poverty, with insufficient formal education and/or low quality of education, poor health leading to short life expectancy. It should also be stressed that SSA countries have economies that are highly dependent on natural resources, and SSA is an area that suffers most from natural disasters and global warming, due to its financial difficulties in adapting to them. In this perspective, due to environmental problems, it is estimated that by 2080, between 9% and 20% of arable land in SSA will become much less suitable for agriculture

[4]

Although SSA is the least integrated and pollutes the environment the least, it is the most vulnerable to future climate change

[2]

. Carbon emissions are also heterogeneously distributed across SSA countries. The main contributors are South Africa, Angola and Nigeria, with 853107.128 kt, 34693.487 kt and 120369.275 kt emitted in 2019 respectively. Guinea-Bissau, Comoros and Sao Tome and Principe represent the lowest levels, with 293.36 kt, 201.685 kt and 121.011kt of CO₂ emitted in 2019

[50]

In several SSA countries, the negative implications of recent economic progress outstrip the capacity of local governments to cope with these consequences. Waste collection and sanitation systems cannot handle the volumes of waste generated by economic activities, leading to significant degradation of urban and aquatic environments

[3]

. Furthermore, theory and evidence suggest that trade openness and FDI promote growth. To take just one example, the remarkable rise of China and the “Four Asian Dragons” is largely due to their integration into the global trading system. Given the many problems associated with an import-substitution industrialization strategy, openness to international trade is now dominant, particularly in developing countries.

Moreover, the ratio of FDI to Gross Domestic Product (GDP) in Africa stands at 5.1%, the highest in the world, confirming the importance of these flows to the continent's economic growth

[20]

. In fact, FDI was on an upward trend between 1995 and 2001, following the opening up to FDI and international trade spurred by the WTO

[38]

. FDI flows to SSA rose to USD 32 billion in 2018, an increase of 13% on the previous two years, which were USD 28.5 billion in 2017 and USD 31.8 billion in 2016

[11]

It is important to emphasize that increasing FDI in SSA countries is an integral part of the international agenda for achieving the Sustainable Development Goals (SDGs). In SSA, the contribution of FDI is relatively higher than investment

[26]

. In addition, the quality of the environment is sensitive to the evolution of FDI

[51]

. Over the period 2000-2015, SSA has relatively attracted capital from outside

[7]

. Since the early 1990s, a large and growing literature has studied the role of foreign direct investment in CO₂ emissions. However, empirical contributions have remained contradictory and inconclusive. Consequently, this study stands out by investigating the causal link between these two variables in SSA.

The remainder of this article is structured around four points: section 2 presents a summary review of the literature; section 3 presents the analytical model; section 4 presents the econometric strategies; section 5 presents the results and interpretations; and section 6 concludes with recommendations.

2. Foreign Direct Investment and CO₂ Emissions: A Review of the Literature

Theoretically, two schools of thought diverge, each supporting a hypothesis. The first school of thought supports the “Pollution Haven Hypothesis”, according to which weak environmental regulation in a host economy attracts FDI and leads to environmental degradation

[10]

. This confirms the positive effect of FDI on CO₂ emissions. The second school of thought supports the “pollution halo hypothesis”, which states that FDI increases energy efficiency and improves environmental quality by reducing carbon emissions through technology transfer and innovative production systems

[16]

This theoretical contradiction has prompted empirical investigations. Empirically, however, the subject is still under discussion around the world due to the different results observed. On the one hand, some studies support the “pollution haven hypothesis” (see for example

[42, 30, 18, 12, 41, 49, 28]

). On the other hand, according to the second approach “pollution halo hypothesis” (see for example

[6, 20, 23]

). In addition, other studies have highlighted a bidirectional relationship between FDI and CO₂ emissions (e.g.

[1, 36]

Despite the surge in empirical literature on these two hypotheses since the early 2000s, empirical results on the impact of FDI on CO₂ emissions are still inconclusive

[5]

. Indeed, according to

[15]

, 54% of studies report a negative effect of FDI on CO₂ emissions versus 46% of studies reporting a positive effect.

3. Presentation of the Analysis Model

Our analysis is carried out in a multivariate framework incorporating variables in the model that are assumed to have effects on CO₂ emissions, such as GDP/h, which allows us to test the existence of the Environmental Kuznets Curve (EKC) hypothesis, and renewable energy consumption. Thus, we empirically examine the dynamic relationship between CO₂ emissions, FDI, renewable energy consumption and GDP for a sample of 38 SSA countries, as the relevant results in the literature appear controversial and ambiguous.

Following the methodology of

[30]

, the relationship between CO₂ emissions per capita and its determinants (FDI, PIB/h and renewable energy consumption) are described in the following model specification:

(1)

Given that we are using a panel approach, equation (1) becomes:

(2)

All variables are expressed in neperian logarithms. CO₂ is carbon dioxide emissions expressed in metric kilotons per capita (lnCO₂); PIB/h is real gross domestic product per capita (lnPIB/h); FDI is inward foreign direct investment as a percentage of PIB (lnIDE); RE is renewable energy consumption as a percentage of total energy consumption (lnER); is the error term that follows a lognormal distribution.

4. Econometric Strategies

The four strategic points in the econometric analysis are as follows:

4.1. Panel Unit Root Tests

This test is based on the classic Augmented Dickey Fuller (ADF) test for the regression of the following equation:

.(3)

Where

is the first difference operator, is the dependent variable, is a white noise disturbance with variance of, and t = 1,..., T.

Which alternative hypothesis corresponds to stationary.

The test is based on the statistic

(where

is the OLS estimator of

in equation (3) and

is its standard error.

[33]

found that the panel approach substantially increases power in finite samples compared to the ADF test equation, and proposed a panel version based on equation (4) that restricts

by keeping it identical across countries as follows:

.(4)

Where i = 1, 2,..., N refers to the countries in the panel.

[33]

tested

With the statistic-based test

The IPS test

[24]

is an extension of the LLC test that relaxes the homogeneity assumption by allowing heterogeneity in the autoregressive coefficients for panel members. This test is based on the mean of the Augmented Dickey-Fuller (ADF) statistics calculated for each individual in the panel. It uses the average of the statistics in equation (3) to perform the following statistic

.(5)

Where,

and

are respectively the mean and variance of each statistic, and they are generated by simulations. Unlike the IPS test, which is parametric and asymptotic,

[34]

and

[9]

propose a simpler, non-parametric unit root test (the Fisher-ADF and Fisher-PP statistics), based on a combination of the p-values of the individual unit root tests. This test is superior to the IPS test

[34]

. Its advantage is that its value does not depend on the different lag lengths in the individual ADF regressions.

[34]

proposed to derive tests that combine p-values from individual unit root tests. If we define as the p-value of any unit root test for the cross-section, then under the null hypothesis of unit root for all cross-sections, we have the following asymptotic result:

.(6)

The null and alternative hypotheses are the same as for

[24]

. Hadri's test is a Lagrange Multiplier (LM) test where the null hypothesis is that there is no unit root in any of the series in the panel versus the alternative of a unit root in the panel. As with the

[32]

test,

[22]

test is based on residuals from individual OLS regressions on a constant, or on a constant and a trend. If we include both the constant and a trend, we derive estimates of:

(7)

Under the null hypothesis of stationarity, the asymptotic result is given by:

;

where

is the cumulative sum of residuals and

is the estimator of

. Hadri (2000) proposes two cases:

and

, if the model includes only the constant;

, if the model includes the constant and the trend.

4.2. Panel Cointegration Test

The second stage of our empirical work concerns the study of the long-term relationship between CO₂ emissions and inward FDI, using the panel cointegration techniques developed by

[39]

. The cointegration relationship is specified by the following equation:

(8)

Where i = 1,..., N denotes the number of individuals in the panel, t = 1,..., T denotes the time dimension,

is the individual specific effect,

is the deterministic trend and

is the estimated residual, which represents deviations from the long-term relationship. The structure of the estimated residuals is as follows:

.(9)

Pedroni proposes seven cointegration tests, four of which are based on the within dimension (intra) and three on the between dimension (inter). Both categories of tests are based on the null hypothesis of no cointegration:

, where

denotes the autoregressive term of the residuals estimated under the alternative hypothesis. In our analysis, in addition to applying

[39]

tests, we also use the cointegration tests proposed by

[27, 25]

to study the relationship between inward FDI and CO₂ emissions in SSA.

The first is the

[27]

test, which is based on the two-stage Engle-Granger procedure, and imposes homogeneity on panel members. The null hypothesis of no cointegration is tested using an ADF-type test.

[27]

proposed the following equation:

.(10)

Where

;

et i = 1, 2, …, N.

One of the tests proposed by

[27]

is based on the ADF test. This test is given by:

.(11)

Where

is selected when the

are uncorrelated under the assumption of no cointegration.

Then, the statistic is:

Where

is the t-statistic of

, and

comes from the covariance matrix

of the bivariate process

We also use Fisher's test to aggregate the p-values of individual cointegration tests from Johansen's maximum likelihood statistics

[34]

. The Fisher test is a non-parametric test that does not assume coefficient homogeneity.

4.3. Estimating the Long-term Panel Relationship: FMOLS and DOLS Estimates

The DOLS and FMOLS estimators are obtained from the following equation:

: i = 1,2,…,N;t = 1,2,…,T(12)

Where

is the logarithm of CO₂ emissions in metric tons per capita and

is the logarithm of the independent variable and

and

are cointegrated with the coefficient

, which may or may not be homogeneous across i. Note that the characteristic equation of the FMOLS and DOLS estimators is an extension of standard regression, in which lags and leads are incorporated into the cointegrating relationship in order to asymptotically reproduce unbiased estimators and avoid the problems associated with estimating nuisance parameters. According to equation (12),

is a stationary vector made up of the estimated residuals from the cointegration regression and differences in FDI attractiveness.

Also,

is the long-term covariance for this vector process, which can be decomposed as follows:

where

denotes the contemporaneous covariance and

is a weighted sum of autocovariances. FMOLS estimators are given by:

.(13)

Where

and

The inter-dimensional estimator is

where

is the FMOLS estimator applied to the ith panel member. The t-statistics are calculated as follows:

where

is the test statistic calculated for the ith panel member. Its asymptotic distribution is the normal distribution centered reduced.

From equation (13), the DOLS estimator is given by:

(14)

where

is a vector of 2(K+1)1.

and

. The DOLS inter-dimensional estimator can be constructed as follows

where

the DOLS estimator applied to the ith panel member is. The t-statistics are:

where

is the long-term variance of the residuals from the DOLS regression is

One of the disadvantages of using the DOLS test is that the degrees of freedom are lower in lags and leads

[35]

, so we note that the FMOLS test requires fewer hypotheses than the DOLS test and therefore tends to deliver more robust results.

4.4. Granger Panel Causality Tests

To test Granger causality in the long term, we use a two-stage process

[14]

. The first step is to estimate the long-run model for equation (11) to obtain the estimated residuals, ε_it(error correction term; henceforth referred to as ECT). The second step is to estimate the Granger causality model with a dynamic error-correction model. The error-correction model is specified as follows:

;(15)

;(16)

;(17)

(18)

where ∆ is the first difference operator; k is the number of delays determined by the Schwarz Information Criterion (SIC);

is the one-period lagged error correction term derived from the long-term cointegration relationship. According to

[14]

, this term materializes recall forces and makes it possible to determine the duration of shock absorption. To determine this duration,

gives the integer part of the number of years of absorption.

We can identify the sources of causality by testing the significance of the coefficients of the lagged dependent variables in equations (15) to (18). The specification of the equation allows us to test both short-term and long-term causality. For example, in the equation for CO₂ emissions (equation (15)), the short-term causality of GDP per capita, inward FDI and renewable energy consumption are tested respectively on the basis of H_A:

and

H_A:

and

and H_A:

and

5. Analysis of Results and Interpretation

In this section, we present the results and interpret the tests before presenting the results of the estimations.

5.1. Results and Interpretation of Unit Root and Panel Cointegration Tests

Before moving on to the interpretation of the cointegration tests, it is worthwhile presenting the unit root tests.

5.1.1. Panel Non-stationarity of the Variables Inward FDI and CO₂ Emissions

It should be noted that the tests for the series used are carried out in level and first difference.

Table 1. Results of panel unit root tests.

Variables		LLC	Breitung	IPS	ADF	PP	Hadri	Hadri_c
In level	lnCO₂	0,652 (0,521)	1,251 (0,854)	2,558* (0,000)	3,594 (0,100)	4,002 (0,336)	4,256 (0,554)	2,245* (0,000)
	LnPIB	0,022 (0,254)	3,701 (0,548)	7,878 (0,772)	-1,282 (0,811)	2,884 (0,996)	4,018 (0,621)	5,021 (0,362)
	lnIDE	2,112 (0,114)	1,111 (0,963)	-0,214 (0,963)	9,245 (0,845)	1,891 (0,145)	5,256* (0,000)	4,114* (0,000)
	LnER	0,215 (0,451)	1,895 (0,571)	0,287 (0,741)	4,253 (0,205)	7,241* (0,000)	8,562* (0,000)	3,587* (0,000)
In first difference	∆lnCO₂	-4,01* (0,000)	-0,559* (0,000)	-1,11* (0,007)	44,33* (0,007)	81,01* (0,007)	5,012* (0,000)	1,230* (0,000)
	∆lnPIB	1,14* (0,001)	0,021* (0,000)	-2,77* (0,000)	49,51* (0,001)	52,12* (0,000)	-1,210* (0,000)	1,237* (0,000)
	∆lnIDE	-0,25* (0,000)	7,21* (0,000)	1,457* (0,000)	10,87* (0,001)	7,96* (0,001)	-1,111 (0,201)	0,827* (0,001)
	∆lnER	-4,28* (0,000)	-5,777* (0,001)	-3,58* (0,003)	65,62* (0,000)	63,22* (0,000)	4,555* (0,000)	0,989* (0,008)

Note: * Indicates that the statistic is significant at the 1% level. ∆ is the first dissimilarity operator. Hadric denotes heteroskedastic coherent Z-stat, ADF and PP denote MW-ADF Fisher Chi-square and MW-PP Fisher Chi-square respectively. The p-values are in brackets.

Source: Author's estimates

For level variables, we find that the variables are non-stationary for the majority of tests. This result of non-stationarity between variables is also found by

[36, 48]

. Moving on to first-difference tests, we find that all series are stationary at the 1% significance level. For the purposes of this analysis, we assume a model with a constant and no trend for the individual statistics. As can be seen from Table 1, the null hypothesis of unit root can be rejected in several cases. The results suggest that the variables are non-stationary in level in the majority of tests. However, some variables are stationary in level, namely renewable energies with the PP and Hadri tests; CO₂ emissions with the IPS and Hadri tests; and FDI with the Hadri tests. On the other hand, all variables are stationary in first difference.

5.1.2. Panel Cointegration of Inward FDI and CO₂ Emissions Variables

The analysis of these results suggests that the null hypothesis of no cointegration cannot be rejected at the 5% significance level for all tests. Therefore, there may be a long-run relationship between inward FDI, CO₂ emissions, GDP per capita and renewable energy consumption in the case of our panel of countries.

Table 2. Results of Pedroni's panel cointegration test (

[39, 40]

Methods

Intra Dimension (panel statistics)

Inter Dimension (individual statistics)

Tests

Stat

Prob

Tests

Stat

Prob

[39]

V Panel statistics

0,589*

0,000

Group statistics ρ

1,735

0,987

ρ Panel statistics

-0,253

0,001

Group statistics pp

-1,589

0,893

PP Panel statistics

-1,411

0,002

Group statistics

-1,719

0,417

V Panel statistics ADF

-0,986

0,003

ADF

[40]

V Panel statistics

1,047

0,413

ρ Panel statistics

-0,868

0,087

PP Panel statistics v

-0,967

0,059

Panel statistics ADF

-0,270

0,234

Notes: The null hypothesis assumes that the variables are not cointegrated. Under the null hypothesis, all statistics are given as normal. ** indicates that the evaluated parameters are significant at the 5% error level.

Source: Author's estimates

The variables are cointegrated with the

[39]

test in the Intra dimension. However, the

[39]

tests in the Inter dimension and

[40]

in the Intra dimension show that the variables are not cointegrated. Therefore, we apply the panel cointegration test of

[27]

. Table 3 presents the results obtained from this test. It is clear that the null hypothesis of no cointegration is rejected at the 5% significance level.

Table 3.

[27]

panel cointegration test.

Model	ADF	P-value
lnCO₂, lnIDE, lnPIB, lnER	-1,205	0,031**

Notes: ADF is the residual based on the ADF statistic

[27]

. ** indicates that the evaluated parameters are significant at the 5% error level.

Source: Author's estimates

In the following work, we apply the Johansen cointegration test to test the cointegration between lnCO₂, lnIDE, lnPIB and lnER. However, before performing the Johansen panel cointegration tests, the optimal lag number is determined from the Akaike (AIC) and Schwarz (SIC) information criteria. The results obtained from the test are presented in Table 4. The decision rule states that if the test statistic is greater than the critical value at a given significance level (1%, 5% or 10%), the null hypothesis is rejected and vice versa.

Table 4. Results of the Johansen panel cointegration test.

Null hypothesis	Alternative hypothesis	Trace Test	Eigenvalue test
r = 0	r > 1	41,01* (0,001)	25,25* (0,001)
r ≤ 1	r > 2	8,12 (0,524)	11,21 (0,241)
r ≤ 2	r > 3	14,13 (0,784)	9,37 (0,793)

Note: * denotes that statistical tests are significant at the 1% level.

Source: Author's estimates

The results are quite conclusive: Fisher's tests (the trace and eigenvalue statistics) favor the presence of cointegration relationships between the study variables. Consequently, there is a cointegration relationship between the four variables of our model. Moreover, we can affirm that the variables (CO₂, FDI, GDP and ER) move together in the long term. The next step in this methodological approach is therefore to estimate this relationship through the FMOLS and DOLS.

5.2. Results of FMOLS and DOLS Estimations and Panel Granger Causality

In this section, we present the FMOLS and DOLS estimations before seeing the results of panel Granger causality between the different variables.

5.2.1. FMOLS and DOLS Estimations and Interpretations

The results obtained show that the estimated coefficients are significant at the 1%, 5% and 10% thresholds (Table 5). Furthermore, since the variables are expressed in logarithms, the coefficients can be interpreted in terms of elasticities.

Table 5. Long-run elasticities from FMOLS and DOLS estimators.

Dependent variable lnCO₂	FMOLS				DOLS
	Independent variables				Independent variables
	lnIDE	lnIDE, lnPIB	lnIDE, lnER	lnIDE, lnPIB, lnER	LnIDE	lnIDE, lnPIB	lnIDE, lnER	lnIDE, lnPIB, lnER
Inter results
C	0,210** (0,031)	0,725** (0,038)	0,001* (0,002)	0,035*** (0,072)	0,238** (0,042)	1,415** (0,024)	0,002* (0,001)	0,058*** (0,071)
With TD	0,471** (0,04)	0,998*** (0,08)	0,999** (0,04)	0,251* (0,007)	0,325*** (0,077)	1,830 (0,214)	0,101** (0,023)	0,011* (0,001)
Intra results
C	0,111** (0,041)	0,994** (0,049)	0,005* (0,001)	0,071*** (0,06)	0,283** (0,044)	0,871** (0,038)	0,004* (0,004)	0,041*** (0,07)
With TD	0,144*** (0,07)	0.771* (0,001)	0,006*** (0,08)	0,035 (0,874)	0,477*** (0,07)	0,610 (1,444)	0,007** (0,031)	0,071 (0,755)

Note: C and TD indicate dependent variable, constant and deterministic trend respectively.

P-values are in parentheses. T-statistics correspond to H0: βi =1. *, **, *** present the significance of the parameters at the threshold at the thresholds of 1%, 5% and 10% respectively.

Source: Author's estimates

Overall, the results of this study show that there is a relationship between CO₂ emissions and inward FDI in SSA. This result corroborates those of

[29]

. The impact of inward FDI on CO₂ emissions in SSA is positive and significant at the 5% level. In terms of elasticity, the results of this study indicate that a 1% increase in inward FDI in SSA increases CO₂ emissions by 0.210%. This therefore shows that the increase in inward FDI in SSA contributes to environmental degradation.. This impact remains positive, however, when the gross domestic product variable is taken into account.

With GDP per capita, a 1% increase in inward FDI leads to an increase in CO₂ emissions of 0.725%. This result confirms the “pollution haven” hypothesis in SSA, i.e. FDI is directed to SSA because of their weak environmental regulations. With renewable energy consumption, a 1% increase in inward FDI leads to a small increase in CO₂ emissions of 0.001%. Despite this positive effect, this result shows that the use of renewable energy leads to a decrease in the rate of CO₂ emissions and therefore tends to improve the quality of the environment in SSA.

[36]

in Turkey, and

[2]

in SSA countries found the same result. Taking into account GDP per capita and renewable energy consumption, a 1% increase in inward FDI leads to an increase in CO₂ emissions of 0.035%.

5.2.2. Short-run and Long-run Directional Causality and Speed of Adjustment

The results obtained from Granger causality are presented in Table 6. Since the variables in the model are cointegrated, the direction of causality can be divided into short-run causality and long-run causality. The optimal lag structure of a period is determined using the Akaike and Schwarz information criteria. In the short run, there is no causality between CO₂ emissions, inward FDI and renewable energy consumption. Furthermore, CO₂ emissions have a negative and statistically significant effect on GDP per capita in SSA. Indeed, a 1% increase in CO₂ emissions would lead to a decrease in real GDP per capita of 0.080%.

Table 6. Results of panel Granger Causality tests.

			Dependent variables				Sense of causality
			∆lnCO_2t	∆lnPIB_t	∆lnIDE_t	∆lnER_t	Sense of causality
Source of Causality (Independent variables)	CT	∆lnCO_2t	-	-0,08** (0,032)	0,024 (0,981)	-0,041 (0,781)	CO₂→PIB
		∆lnPIB_t	0,017* (0,001)	-	0,001** (0,032)	0,017 (0,741)	PIB→CO₂, IDE
		∆lnIDE_t	0,214* (0,002)	0,141* (0,001)	-	0,003* (0,002)	IDE→CO₂, PIB, ER
		∆lnERt	-0,008* (0,001)	0,049* (0,005)	0,013 (0,897)	-	ER→CO₂, PIB
	Test CJ (CT et LT)	∆lnCO_2t-1, ECT_t-1	-	4,014* (0,001)	2,014* (0,004)	3,893** (0,041)	CO₂→PIB, IDE et ER
		∆lnPIB_t-1, ECT_t-1	4,251* (0,004)	-	2,111* (0,001)	5,625*** (0,052)	PIB→CO₂, IDE et ER
		∆lnIDE_t-1, ECT_t-1	3,333* (0,007)	2,111* (0,002)	-	2,241** (0,024)	IDE→CO₂, PIB et ER
		∆lnER_t-1, ECT_t-1	2,582* (0,002)	4,257** (0,032)	2,258*** (0,08)	-	ER→CO₂, PIB et IDE

Note: X→Y means that variable X Granger-causes variable Y. CT, LT, and CJ denote short-run, long-run, and joint, respectively. * denotes that the estimated parameters are significant at the 1%, 5%, and 10% level, respectively.

Source: Author's estimates

Given that our variables of interest are CO₂ emissions and inward FDI in this article, therefore, we will not dwell at length on the other two variables (GDP per capita and renewable energy consumption).

Thus, our results highlight the fact that in the short term, CO₂ emissions have no effect on the attractiveness of FDI in SSA, and on the other hand, inward FDI impacts the environment in SSA through the increase in CO₂ emissions. This reflects a unidirectional causality from inward FDI to CO₂ emissions in SSA. This result corroborates that of

[29]

Furthermore, the long-run dynamics which is based on the statistical significance of the error correction terms indicates the speed of adjustment to the long-run equilibrium. This speed indicates that inward FDI, CO₂ emissions, GDP per capita and renewable energy consumption respond to deviations from the long-run equilibrium. Therefore, there is a bidirectional causality between inward FDI and CO₂ emissions in SSA in the long run. This result is identical to those found by

[36]

in Turkey;

[1]

in the Mediterranean region, and

[47]

in Vietnam. According to

[17]

, there is a positive long-term relationship between CO₂ emissions and FDI inflows. Following the results obtained from the cointegration tests, which indicated the presence of a long-run relationship between the variables, we adopt the error correction method by long-run adjustment between the variables.

Table 7. Results of the panel causality test (k = 1).

		Dependent variables
		∆lnCO_2t	∆lnPIB_t	∆lnIDE_t	∆lnER_t
Source of causality (independent variables)	∆lnCO_{2t -1}	-0,041 (0,414)	0,451 (0,541)	0,631 (0,451)	0,251 (0,123)
	∆lnPIB_{t – 1}	0,101*** (0,07)	-0,521 (0,691)	0,004** (0,03)	0,023* (0,003)
	∆lnIDE_{t -1}	0,204** (0,04)	0,127** (0,03)	-0,125 (0,631)	0,0023 (0,862)
	∆lnER_{t -1}	-0,147* (0,001)	0,025 (0,571)	0,011 (0,177)	-0,638 (0,477)
	ECT_t-1	-0,318* (0,001)	-0,396* (0,004)	-0,175* (0,001)	-0,155* (0,003)
	C	1,0021** (0,042)	1,314** (0,036)	0,993** (0,041)	0,931** (0,032)

Note: ECT denotes the error correction term. *, ** and *** denote that the estimated parameters are significant at the 1%, 5% and 10% levels respectively.

Source: Author's estimates

Thus, the ECT coefficient of -0.175 indicates that if the attractiveness of investments suffers a shock, it will take about five years and eight months for the shock to be absorbed and allow a return to the long-term equilibrium of incoming FDI in SSA. Similarly, the ECT coefficient of -0.396 indicates that if the economic system suffers a shock, it will take about two years and six months for the shock to be absorbed and allow a return to the long-term equilibrium of GDP per capita in SSA. And the ECT coefficient of -0.155 indicates that if the energy sector suffers a shock, it will take about six years and five months for the shock to be absorbed and for renewable energy consumption to return to the long-term equilibrium in SSA. And finally, an ECT of -0.318 indicates that if the environmental sector suffers a shock, it will take about three years and two months for the shock to be absorbed and for the CO₂ emission rate to return to long-term equilibrium.

6. Conclusion and Policy Recommendations

Moreover, this paper set itself the main objective of empirically assessing the nature of the relationship between inward FDI and CO₂ emissions in SSA. To this end, using a methodological approach based on recent econometric techniques, this paper determined the relationship between the different variables of the analysis model. Several tests were used, namely cointegration tests and stationarity tests. It emerged that the variables move together in the long term and that the increase in inward FDI in SSA contributes to environmental degradation.

This result, moreover, confirms the "pollution haven hypothesis" in SSA. By taking into account the control variables (gross domestic product and consumption of renewable energy), this result was nevertheless positive. However, taking into account the consumption of renewable energy reduces the positive effect of FDI on CO₂ emissions in SSA.

Overall, this shows that the use of renewable energies leads to a reduction in CO₂ emissions and consequently to an improvement in environmental quality in SSA. Applying panel causality tests, we find that if there is a shock to investment attractiveness, it will take around five years and eight months for the shock to be absorbed and for inward FDI in SSA to return to long-term equilibrium. On the other hand, if the environmental sector suffers a shock, it will take around three years and two months for the shock to be absorbed and the CO₂ emission rate to return to long-term equilibrium.

In order to reduce CO₂ emissions, SSA must implement policies and strategies that guarantee growth without abandoning the environment; put in place effective environmental rules to guide inward FDI; put in place strategies to exploit and enhance its energy potential.

Abbreviations

ADF	Augmented Dickey-Fuller
CO₂	Carbon Dioxide Emission
DOLS	Dynamic Ordinary Least Squares
ECT	Error Correction Term
EKC	Environmental Kuznets Curve
FDI	Foreign Direct Investment
FMOLS	Fully Modified Ordinary Least Squares
GDP	Gross Domestic Product
WDI	World Development Indicator
SSA	Sub-saharan Africa

Author Contributions

Mohamadou Oumarou: Conceptualization, Formal Analysis, Investigation, Methodology, Resources, Writing – original draft

Mohammadou Nourou: Data curation, Methodology, Supervision, Validation, Visualization

Ibrahim: Formal Analysis, Methodology, Supervision, Validation

Votsoma Philemon: Data curation, Funding acquisition, Methodology, Supervision, Validation

Conflicts of Interest

The authors declare no conflicts of interest.

References

[1]	Abdouli. M. and Omri A. “Exploring the Nexus Among FDI Inflows, Environmental Quality, Human Capital, and Economic Growth in the Mediterranean Region.” J Knowl Econ 12, 788-810 (2021). https://doi.org/10.1007/s13132-020-00641-5
[2]	Acheampong A. O., Samuel A. and Elliot B. «Do globalization and renewable energy contribute to carbon emissions mitigation in Sub-Saharan Africa? », Science of the Total Environment, 2019. https://doi.org/10.1016/j.scitotenv.2019.04.353
[3]	Andriamahery A., Danarson J. and Qamruzzaman Md. «Nexus between trade and environmental quality in sub-saharan Africa: Evidence from panel GMM», Frontiers in Environmental Science, 10. 10.3389/fenvs.2022.986429, 2022.
[4]	Awad A. and Warsam M. «Climate changes in Africa: does economic growth matter? A semi-parametric approach», International Journal of Energy Economic and Policy, 7, 1-8, 2017.
[5]	Balsalobre-Lorente D., Gokmenoglu K. K., Taspinar N. et al. “An approach to the pollution haven and pollution halo hypotheses in MINT countries”. Environ Sci Pollut Res 26, 23010–23026 (2019). https://doi.org/10.1007/s11356-019-05446-x
[6]	Bildirici M. and Gokmenoglu S. M. «The impact of terrorism and FDI on environmental pollution: evidence from Afghanistan, Iraq, Nigeria, Pakistan, Philippines, Syria, Somalia, Thailand and Yemen», Environmental Impact Assessment Review, 81, 106340, 2020. https://doi.org/10.1016/j.eiar.2019.106340
[7]	Bobbo A. «Volatilité de l’inflation, gouvernance et investissements directs étrangers entrants en Afrique sub-saharienne», African Development Review, 30, 86-99, 2018. https://doi.org/10.1111/1467-8268.12314
[8]	Chirambo D. «Towards the achievement of SDG 7 in sub-Saharan Africa: Creating synergies between Power Africa, Sustainable Energy for All and climate finance in-order to achieve universal energy access before 2030», Renewable and Sustainable Energy Reviews, 94, 600-608, 2018. https://doi.org/10.1016/j.rser.2018.06.025
[9]	Choi I. «Unit Root Tests for Panel Data», Journal of International Money and Finance, 20, 249-272, 2001. https://doi.org/10.1016/S0261-5606(00)00048-6
[10]	Cole M. A. and Neumayer E. “Examining the Impact of Demographic Factors on Air Pollution”. Population and Environment, 5–21 (2004). https://doi.org/10.1023/B:POEN.0000039950.85422.eb
[11]	CNUCED «Rapport sur le commerce et le développement 2019: Le financement d’une Nouvelle donne écologique mondiale», 2019.
[12]	Demena B. A. and Afesorgbor S. K. «The effect of FDI on environmental emissions: Evidence from a meta-analysis», Energy Policy, 138, 111192, 2020. https://doi.org/10.1016/j.enpol.2019.111192
[13]	Emrah K. and Aykut Ş. «The impact of foreign direct investment on CO₂ emissions in Turkey: new evidence from cointegration and bootstrap causality analysis», Environmental Science and Pollution Research International, 25, 790-804, 2018. https://doi.org/10.1007/s11356-017-0468-2
[14]	Engle R. F. and Granger C. W. J. «Cointegration and Error-Correction: Representation, Estimation and Testing», Econometrica, 64, 813-836, 1987. https://doi.org/10.2307/1913236
[15]	Eriandani R., Anam., Prastiwi D. and Triani A. N. N. «The Impact of Foreign Direct Investment on CO₂ Emissions in ASEAN Countries», International Journal of Energy Economics and Policy, 10, 584-592, 2020. https://doi.org/10.32479/ijeep.10230
[16]	Eskeland G. S. and Harrison A. E. «Moving to greener pastures? Multinationals and The pollution haven hypothesis», Journal of Development Economics, 70, 1-23, 2003. https://doi.org/10.1016/S0304-3878(02)00084-6
[17]	Essandoh O. K., Islam M. and Kakinaka M. «Linking international trade and foreign direct investment to CO₂ emissions: Any differences between developed and developing countries?», Science of The Total Environment, 712, 136437, 2020. https://doi.org/10.1016/j.scitotenv.2019.136437
[18]	Farooq F., Chaudhry S., Yusop Z. and Habibullah M. S. «How Do Globalization and Foreign Direct Investment Affect Environmental Quality in OIC Member Countries», Pakistan Journal of Commerce and Social Sciences, 14, 551-568, 2020.
[19]	Ganda F. «The influence of growth determinants on environmental quality in SubSaharan Africa states», Environment, Development and Sustainability. 23, 7117-7139, 2021. https://doi.org/10.1007/s10668-020-00907-7
[20]	Gharnit S., Bouzahzah M. and Ait Soussane J. «Impact of foreign direct investment on CO2 emissions in Morocco: An empirical investigation», Journal of Economic Benchmarks and Perspectives, 4, 213-226, 2020.
[21]	Grossman G. M. and Krueger A. B. «Environmental Impacts of North American Free Trade Agreement», Discussion Papers in Economics, No. 158, Woodrow Wilson School of Public and International Affairs, Princeton, NJ, 1991. https://doi.org/10.3386/w3914
[22]	Hadri K. «Testing for Stationarity in Heterogeneous Panel Data», Econometric Journal, 3, 148-161, 2000. https://doi.org/10.1111/1368-423X.00043
[23]	Hille E., Shahbaz M. and Mose I. «The impact of FDI on regional air pollution in the Republic of Korea: a way ahead to achieve the green growth strategy?», Energy Economics, 81, 308-326, 2019. https://doi.org/10.1016/j.eneco.2019.04.004
[24]	Im KS., Pesaran MH. And Shin Y. «Testing for unit roots in heterogeneous panels», Journal of econometrics, 115, 53-74, 2003. https://doi.org/10.1016/S0304-4076(03)00092-7
[25]	Johansen S. «Statistical analysis of cointegration vectors», Journal of economic dynamics and control, 12, 231-254, 1988. https://doi.org/10.1016/0165-1889(88)90041-3
[26]	Jugurnath B., Chuckun N. and Fauzel S. «Foreign Direct Investment & Economic Growth in Sub-Saharan Africa: An Empirical Study», Theoretical Economics Letters, 6, 798-807, 2016. https://doi.org/10.4236/tel.2016.64084
[27]	Kao C. «Spurious Regression and Residual-Based Tests for Cointegration in Panel Data», Journal of Economic, 90, 1-44, 1999. https://doi.org/10.1016/S0304-4076(98)00023-2
[28]	Khan M. A. and Ozturk I. «Examining foreign direct investment and environmental pollution linkage in Asia», Environmental Science and Pollution Research, 27, 7244-7255, 2020. https://doi.org/10.1007/s11356-019-07387-x
[29]	Kivyiro P. and Arminen H. «Carbon dioxide emissions, energy consumption, economic growth, and foreign direct investment: Causality analysis for Sub-Saharan Africa», Energy, 74, 595-606, 2014. https://doi.org/10.1016/j.energy.2014.07.025
[30]	Kostakis I., Lolos S. and Sardianou E. «Foreign direct investment and environmental degradation: Further evidence from Brazil and Singapore», Journal of Environmental Management and Tourism, 8, 45-59, 2017. https://doi.org/10.14505//jemt.v8.1(17).04
[31]	Kriaa I., Ettbib R. and Akrout Z. «Foreign direct investment and industrialization of Africa», International Journal of Innovation and Applied Studies, 21, 477-491, 2017.
[32]	Kwiatkowski D., Phillips C. B., Schmidt P. and Shin Y. «Testing the null hypothesis of stationarity against the alternative of a unit root: How sure are we that economic time series have a unit root? », Journal of Econometrics, 54, 159-178, 1992. https://doi.org/10.1016/0304-4076(92)90104-Y
[33]	Levin A., Lin C-F. and Chu C-SJ. «Unit root tests in panel data: asymptotic and finitesample Properties», Journal of econometrics, 108, 1-24, 2002. https://doi.org/10.1016/S0304-4076(01)00098-7
[34]	Maddala G. S. and Wu S. «A Comparative Study of Unit Roots with Panel Data and a New Simple Test», Oxford Bulletin of Economics and Statistics, 61, 631-651, 1999. https://doi.org/10.1111/1468-0084.0610s1631
[35]	Maeso-Fernandez F., Osbat C. and Schnatz B. «Towards the estimation of equilibrium exchange rates for transition economics: Methodological issues and a panel cointegration perspective», Journal of Comparative Economics, 34, 499-517, 2006. https://www.ecb.europa.eu//pub/pdf/scpwps/ecbwp353.pdf
[36]	Mert M. and Caglar A. E. «Testing pollution haven and pollution halo hypotheses for Turkey: a new perspective», Environmental Science and Pollution Research, 27, 32933-32943, 2020. https://doi.org/10.1007/s11356-020-09469-7
[37]	Ngwenya B., Oosthuizen J., Cross M., Frimpong K. and Chaibva C. N. «A review of heat stress policies in the context of climate change and its impacts on outdoor workers: evidence from Zimbabwe», International Journal of social Ecology and Sustainable Development, 9, 1-11, 2018. https://doi.org/10.4018/IJSESD.2018010101
[38]	Ongo Nkoa B. E. and Song S. J. «Africa’s fragility: An explanation by foreign direct investment?», Developing Worlds, 193, 47-68, 2021. https://doi.org/10.3917/med.193.0051
[39]	Pedroni P. «Critical Values for Cointegration Tests in Heterogeneous Panels with Multiple Regressors», Oxford Bulletin of Economics and Statistics, 61, 653-670, 1999. https://doi.org/10.1111/1468-0084.0610s1653
[40]	Pedroni P. «Panel Cointegration: Asymptotic and Finite Sample Properties of Pooled Time Series Tests with an Application to the PPP Hypothesis: New Results», Econometric Theory, 20, 597-625, 2004. https://doi.org/10.1017/S0266466604203073
[41]	Rahman Z. U., Chongbo W. and Ahmad M. «An (A)symmetric Analysis of the Pollution haven Hypothesis in the Context of Pakistan: A Non-linear Approach», Carbon Management, 10, 227-239, 2019. https://doi.org/10.1080/17583004.2019.1577179
[42]	Sabir S., Qayyum U. and Majeed T. «FDI and environmental degradation: the role of political institutions in South Asian countries», Environmental Science and Pollution Reserach, 27, 32544-32553, 2020. https://doi.org/10.1007/s11356-020-09464-y
[43]	Shahbaz M., Nasir M. A. and Roubaud D. «Environmental degradation in France: The effects of FDI, financial development, and energy innovations», Energy Economics, 74, 843-857, 2018. https://doi.org/10.1016/j.eneco.2018.07.020
[44]	Shahbaz M., Samia N. and Faisal A. «Does foreign direct investment impede environmental quality in high-, middle-, and low-income countries?», Energy Economics, 51, 275-287, 2015.
[45]	Sotamenou J. et Nguepdjio C. N. «Energy consumption, economic growth and CO2 emissions in Cameroon: a causal analysis», African Intergration and Development Review, 11, 82-100, 2019.
[46]	Syed Q. R., Bhowmik R., Adedoyin F. F., Alola A. A. and Khalid N. «Do economic Policy Uncertainty and Geopolitical Risk Surge CO₂ Emissions ? New Insights from Panel Quantile Regression Approach», Environmental Science and Pollution Research International, 29, 27845-27861, 2022. https://doi.org/10.1007/s11356-021-17707-9
[47]	Tang C. F. and Tan B. W. «The impact of energy consumption, income and foreign direct investment on carbon dioxide emissions in Vietnam», Energy, 79, 447-454, 2015. https://doi.org/10.1016/j.energy.2014.11.033
[48]	Terzi H. and Pata K. «Is the pollution haven hypothesis (PHH) valid for Turkey?», Panoeconomicus, 67, 93-109, 2020. 67(1): 93-109. https://doi.org/10.2298/PAN161229016T
[49]	To H. A., Dao Ha T-T., Nguyen H. M. and Vo Hong D. «The Impact of Foreign Direct Investment on Environment Degradation: Evidence from Emerging Markets in Asia», International Journal of Environmental Research and Public Health, 16, 1636, 2019. https://doi.org/10.3390/ijerph16091636
[50]	World Bank «World development indicators». World Bank, Washington DC, 2023. http://data.worldbank.org/data-catalog
[51]	Zafar M. W., Qin Q., Malik M. N. and Zaidi S. A. H. «Foreign direct investment and education as determinants of environmental quality: The importance of post Paris Agreement (COP21)», Journal of Environmental Management, 270, 110827, 2020. https://doi.org/10.1016/j.jenvman.2020.110827

Cite This Article

Plain Text BibTeX RIS

APA Style

Oumarou, M., Nourou, M., Ibrahim, Philemon, V. (2024). Foreign Direct Investment and CO2 Emissions in Sub-Saharan Africa: A Heterogeneous Panel Causality Analysis. International Journal of Economy, Energy and Environment, 9(5), 105-118. https://doi.org/10.11648/j.ijeee.20240905.11

Copy | Download

ACS Style

Oumarou, M.; Nourou, M.; Ibrahim; Philemon, V. Foreign Direct Investment and CO2 Emissions in Sub-Saharan Africa: A Heterogeneous Panel Causality Analysis. Int. J. Econ. Energy Environ. 2024, 9(5), 105-118. doi: 10.11648/j.ijeee.20240905.11

Copy | Download

AMA Style

Oumarou M, Nourou M, Ibrahim, Philemon V. Foreign Direct Investment and CO2 Emissions in Sub-Saharan Africa: A Heterogeneous Panel Causality Analysis. Int J Econ Energy Environ. 2024;9(5):105-118. doi: 10.11648/j.ijeee.20240905.11

Copy | Download

@article{10.11648/j.ijeee.20240905.11,
  author = {Mohamadou Oumarou and Mohammadou Nourou and Ibrahim and Votsoma Philemon},
  title = {Foreign Direct Investment and CO2 Emissions in Sub-Saharan Africa: A Heterogeneous Panel Causality Analysis


},
  journal = {International Journal of Economy, Energy and Environment},
  volume = {9},
  number = {5},
  pages = {105-118},
  doi = {10.11648/j.ijeee.20240905.11},
  url = {https://doi.org/10.11648/j.ijeee.20240905.11},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ijeee.20240905.11},
  abstract = {Following the density of the literature and the consensus in empirical studies, the aim of this article is to examine the nature of the relationship between foreign direct investment (FDI) and carbon dioxide (CO2) emissions in sub-Saharan Africa (SSA). To this end, the methodological strategy employed is based not only on a theoretically sound multivariate framework, but also on recent developments in panel data econometrics, namely fully modified ordinary least squares (FMOLS) estimators, dynamic ordinary least squares (DOLS) estimators and the vector error correction model. In addition, the stationarity properties of the panel variables are examined, and the panel cointegration technique is used to test cointegrating relationships in the series of variables. The panel is composed of 38 SSA countries over the period 2000-2022. The main results show that in SSA: the variables move together in the long term. A 1% increase in inward FDI increases CO2 emissions by 0.210%. This result suggests that FDI has flowed to SSA because of its weak environmental regulations, thus verifying the pollution haven hypothesis. In the long term, there is a bidirectional relationship between inward FDI and CO2 emissions. In all the models used, renewable energy consumption reduces CO2 emissions. Therefore, SSA needs to put in place effective environmental rules to better guide FDI; put in place strategies to harness and add value to its energy sector, implement policies and strategies that ensure FDI attractiveness without abandoning the environment.
},
 year = {2024}
}

Copy | Download

TY  - JOUR
T1  - Foreign Direct Investment and CO2 Emissions in Sub-Saharan Africa: A Heterogeneous Panel Causality Analysis



AU  - Mohamadou Oumarou
AU  - Mohammadou Nourou
AU  - Ibrahim
AU  - Votsoma Philemon
Y1  - 2024/09/23
PY  - 2024
N1  - https://doi.org/10.11648/j.ijeee.20240905.11
DO  - 10.11648/j.ijeee.20240905.11
T2  - International Journal of Economy, Energy and Environment
JF  - International Journal of Economy, Energy and Environment
JO  - International Journal of Economy, Energy and Environment
SP  - 105
EP  - 118
PB  - Science Publishing Group
SN  - 2575-5021
UR  - https://doi.org/10.11648/j.ijeee.20240905.11
AB  - Following the density of the literature and the consensus in empirical studies, the aim of this article is to examine the nature of the relationship between foreign direct investment (FDI) and carbon dioxide (CO2) emissions in sub-Saharan Africa (SSA). To this end, the methodological strategy employed is based not only on a theoretically sound multivariate framework, but also on recent developments in panel data econometrics, namely fully modified ordinary least squares (FMOLS) estimators, dynamic ordinary least squares (DOLS) estimators and the vector error correction model. In addition, the stationarity properties of the panel variables are examined, and the panel cointegration technique is used to test cointegrating relationships in the series of variables. The panel is composed of 38 SSA countries over the period 2000-2022. The main results show that in SSA: the variables move together in the long term. A 1% increase in inward FDI increases CO2 emissions by 0.210%. This result suggests that FDI has flowed to SSA because of its weak environmental regulations, thus verifying the pollution haven hypothesis. In the long term, there is a bidirectional relationship between inward FDI and CO2 emissions. In all the models used, renewable energy consumption reduces CO2 emissions. Therefore, SSA needs to put in place effective environmental rules to better guide FDI; put in place strategies to harness and add value to its energy sector, implement policies and strategies that ensure FDI attractiveness without abandoning the environment.

VL  - 9
IS  - 5
ER  -

Copy | Download

Author Information

Mohamadou Oumarou

Department of Public Economics, Faculty of Economics and Management, University of Garoua, Garoua, Cameroon

Contact Email
Mohammadou Nourou

Department of Quantitative Methods, Faculty of Economics and Management, University of Garoua, Garoua, Cameroon

Contact Email

http://orcid.org/0000-0001-8440-6988
Ibrahim

Department of Public Economics, Faculty of Economics and Management, University of Garoua, Garoua, Cameroon
Votsoma Philemon

Banking and Monetary Economics Department, Faculty of Economics and Management, University of Garoua, Garoua, Cameroon

Contact Email

Download PDF

Plain Text BibTeX RIS

APA Style

Oumarou, M., Nourou, M., Ibrahim, Philemon, V. (2024). Foreign Direct Investment and CO2 Emissions in Sub-Saharan Africa: A Heterogeneous Panel Causality Analysis. International Journal of Economy, Energy and Environment, 9(5), 105-118. https://doi.org/10.11648/j.ijeee.20240905.11

Copy | Download

ACS Style

Oumarou, M.; Nourou, M.; Ibrahim; Philemon, V. Foreign Direct Investment and CO2 Emissions in Sub-Saharan Africa: A Heterogeneous Panel Causality Analysis. Int. J. Econ. Energy Environ. 2024, 9(5), 105-118. doi: 10.11648/j.ijeee.20240905.11

Copy | Download

AMA Style

Oumarou M, Nourou M, Ibrahim, Philemon V. Foreign Direct Investment and CO2 Emissions in Sub-Saharan Africa: A Heterogeneous Panel Causality Analysis. Int J Econ Energy Environ. 2024;9(5):105-118. doi: 10.11648/j.ijeee.20240905.11

Copy | Download

@article{10.11648/j.ijeee.20240905.11,
  author = {Mohamadou Oumarou and Mohammadou Nourou and Ibrahim and Votsoma Philemon},
  title = {Foreign Direct Investment and CO2 Emissions in Sub-Saharan Africa: A Heterogeneous Panel Causality Analysis


},
  journal = {International Journal of Economy, Energy and Environment},
  volume = {9},
  number = {5},
  pages = {105-118},
  doi = {10.11648/j.ijeee.20240905.11},
  url = {https://doi.org/10.11648/j.ijeee.20240905.11},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.ijeee.20240905.11},
  abstract = {Following the density of the literature and the consensus in empirical studies, the aim of this article is to examine the nature of the relationship between foreign direct investment (FDI) and carbon dioxide (CO2) emissions in sub-Saharan Africa (SSA). To this end, the methodological strategy employed is based not only on a theoretically sound multivariate framework, but also on recent developments in panel data econometrics, namely fully modified ordinary least squares (FMOLS) estimators, dynamic ordinary least squares (DOLS) estimators and the vector error correction model. In addition, the stationarity properties of the panel variables are examined, and the panel cointegration technique is used to test cointegrating relationships in the series of variables. The panel is composed of 38 SSA countries over the period 2000-2022. The main results show that in SSA: the variables move together in the long term. A 1% increase in inward FDI increases CO2 emissions by 0.210%. This result suggests that FDI has flowed to SSA because of its weak environmental regulations, thus verifying the pollution haven hypothesis. In the long term, there is a bidirectional relationship between inward FDI and CO2 emissions. In all the models used, renewable energy consumption reduces CO2 emissions. Therefore, SSA needs to put in place effective environmental rules to better guide FDI; put in place strategies to harness and add value to its energy sector, implement policies and strategies that ensure FDI attractiveness without abandoning the environment.
},
 year = {2024}
}

Copy | Download

TY  - JOUR
T1  - Foreign Direct Investment and CO2 Emissions in Sub-Saharan Africa: A Heterogeneous Panel Causality Analysis



AU  - Mohamadou Oumarou
AU  - Mohammadou Nourou
AU  - Ibrahim
AU  - Votsoma Philemon
Y1  - 2024/09/23
PY  - 2024
N1  - https://doi.org/10.11648/j.ijeee.20240905.11
DO  - 10.11648/j.ijeee.20240905.11
T2  - International Journal of Economy, Energy and Environment
JF  - International Journal of Economy, Energy and Environment
JO  - International Journal of Economy, Energy and Environment
SP  - 105
EP  - 118
PB  - Science Publishing Group
SN  - 2575-5021
UR  - https://doi.org/10.11648/j.ijeee.20240905.11
AB  - Following the density of the literature and the consensus in empirical studies, the aim of this article is to examine the nature of the relationship between foreign direct investment (FDI) and carbon dioxide (CO2) emissions in sub-Saharan Africa (SSA). To this end, the methodological strategy employed is based not only on a theoretically sound multivariate framework, but also on recent developments in panel data econometrics, namely fully modified ordinary least squares (FMOLS) estimators, dynamic ordinary least squares (DOLS) estimators and the vector error correction model. In addition, the stationarity properties of the panel variables are examined, and the panel cointegration technique is used to test cointegrating relationships in the series of variables. The panel is composed of 38 SSA countries over the period 2000-2022. The main results show that in SSA: the variables move together in the long term. A 1% increase in inward FDI increases CO2 emissions by 0.210%. This result suggests that FDI has flowed to SSA because of its weak environmental regulations, thus verifying the pollution haven hypothesis. In the long term, there is a bidirectional relationship between inward FDI and CO2 emissions. In all the models used, renewable energy consumption reduces CO2 emissions. Therefore, SSA needs to put in place effective environmental rules to better guide FDI; put in place strategies to harness and add value to its energy sector, implement policies and strategies that ensure FDI attractiveness without abandoning the environment.

VL  - 9
IS  - 5
ER  -

Copy | Download