Sage Journals: Discover world-class research

Abstract

In order to stimulate economic growth, most African countries have engaged the industrial sector to reduce their dependence on importation. This prompted the need for this study to investigate the impact of energy consumption on industrial performance. The study utilized panel data across 32 Sub Saharan African countries from 2002 to 2019. The Fully Modified Ordinary Least Squares (FMOLS), Pooled Mean Group (PMG) and Generalized Linear Model (GLM) techniques were employed to evaluate the long-term impacts. To ensure an in-depth study, the paper focused on two groups of countries; the lower-middle-income and low income countries and merged them into a single group. The major findings from the study is the positive associations between energy consumption, renewable energy consumption and fossil fuel consumption on industrial sector performance. This means that increased energy consumption enhances the productive capacities of industries, which steers economic growth across the region. Further findings revealed a negative association between foreign direct investment and industrial sector performance. The study recommended the usage of clean energy since increased energy consumption in form of fossil fuels influences climate change.

Plain language summary

Energy consumption and industrial sector performance

This study examined the impact of energy consumption on industrial performance in 32 Sub Saharan African countries from 2002 to 2019. The Fully Modified Ordinary Least Squares (FMOLS), Pooled Mean Group (PMG) and Generalized Linear Model (GLM) techniques were adopted to evaluate the long-term impacts. Empirical investigation showed a positive association between energy consumption, renewable energy consumption and fossil fuel consumption on industrial sector performance. Therefore, the study recommended the usage of clean energy since increased energy consumption in form of fossil fuels influences climate change.

Keywords

agricultural and natural resource economics economic science social sciences industrial organization economic systems energy consumption industrial output SSA

Introduction

The industrial sector is one among the sectors (alongside agricultural and service sector) that make up a country’s economy. In every economy, the industrial sector mostly transforms raw materials to finished or semi-finished products through the construction and manufacturing industries, and sell to the end user or distributed for further processing. Economists opine that the manufacturing industry contributes more to an economy as compared to the service sector. As such, countries that export manufactured products usually record massive marginal GDP growth that leads to economic development, translating to a better quality of life for its citizens. This facilitates inclusive growth by reducing unemployment and enhancing agricultural productivity (Ologbenla, 2020). If agricultural outputs must be utilized to sustain growth in the region, the industrial sector must be robust to engage the outputs. According to Ranis (1973), the success of this depends on how business climate is improved, as well as enhanced technical know-how and exploitation of recent technologies.

Therefore, for the industrial sector to meet up with the growing demands to function to its full capacity due to population growth and urbanization, there must be increased energy consumption. Industrial activities consume a substantial share of energy, accounting for 25% of total energy consumption at the global level in 2017 (Bataille & Melton, 2017). It is likely impossible to produce, deliver, or use mainstream commodities without consuming energy. Hence, Yildirim (2017) observed that insufficient energy would negatively impact the performance of different sectors of the economy such as transport and a country’s social life. The World Bank defines energy consumption as the use of primary energy before transformation to other end-use fuels, which include the production of electric power. It is imperative to note that the World Bank’s definition of energy consumption encompasses all primary energy sources, including fossil fuels (such as coal, oil, and natural gas), renewables (such as hydro, wind, solar, and biomass), and nuclear energy. In other words, energy consumption is the total amount of energy required for a given process. Industrial energy consumption contributes to the emission of greenhouse gases by relying mainly on energy produced from fossil fuels (Gozgor et al., 2018). While considerable progress is being recorded by shifting to renewable energy sources, industrial development and growth have increased energy consumption and have further increased emission levels. Improving the industrial sector’s energy efficiency (EE) is crucial for decoupling economic growth from the negative environmental and climate impacts of industrial development (United Nations Industrial Development Organization, 2020).

There is no denying that industries play a pivotal role in the growth and development of a nation. This is because, the industrial sector is responsible for transforming raw materials into finished products. These activities are growth-enhancing as they generate employment opportunities and stimulate other sectors of the economy, such as the agricultural and mining sectors. To further support this claim, Eke et al. (2018) opined that the development of industrial activities is considered a catalyst for rapid growth. The study revealed that the quality of domestic institutions and energy consumption mix are the two major determinants of industrial performance (Eke et al., 2018). Furthermore, Oyerinde (2019), noted that industrialization is an essential element that bridges the gap between underdeveloped and developed nations. However, the process of industrialization has largely been driven by the consumption of fossil sources of energy such as petrol and diesel in most parts of Africa. This is largely due to the inadequate supply of electricity from the national grid of the countries. With this, African generator market is expected to grow due to the increasing demand for uninterrupted and reliable power supply and rise in the activities of the industrial sector.

There is no doubt that energy is critical in every sector of the economy. Ekong et al. (2021) affirmed this and noted that for sectoral growth and development, energy is imperative. At the heart of achieving sustainable growth and development in an economy is energy consumption and industrial development. This explains why the sustainable development goal 7 is dedicated to ensuring affordable and reliable modern energy for all. Consequently, in the industrial sector, the role of energy cannot be overemphasized. The industrial sector is one of the highest energy consuming sectors of an economy. This is due to its direct relationship with economic growth (Abid & Mraihi, 2015; Dash, 2016; Ziramba, 2009). According to Qazi et al. (2012), the efficient utilization of energy resources and its availability together with inputs of labor and capital are the major determinants of industrial sector output. Therefore, industries depend on energy as a source of inputs for their production activities. The productive capacities of these industries depend on the level of infrastructural investment. Studies such as Azolibe and Okonkwo (2020) found that due to poor infrastructure in SSA, there exist a low level of industrial productivity. This is a major challenge which has limited its potentials in the region. Most developing countries depend on energy imports such as crude oil, natural gas, and coal to meet the increasing demand for industrial energy inputs, transportation, or electrical generation (Polat, 2021). As such, Africa is not an exception, as reflected in the figure below.

Figure 1 below reports the primary energy consumption in Africa. It indicates that the level of energy consumed in Africa between 2002 and 2019 has been on the constant rise. This further amplifies the importance of energy in the usefulness of industrial performance. Energy consumption is undoubtedly pivotal to both sectoral and economy-wide performance, growth and development as it plays multi-dimensional roles in the development process at different levels and in different ways (Iwayemi, 1983, 1993). First, energy and its consumption has been viewed as a basic input (factor of production) by Stern (2019) in the production process. Second, economic agents (the decision-making units) in the economy need energy in order to fully maximize their utility and productivity. Third, arising from the above premise, it means that energy is core to output generation in every econom. According to Ekong et al. (2021), energy availability determines the production capacity, such as; volume of output per time, quantity and quality of products and finally, future plans for expansion. Therefore, access to energy and its availability reduces the marginal cost of production for firms. The energy sector of the economy is made up of electricity sub-sector, petroleum subsector, biofuel sub-sector, coal sub-sector among other sub-sectors.

Figure 1.

Primary energy consumption per capita (kWh/person).

Researchers have advocated for industrialization to solve the many challenges facing the region. For instance, SSA is experiencing rapid population increase than any other region in the world and only massive industrialization effort will be able to provide the needed job opportunities for the teeming young population (Morris & Fessehaie, 2014). The state of the industrial sector in the region is weak and underperforming and this is shown by the increased imports of several commodities including household items. These imports would turn to exports only if the underperforming industrial sector is galvanized to play its essential roles on unemployment reduction, poverty mitigation, trade promotion, exchange of goods, and services and economic growth etc. (Maroof et al., 2018). Therefore, it is imperative to understand the extent to which energy consumption drives industrial performance in SSA.

It is against this backdrop that the objectives of this paper aim to explore the relationship between energy consumption and industrial sector performance, as well as check for their causalities across different groups of countries in SSA. This study enriches existing literature in many ways. To begin with, unlike previous studies, sample of 32 SSA countries are carefully considered, and it comprises of low-income and lower-middle-income (see Appendix Table A2 for country classifications). This is necessary not only to understand their economic trends but also to better understand the socioeconomic status of these countries, unlike previous studies that focused strictly on regions. Unlike most empirical literatures in the subject that proxied energy consumption with electricity consumption, this study utilizes primary energy consumption for a more holistic coverage. In addition to these, the study adds to the scarce literatures to have recently explored energy consumption and industrial sector performance in SSA. Furthermore, the study includes CO₂ emission to report the extent to which fossil fuel consumption impacts the sectoral performance. Consequently, the study employs data from 2002 to 2019 and adopts the Fully Modified Ordinary Least Squares (FMOLS) method and the Dumitrescu and Hurling causality test to estimate the long-run association of the variables as well as their respective causalities. The study further proceeds to answer the question of whether energy consumption, renewable energy consumption and fossil fuel consumption impacts industrial performance.

Literature Review

Extant literature have explored the nexus between energy consumption and industrial performance. From the review, we discovered that, electricity consumption was used to proxy energy consumption in most of the studies. It is worthy to note that electricity consumption is only a part of energy consumption. Therefore, using it as a proxy may not give a broader coverage of energy consumption. More so, none of these studies presented their research as this study did in terms of grouping the countries by their income levels. This is necessary for comparative analysis and also to provide a better understanding of the economic conditions of the region. For example, Abokyi et al. (2018) researched on electricity consumption and industrial growth in the Ghanaian economy between 1971 and 2014. The study utilized the autoregressive distributed lag (ARDL) model and found that electricity consumption impacts negatively on industrial growth in both the long-run and the short-run. Results show the presence of co-integration and unidirectional causality from the consumption of electricity to industrial growth, supporting the growth hypothesis in Ghana. Mawejje and Mawejje (2016) examined the causal relationship between electricity consumption and sectoral output growth in Uganda from 2005 to 2015. The study utilized the vector error correction modeling (VECM) framework. From the macro level, the result shows the presence of causality running from electricity consumption to GDP. On the sectoral level, long-run causality runs from electricity consumption to industry, that is, indicating growth hypothesis for the sector, short-run causality from the services sector to electricity consumption, and no causality for agriculture. Qazi et al. (2012) further examined the relationship between disaggregated energy consumption and industrial output in Pakistan. Empirical findings using the Johansen cointegration methodology, showed that the long run coefficients of disaggregate energy consumption are significant and positively related to industrial output. Focusing also in Pakistan, Liew et al. (2012) explored the interdependence between energy consumption and sectoral output for the period 1980 to 2007 using co-integration and Granger causality tests. The findings show no evidence of a long-run relationship between energy consumption and industrial output. Bi-directional relationship was reported between energy consumption and industrial output. Similarly, Nwosa and Akinbobola (2012) also reported a bi-directional causality between energy consumption and sectoral output in Nigeria using vector autoregression. On the contrary, Chebbi and Boujelbene (2008) utilized vector error correction model and reported the existence of a uni-directional relationship between energy consumption and sectoral output in Tunisia.

Abid and Mraihi (2015) in a country specific study, investigated the relationship between energy consumption and industrial production in Tunisia. The study showed the existence of unidirectional causality from total energy consumption to Industrial added value (IAV) in the short run and no long run causality. At the disaggregated level, there is absence of causality between oil consumption and IAV both in the short and long run. Further findings found that bidirectional causality exists between industrial electricity consumption and IAV in the short and long run. Similarly, Rahman and Kashem (2017) employed ARDL cointegration and Granger causality analysis to explore industrial growth in Bangladesh. The Granger causality test revealed the existence of a unidirectional causality running from industrial development to energy consumption and energy consumption to CO2 emissions. Suggesting that industrial development influences the increases in energy consumption. Abokyi et al. (2018) using the ARDL bounds test found evidence of long-run relationship between consumption of electricity and industrial growth in Ghana. Also, both in the long run and short run, electricity consumption has a negative relationship with industrial growth. The Toda-Yamamoto modified granger causality showed a one-way causal relationship from electricity consumption to industrial growth.

Omosebi et al. (2019) explored the linkage between disaggregated energy consumption and industrial output in Nigeria. The study showed that there exists a positive relationship between industrial output, premium motor spirit, diesel, coal, and human capital. This implies that the variables contribute significantly to industrial output. Also, the study showed that consumption of gas, electricity, kerosene and capital stock contributed significantly to decrease in industrial output. This is due to their inadequate and irregular supply in Nigeria. In the same vein, Abbasi et al. (2021) investigated the impact of energy consumption on industrial sectoral growth between 1970 and 2018. Electricity usage was the favored proxy for energy consumption. Utilizing the Vector Error Correction Mechanism and Variance Decomposition, the findings showed that energy consumption positively impacts industrial output in Pakistan. The authors called on the government to increase their investment level in power projects as it will help maintain energy demands and supply, which is also beneficial to the overall economy at large. Further empirical studies by Zehra et al. (2020) examined the productive performance of the Pakistani industrial sector. They employed the error correction model data set of renewable and non-renewable energy products between 1980 and 2016. Their findings also reported a strong positive association between longrun energy consumption and industrial sector output.

Only few cross-country studies between energy consumption and industrial output have been explored. For example, Danmaraya and Hassan (2016) utilized panel data to examine the effect of energy consumption, capital and labor on manufacturing industry performance in seven low-income sub Saharan African countries. The study adopted the fully modified Ordinary Least square (FMOLS) procedure. Result of the study showed a positive significant relationship between manufacturing industry performance and energy consumption as well as between manufacturing industry performance and capital. This shows that energy consumption stimulates manufacturing performance.

Methodology

The model in the equation below was estimated using panel data for a sample of 32 Sub Saharan African countries. Two groups of countries (lower-middle-income and low income) based on the classifications of United Nations (2022) were merged together in the study. These countries have been classified by their level of development as measured by per capita gross national income—GNI—(United Nations, 2022). To maintain compatibility with similar classifications used elsewhere, the threshold levels of GNI per capita are those established by the World Bank. Countries with less than $1,046 GNI per capita are classified as low-income countries, those with between $1,046 and $4,095 as lower-middle income countries, those with between $4,096 and $12,695 as upper-middle-income countries. These countries were chosen for the study based on data availability.

Data Description

This study examines the impact of energy consumption on industrial sector performance from 2002 to 2019. To further proceed with the analysis, two control variables including official exchange rates and foreign direct investment (FDI) were included in the model. The choice of these variables as presented in Table 1 below are for the following reasons. Firstly, as explained by WDI (2023) dataset, the annual growth rate for industrial (including construction) value added is based on constant local currency. These aggregates are based on constant 2015 prices, expressed in U.S. dollars. Industry corresponds to International Standard Industrial Classification (ISIC) divisions and includes manufacturing. It comprises value added in mining, manufacturing (also reported as a separate subgroup), construction, electricity, water, and gas. Value added is the net output of a sector after adding up all outputs and subtracting intermediate inputs. It is calculated without making deductions for depreciation of fabricated assets or depletion and degradation of natural resources. The origin of value added is determined by the ISIC. In this study, industry (including construction), value added (annual % growth) is one of the two core variables. This is guided by the fact that growth in industrial sector is interpreted as increase in industrialization. Thus, this variable captures the growth of this sector’s performance on an annual basis to suit the objectives of this study.

Table 1.

Data Source.

Variables	Data sources
Industry (including construction), value added (annual % growth)	World Development Indicators, World Bank
Primary energy consumption per capita (kWh/person)	BP Statistical Review of World Energy
Renewable energy consumption (% of total final energy consumption)	World Development Indicators, World Bank
CO₂ emissions kiloton (kt)	World Development Indicators, World Bank
Foreign direct investment (FDI), net inflows (BoP, current US$)	World Development Indicators, World Bank
Official exchange rate (LCU per US$, period average)	World Development Indicators, World Bank

Secondly, primary energy consumption per capita (kWh/person) is the second core variable in this research. BP Statistical Review of World Energy defines primary energy as the energy available as resources—such as the fuels burnt in power plants—before it has been transformed. This relates to the coal before it has been burned, the uranium, or the barrels of oil. It includes energy that the end user needs in the form of electricity, transport and heating, plus inefficiencies and energy that is lost when raw resources are transformed into a usable form. For the purpose of this study, it captures the industrial energy consumption, which it output determines the sector’s performance.

Another core variable in the study is the renewable energy consumption which is defined as the share of renewable energy in total final energy consumption. This indicator includes energy consumption from all renewable resources such as; hydro, solid biofuels, wind, solar, liquid biofuels, biogas, geothermal, marine, and waste.

The fourth core variable—carbondioxide (CO₂) emission —is a proxy for fossil fuel consumption. CO₂ emissions are those stemming from the burning of fossil fuels and the manufacture of cement. They include carbon dioxide produced during consumption of solid, liquid, and gas fuels and gas flaring. These emissions affects industrial performance directly and indirectly. For example, industries can be affected by the physical impacts of climate change, including extreme weather conditions, sea-level rise and changes in temperature patterns. Such impacts can disrupt supply chains, damage infrastructure, and affect the availability of resources, potentially leading to disruptions in industrial operations.

Considering the control variables, foreign direct investment (FDI) are the net inflows of investment to acquire a lasting management interest (10% or more of voting stock) in an enterprise operating in an economy other than that of the investor (WDI, 2023). It is the sum of equity capital, reinvestment of earnings, other long-term capital, and short-term capital as shown in the balance of payments. This series shows net inflows (new investment inflows less disinvestment) in the reporting economy from foreign investors, and is divided by GDP.

Finally, official exchange rate refers to the exchange rate determined by national authorities or to the rate determined in the legally sanctioned exchange market. It is calculated as an annual average based on monthly averages (local currency units relative to the U.S. dollar). The exchange rate is a critical factor influencing industrial performance through its effects on export competitiveness, import costs, profitability, investment decisions, inflation, global value chains and trade balances. Industries and policymakers closely monitor it to understand its implications and make informed decisions to support economic growth and competitiveness.

Model Specifications

This section lays down a model which will help to understand the relationship between industrial performance and energy consumption in sub Saharan Africa.

$\begin{matrix} lnIN D_{it} = β_{0} + β_{1} lnE C_{it} + β_{2} lnRE C_{it} + β_{3} lnC O_{2 it} \\ + β_{4} FD I_{it} + β_{5} lnEX R_{it} + β_{6} lnIN D_{it - 1} \\ + β_{7} ζ_{it - 1} + f_{it} + μ_{it} \end{matrix}$ (1)

where lnIND_it is the of dependent variable (industrial performance) for the region i and period t; lnEC is energy consumption for region i at period t; lnREC_it is renewable energy consumption (% of total final energy consumption); lnCO_2it is carbon monoxide emission for region i at period t; FDI_it is foreign direct investment, net inflows (BoP, current US$) and lnEXR_it is exchange rate for region i at period t. lnIND_it-1 is the lagged value of the dependent variable; ζ_it-1 is the lagged values of the vector of independent variables. β ₁ , β ₂ , β₃β₄, and β₅ are the parameters; ϕ_it represents the entity-specific fixed effects to capture unobserved heterogeneity across entities; and μ_it is the white noise/error term. Apart from FDI, all the other variables are presented in their natural logarithmic form.

Estimation Technique

The panel Fully Modified Ordinary Least Squares (FMOLS) estimation method is utilized to examine the relationship between energy consumption and industrial performance in SSA. Phillips and Hansen (1990) first proposed the FMOLS, which Phillips and Moon (1999), Pedroni (2001) and Kao and Chiang (2001) further expanded. Phillips and Hansen (1990) proposed an estimator that utilizes a semi-parametric correction to remove the deficiencies caused by the longrun correlation between the cointegrating equation and stochastic regressors innovations. The resulting Fully Modified OLS (FMOLS) estimator is asymptotically unbiased and has fully efficient mixture of normal asymptotics allowing for standard Wald tests using asymptotic Chi-square statistical inference. However, the study used the Pedroni (2001) heterogeneous FMOLS estimator for the panel cointegration regression as it has the advantage of correcting endogeneity bias and serial correlation (Ozcan, 2013). According to Hamit-Haggar (2012), FMOLS is the most suitable technique for the panel, which includes heterogeneous cointegration.

The FMOLS is superior to the Ordinary Least Squares (OLS) for some key reasons. First, OLS estimates are super-consistent, but the t-statistic gotten without stationarity at level terms are only approximately normal. Even though OLS is super-consistent, in the presence of a large finite sample bias, convergence of OLS can be low in finite samples. Secondly, OLS estimates may suffer from serial correlation and heteroskedasticity since the omitted dynamics are captured by the residual, hence, the inference using the normal tables will not be valid—even asymptotically. Therefore, t-statistics for the OLS estimates are useless. Thirdly, FMOLS take care of endogeneity. They involve additional transformations to ensure that the estimated coefficients are unbiased and consistent when cointegration is present. The method addresses endogeneity and autocorrelation issues in time-series data, providing more robust estimates, thus, ensuring that the results are valid and statistically efficient.

Presentation of Results

Descriptive Statistics

Table 2 presents the summary of descriptive statistics of the variables employed in the study. Careful observation shows that the industrial growth rate averaged 21.27% ranging from 17.02% to 25.33%. In similar fashion, the average rate of energy consumption under the period of review is 7.1%, as it fluctuated between 4.98% to 8.95%. Renewable energy consumption on the other hand averaged 4.25% with a consistent fluctuation between 3.03% to 4.59%. Fossil fuel consumption on the other hand averaged 7.67% growth and revolved between 4.09% and 11.66% during the period under review. The Table also shows the distribution of the variables. Four of the variables (lnLIND, lnLEC, lnCO₂, and lnEXR) are normally distributed as depicted by the value of Jarque-Bera. The Jarque-Bera test is based on the sample skewness and kurtosis. For a sample to be approximately normally distributed, the skewness should be close to zero, and the kurtosis should be close to 3 (which is the kurtosis of a normal distribution). The values of skewness and kurtosis further reveal a relatively low standard deviation that suggests a small deviation from their respective mean values.

Table 2.

Descriptive Statistics.

Statistics	lnIND	lnEC	lnREC	lnLCO2	FDI	lnEXR
Mean	21.27125	7.100926	4.247380	7.670540	4.922459	5.341596
Median	21.36055	7.085603	4.378959	7.705252	2.693126	6.155386
Maximum	25.33238	8.948052	4.587719	11.65512	103.3374	9.125205
Minimum	17.02230	4.983516	3.033991	4.094345	−11.19898	0.701165
Std. Dev.	1.739527	0.869095	0.331820	1.508717	9.208341	1.974170
Skewness	−0.137935	−0.031825	−1.677010	0.136611	5.766007	−0.463621
Kurtosis	2.782223	2.249080	5.373179	2.977395	48.79299	2.329733
Jarque-Bera	2.964746	13.63037	405.1544	1.803881	53519.64	31.36235
Probability	0.227098	0.001097	0.000000	0.405781	0.000000	0.000000
Sum	12,252.24	4,090.133	2,446.491	4,418.231	2,835.336	3,071.418
Sum Sq. Dev.	1,739.923	434.3124	63.31007	1,308.830	48,756.29	2,237.078
Observations	576	576	576	576	576	575

Source. Authors’ computation.

Multicollinearity

This test is conducted to check the collinearity among the exogenous variables. If there exists a relationship of such among the regressors, it becomes difficult to determine their coefficients. The results show that there is no threat of collinearity since none of the variables have a relationship of up to 0.8 as seen in Table 3 below.

Table 3.

Correlation Matrix.

Variables	lnLEC	lnREC	lnCO₂	FDI	lnEXR
lnEC	1.000000	−0.649277	0.233151	0.108167	−0.536182
lnREC	−0.649277	1.000000	0.120120	−0.027648	0.289975
lnCO₂	0.233151	0.120120	1.000000	−0.097224	−0.026262
FDI	0.108167	−0.027648	−0.097224	1.000000	−0.103061
lnEXR	−0.536182	0.289975	−0.026262	−0.103061	1.000000

Source. Authors’ computation.

Unit Root Test

The study adopted the two famous panel unit root tests viz, LLC, (Levin et al., 2002) and Im et al. (2003) tests for the stationarity. These two tests were applied to the chosen variables at their levels and first difference, with IPS test used to augment the LLC test. The study combines LLC (Levin et al., 2002) as well as IPS (Im et al., 2003) panel unit root tests to ascertain the stability of the variables. These tests are structured differently to capture the various aspects of unit root testing in panel data. LLC tests essentially focuses on the existence of a common unit root. That is, whether common unit roots exist in the panel as a whole. On the other hand, IPS primarily examines whether there are unit roots at the individual (cross-sectional) level. Combining these two tests enhances the robustness and ensures a more elaborate assessment of unit root behavior in the panel.

The FMOLS requires stationarity at first difference for all the variables. According to Erdal and Erdal (2020), the panel unit root test has some advantages, namely, the provision of a large number of point data that increases the value of the degree of freedom as well as reducing multicollinearity between the regressors. In addition, the panel unit root test provides more powerful test statistics that asymptotically follow a normal distribution (Erdal & Erdal, 2020). The results are presented in Table 4 below. Both results show that all the variables are stationary at first difference, I(1).

The panel unit root test results presented in Table 4 clearly demonstrates that not all the variables are I(0) variables, but they are all stationary at first difference, The results enables the study to proceed to the panel cointegration tests to ascertain whether or not there exists a cointegration equation among the variables.

Table 4.

Unit Root Test.

Variable	Test	Level		First difference
Variable	Test	Statistic	Probability	Statistics	Probability
lnIND	LLC	−3.75068	.0001***	−14.3864	.0000***
	IPS	2.07206	0.9809	−11. 4,337	.0000***
lnEC	LLC	−3.33006	.0004***	−14. 6,156	.0000***
	IPS	0.31680	.3757	−14.7605	.0000***
lnREC	LLC	1. 99,398	.9769	−14.6327	.0000***
	IPS	4.7948	1.0000	−11.7353	.0000***
lnCO₂	LLC	−0.78168	.2172	−15.1465	.0000***
	IPS	4.43462	1.0000	−14.0918	.0000***
FDI	LLC	−6.47834	.0000***	−25.2843	.0000***
	IPS	−6.90259	.0000***	−21.9231	.0000***
EXR	LLC	2.57278	0.9950	−10.2137	.0000***
	IPS	5.17161	1.0000	−8.42266	.0000***

***

**, and * denotes significance at 1%, 5%, and 10% respectively.

Source. Author’s compilations.

Note. lnIND stands for the industrial sector annual growth; lnEC stands for primary energy consumption per capita; lnREC implies renewable energy consumption (% of total final energy consumption); lnCO₂ stands for carbon emissions (kt); FDI stands for Foreign direct investment, net inflows (% of GDP); lnEXR stands for official exchange rates.

Panel Co-integration Test

For the same order of integration—I(1) variables—, the study employs Pedroni (1999) Residual Cointegration Test and Kao (1999) residual cointegration test to find out the existence of cointegrating relationship to enable an unbiased FMOLS results. The two cointegration tests are employed for robustness checks and ensures a more detailed and consistent assessment of the results. Pedroni (1999) cointegration tests is carried out by counting the p-values from all the panel and group statistics. The conclusion depends on the number of significant values found out of the 11 results reported. The results showed that more than half (six) of the total (11) statistics show statistical significance.

According to Kao (1999) Residual Cointegration Test, a p-value of .0000 shows significance 1% significance level. The results are reported in Table 5. Hence, according to the p-value, there is cointegration among these variables.

Table 5.

Cointegration Tests. Null Hypothesis: No cointegration. Included observations: 576. Cross-sections included: 32.

Panel A: Pedroni Residual Cointegration Test
	Weighted
	Statistic	Prob.	Statistic	Prob.
Panel v-Statistic	0.578289	.2815	0.689365	.2453
Panel rho-Statistic	5.131745	1.0000	5.561480	1.0000
Panel PP-Statistic	−3.714334	.0001***	−2.950957	.0016***
Panel ADF-Statistic	−4.078595	.0000***	−5.247747	.0000***
Alternative hypothesis: individual AR coefs. (between-dimension)
Group rho-Statistic	7.410494	1.0000
Group PP-Statistic	−6.523903	.0000***
Group ADF-Statistic	−4.302172	.0000***
Panel B: Kao Residual Cointegration Test
F(lnIND, lnEC, lnREC, lnCO₂, FDI, EXR)	−3.975913	.0000***

***

**, and * denotes significance at 1%, 5% and 10% respectively.

Source. Author’s compilations.

Note. lnIND stands for the industrial sector annual growth; lnEC stands for primary energy consumption per capita; lnREC implies renewable energy consumption (% of total final energy consumption); lnCO2 stands for carbon emissions (kt); FDI stands for Foreign direct investment, net inflows (% of GDP); lnEXR stands for official exchange rates.

Panel FMOLS and Robustness Checks Results

Table 6 illustrates the outcomes of the FMOLS and robustness estimates for the model. The robustness estimates include the Pooled Mean Group (PMG) or panel autoregressive distributed lag (ARDL) model and the Generalized Linear Models (GLM). All these models report similar findings in terms of the signs and significance of their coefficients.

Table 6.

Results of FMOLS, PMG and GLM.

Dependent variable: lnIND
			Robustness checks
	FMOLS		PMG – ARDL(2, 1, 1, 1, 1, 1)		GLM
Variables	Coefficient	p-Values	Coefficient	p-Values	Coefficient	p-Values
			Long-run
lnEC	0.648319	.0000***	0.663588	.0000***	0.653303	.0000***
lnREC	2.140918	.0000***	1.745173	.0000***	2.122422	.0000***
lnCO₂	0.964647	.0000***	1.162284	.0000***	0.967654	.0000***
FDI	−0.017359	.0000***	−0.002447	.0038***	−0.016923	.0000***
EXR	0.046045	.0000***	0.125222	.0000***	0.049299	.0026***
			Short-run equation
ECM	-	-	−0.057381	.0145**	-	-
D(LEC)	-	-	0.013278	.8865	-	-
D(LREC)	-	-	1.201581	.1143	-	-
D(LCO₂)	-	-	0.266613	.0040***	-	-
D(FDI)	-	-	−0.001703	.5045	-	-
D(EXR)	-	-	−0.097091	.1736	-	-

***

**, and * denotes significance at 1%, 5%, and 10%, respectively.

Source. Author’s compilations.

According to the FMOLS results, there exist a positive and significant relationship between energy consumption and industrial performance. That is, as energy consumption increases, industrial performance also increases. In this case, a percentage increase in energy consumption increases industrial performance by 67.8%. These results corroborates the findings of Zehra et al. (2020), Abbasi et al. (2021), Abokyi et al. (2018), Ekong et al. (2021). This therefore means that increased consumption of energy necessitates the industry to increase its production toward attaining full capacity. Intuitively, when industries have access to energy, it steers their production, which impacts positively on economic growth. The reason is that there will be more goods produced and available in the market for exports. This will generate revenue for the country by triggering positive balance of payments that translates to economic growth. In addition, increased output production requires more labor force. When the services of both skilled and unskilled labor are employed by the industries to speed up production, the result is a decrease in unemployment rate. The implication is that the economy tends toward the growth trajectory. Little wonder that Eke et al. (2018) considered the development of industrial activities a catalyst for rapid growth.

Renewable energy consumption positively impacts industrial performance. Similar to the findings above, increased consumption of renewable energy enhances the productive capacities of industries since it is a major determinant of industrial activities. It can take the form of solar energy, wind energy, hydropower, biomass energy, and geothermal energy. These forms of renewable energy offer sustainable alternatives to fossil fuels, contributing to reduced greenhouse gas emissions, energy security and environmental conservation. In this context, the United Nations Energy Statistics Pocketbook (2022) confirms that Africa is one of the regions in the world where renewable energy sources play leading roles in energy consumption. Some low income countries in the region of this study like Uganda, Central African Republic, Somalia, and Democratic Republic of Congo were the only countries in the world in 2019 to have renewable energy sources make up to 90% of total consumption. The latter even came close to 100%, with 96.2%. Africa could be on the right trajectory toward achieving a clean atmosphere if more renewable energy sources are exploited and fossil fuel consumption is minimized.

Further findings reveal that CO₂ emission is also of positive impact to the sector. Increased CO₂ emission enhances the sectoral performance by 96.5%. The findings align with Bernard and Adenuga (2016) who also reported a positive impact of CO₂ emission on industrial performance. This implies that, as industries are utilized for production, they emit CO₂. Higher emissions could imply increase in industrial activities during the production process. The more they produce, the more CO₂ is discharged. This is very popular among developing countries. According to Raihan and Tuspekova (2022), many countries will consistently increase CO₂ emissions as they strive toward economic growth through the industrial sector. Aye and Edoja (2017) opine that as economies embark on growth trajectories, industrial activities rise; the increase in their activities triggers more CO₂ emissions responsible for environmental degradation impacting the ecosystem and human populations. According to Otim et al. (2022), these emissions are responsible for global warming and climate change.

Similarly, exchange rate is also of positive influence to the industrial sector performance. It accounts for an improved performance of about 0.79% for every unit increase. Other studies (including; Akinmulegun & Falana, 2018; Ilechukwu et al., 2015; Jongbo, 2014; Musa & Sanusi, 2013) reported similar findings. This is to say that whenever the currency appreciates against the dollar, there are huge tendencies that industrial activities will be enhanced by attracting foreign investors through Foreign Direct Investments (FDI). FDI has proven its importance over the years to both the host economy and the foreign investor. The host economy enjoys growth in business activities; boost the exportation of goods and services, clamps down unemployment and accelerates economic growth and development. FDI also triggers development financing and growth by boosting the total investments in the receiving or host country and thus enhances gains in productivity via skills in management and technology (Emako et al., 2023). However, the results from this study contradicts this claim. This negative impact of FDI on industrial performance can be due to the possibility of existence of a direct relationship between FDIs and the absorptive capacities in the host countries as noted by Görgülü and Akcay (2015). The absorptive capacities refer to the ability to recognize the value of new information, assimilate it and apply it to commercial ends. Therefore, in the case where these absorptive capacities are insufficient, FDI may have negative effects on economic growth through the industrial sector. The view that justifies the assertion that FDI brings forward the increase in general productivity of the host countries following through externalities in the form of the technological spillovers depends on some political regulations and in some specific environments. That is, countries with comprehensive administrative structures and well-organized markets have high absorptive capacities and are able to benefit as much as possible from FDI while economies that lack such solid administrative and financial structures are not able to extract such positive effects out of FDIs. This is a testament among these countries sampled in this study, as they are characterized by weak structures leading to over-dependency on foreign capital which makes their economies vulnerable to external shocks. Thus, in a case where the foreign companies relied upon repatriate a chunk of their profits back to their home countries, resources become drained for domestic investment and economic growth and development. Herzer (2012) and Azman-Saini et al. (2010) suggested that dependence of foreign investors upon the economic policies in the host economies support the idea that FDI itself does not carry direct positive association in the emerging economies. Other factors like resource course contributes to the negative relationship between FDI and industrial performance as abundant natural resources may discourage investment in the sector. This can lead to a situation where high levels of FDI in natural resource extraction industries negatively impact the development of the industrial sector. In like manner, FDI inflows into resource-rich sectors can also lead to currency appreciation, making non-resource sectors, including manufacturing, less competitive in international markets. This can hinder the growth of domestic industries by making their exports more expensive and imports cheaper, thereby reducing competitiveness and hindering industrial development.

The study further employs the pooled mean group (PMG) and generalized linear model (GLM) for robustness checks to ascertain the consistency of the results. The results of the two models are almost identical with the FMOLS, therefore validating the findings. The choice of the PMG over the traditional fixed effect (FE) or random effect (RE) models is that it offers greater efficiency. In addition, it is relatively robust to misspecification of the underlying data generating process, thus, making it suitable for empirical applications where the true model specification may be uncertain. Furthermore, PMG accounts for potential cross-sectional dependence, which is common in panel data where observations across entities may be correlated. By allowing for cross-sectional dependence, it improves the efficiency of parameter estimates and enhances the reliability of inference while controlling for individual-specific effects and time-specific dynamics. Just like the FMOLS, PMG addresses endogeneity and serial correlation issues commonly encountered in panel data analysis by accounting for the lagged dependent variables and other potential sources of endogeneity, therefore, improving the validity of estimated coefficients. Finally, it integrates both time series and cross-sectional variation, capturing both within-entity and between-entity variations over time. This comprehensive approach allows for a more complete analysis of panel data and facilitates the exploration of complex economic relationships. In conclusion, the PMG provides short-run results. However, only CO₂ emissions reports statistical significance among the independent variables. The error correction model (ECM) reports the speed of adjustment of the model back to equilibrium at 5.7%. The negative sign and significance validates the correctness of the long-run results.

The GLM on the other hand extend the linear regression framework to accommodate a wide range of response distributions beyond the normal distribution. Even in the presence of outliers, it provides efficient estimates of model parameters and their standard errors. This flexibility enables researchers to analyze a diverse set of data types and address different research questions.

Causality Test Results

Following the estimated long-run results, the study employed the Pairwise Dumitrescu Hurlin Panel causality tests to ascertain whether there is a relation of causality between industrial performance, energy consumption, renewables, CO₂ emission and exchange rate (See Appendix Table A1). The panel causality tests results reveal that there is no causality running from energy consumption to industrial performance, but a uni-directional causality running from industrial performance to energy consumption. This suggests that the growth of the sector determines the level of energy to consume. Further findings reveal another uni-directional causality from industrial performance to renewable energy consumption. Since renewable energy consumption is part of the total energy consumption, the industrial sector’s capacity dictates the quantity required of it to be used in the production process. Similarly, another uni-directional causality runs from industrial performance to CO₂ emission. Fossil fuel is another form of energy consumed by the sector, the demand for it strictly depends on the size of the industrial sector. Conversely, unidirectional causality runs from FDI to industrial performance. That is, the growth of the sector relies reasonably on the activities of FDI. It further emphasizes on the vulnerability of industries to external shocks. Finally, the results show a direct causality running from industrial performance to exchange rates. Suggesting that industrial growth boosts local currencies through the exportation of the goods produced by the local industries.

Conclusion

The motivation of this research is to study the performance of the industrial sector in SSA. With studies confirming a positive impact of industrial sector on economic growth across several countries, the analysis of this paper explores the drivers of industrial performance in the region. The FMOLS results report a positive nexus between energy consumption, renewable energy consumption, fossil fuel consumption and exchange rates on annual industrial sector growth. These findings were further validated by the PMG and GLM estimates that were employed for robustness checks. Consequently, the main conclusion of the study is that energy consumption in whichever form is a key driver in the region. This paper differs from previous studies by using the income classification of countries in SSA by the World Economic Situation and Prospects (2022). The primary focus was on countries classified into income groups to include lower-middle-income and low income between 2002 and 2019. However, the study did not conduct separate analysis for the two groups since they share similar economic and political structures. This is important not only for comparative reasons but also to understand the economic conditions and trends of these group of SSA countries. While previous studies focused on the relationship between electricity consumption and economic growth or manufacturing sector, this study examined the impact of energy consumption on industrial sector performance. The FMOLS, PMG and GLM estimates were utilized to examine the long run relationship between energy consumption, renewable energy consumption, fossil fuel consumption, FDI, and exchange rate on industrial performance in SSA.

The three econometric models revealed positive impacts of energy consumption, renewable energy consumption, fossil fuel consumption and exchange rate on industrial performance. However, a negative relationship is reported between FDI and industrial performance.

The Dumitrescu-Hurlin panel causality test reports a uni-directional causality running from industrial sector performance to energy consumption, renewable energy consumption, exchange rates and fossil fuel consumption. However, only FDI causes industrial sectoral performance.

With the subsequent emergence of more recent data, further studies should be embarked upon to investigate this relationship in order to ascertain the consistency in the findings. This study can be implemented in the SSA region to observe the linkage between industrial performance, energy consumption, exchange rate and FDI, especially in this era where CO₂ emission through fossil fuel consumption in countries across the world have triggered climate change. These studies would be country-specific and cross-country in order to enrich the scarce literature. Different econometric techniques could also be used to ensure robustness of the results. The studies are potentially useful in this modern era for decision-making, where countries of the world are shifting toward a green economy.

This study recommends the following; firstly, an in-depth examination of the level of CO₂ emission, especially among lower-middle-income and low income countries. This is because its excessive discharge has sparked global climate concerns, therefore, the governments in these countries should encourage the consumption of clean energy by incentivizing industries that use renewable energy. Secondly, the findings report a positive linkage between exchange rate and industrial growth, meaning that exchange rate is a key determinant of industrial performance. For this reason, industrial outputs should be exported to other continents of the world in order to strengthen the currencies used in these countries. This would reduce imports and dependence on consumables from abroad. Therefore, undermining the importance of exchange rate on industrial output would have detrimental effects on economic growth and development. The activities of FDI must be closely examined by the governments of the host countries to avert potential detrimental effects on the growth of the sector. Finally, to coax industrial activities, efficient use of energy in powering the industries must be closely monitored to ensure that these industries have all it takes to attain maximum production levels with minimal CO₂ emissions.

These policy recommendations are essential in improving industrial performance, as it is the major driver of industrialization. Industrialization is strongly linked to domestic economies. It will promote and help African countries achieve high growth rates, economic diversification and reduce their vulnerability and exposure to external shocks as a result of overdependence on foreign aid. This will substantially contribute to poverty reduction through employment and wealth creation.

Footnotes

Appendices

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research,authorship,and/or publication of this article.

Funding

The author(s) received no financial support for the research,authorship,and/or publication of this article.

ORCID iDs

Ogochukwu Christiana Anyanwu

Keghter Kur

Data Availability Statement

Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.

References

Abbasi

K. R.

Hussain

Abbas

Adedoyin

F. F.

Shaikh

P. A.

Yousaf

Muhammad

(2021). Analyzing the role of industrial sector's electricity consumption, prices, and GDP: A modified empirical evidence from Pakistan. AIMS Energy, 9(1), 29–49.

Abid

Mraihi

(2015). Energy consumption and industrial production: Evidence from Tunisia at both aggregated and disaggregated levels. Journal of the Knowledge Economy, 6(4), 1123–1137.

Abokyi

Appiah-Konadu

Sikayena

Oteng-Abayie

E. F.

(2018). Consumption of electricity and industrial growth in the case of Ghana. Journal of Energy, 2018, 1–11.

Akinmulegun

S. O.

Falana

O. E.

(2018). Exchange rate fluctuation and industrial output growth in Nigeria. International Journal of Economics and Financial Research, 4(5), 145–158.

Aye

G. C.

Edoja

P. E.

(2017). Effect of economic growth on CO2 emission in developing countries: Evidence from a dynamic panel threshold model. Cogent Economics & Finance, 5(1), 1–22.

Azman-Saini

W. N. W.

Baharumshah

A. Z.

Law

S. H.

(2010). Foreign direct investment, economic freedom and economic growth: International evidence. Economic Modelling, 27(5), 1079–1089.

Azolibe

C. B.

Okonkwo

J. J.

(2020). Infrastructure development and industrial sector productivity in Sub-Saharan Africa. Journal of Development Economics, 22(1), 91–109.

Bataille

Melton

(2017). Energy efficiency and economic growth: A retrospective CGE analysis for Canada from 2002 to 2012. Energy Economics, 64, 118–130.

Bernard

O. A.

Adenuga

(2016). Impact of economic reforms on performance of the manufacturing sector in Nigeria. SSRN. https://doi.org/10.2139/ssrn.2717655

10.

Chebbi

H. E.

Boujelbene

(2008). Agriculture and non-agriculture outputs and energy consumption in Tunisia: Empirical evidences from Co-integration and Causality. 12th Congress of the European Association of Agriculture Economists – EAAE.

11.

Danmaraya

I. A.

Hassan

(2016). Energy consumption and Manufacturing industry performance: Evidence from low-income Sub-Saharan African countries. International Energy Journal, 16, 57–64.

12.

Dash

(2016). An analysis of factors affecting energy consumption in Urban Odisha. International Journal of Research and Review, 3(12), 25–32.

13.

Eke

F. A.

Okoi

O. B.

Eke

I. C.

(2018). Institutional quality, energy consumption mix and industrial sustainability in Nigeria. African Journal of Applied and Theoretical Economics, 4(1), 73–94.

14.

Ekong

U. M.

Isaac

U. E.

Isaac

S. E.

(2021). Energy consumption and sectoral performance in Nigeria: New growth evidence from an ARDL model for consideration. Asian Basic and Applied Research Journal, 4(2), 103–119.

15.

Emako

Nuru

Menza

(2023). The effect of foreign direct investment on capital accumulation in developing countries. Development Studies Research, 10(1), 1–15. https://doi.org/10.1080/21665095.2023.2220580

16.

Erdal

(2020). Panel FMOLS model analysis of the effects of livestock support policies on sustainable animal presence in Turkey. Sustainability, 12(8), 3444.

17.

Görgülü

M. E.

Akcay

(2015). The effects of foreign direct investments on developing countries. Eurasian Studies, Republic of Turkey Prime Ministry Turkish Cooperation and Coordination Agency, 47(1), 7–36.

18.

Gozgor

Lau

C. K. M.

(2018). Energy consumption and economic growth: New evidence from the OECD countries. Environmental Science and Pollution Research International, 153(13), 27–34.

19.

Hamit-Haggar

(2012). Greenhouse gas emissions, energy consumption and economic growth: A panel cointegration analysis from Canadian industrial sector perspective. Energy Economics, 34(1), 358–364.

20.

Herzer

(2012). How does foreign direct investment really affect developing countries' growth? Review of International Economics, 20(2), 396–414.

21.

Ilechukwu

I. N.

Nwokoye

E. S.

(2015). Long run impact of exchange rate on Nigeria’s industrial output. Journal of economics and Finance, 6(5), 75–86.

22.

K. S.

Pesaran

M. H.

Shin

(2003). Testing for unit roots in heterogeneous panels. Journal of econometrics, 115(1), 53–74.

23.

Iwayemi

(1983). Energy in West Africa issues and policy approaches. Energy Policy, 11(3), 235–249.

24.

Iwayemi

(1993). Energy development and Sub-Saharan African economies in a global perspective. OPEC Review, 17(1), 29–45.

25.

Jongbo

O. C.

(2014). The impact of real exchange rate fluctuation on industrial output in Nigeria. Transport, 2(2.4), 3–1.

26.

Kao

(1999). Spurious regression and residual-based tests for cointegration in panel data. Journal of econometrics, 90(1), 1–44.

27.

Kao

Chiang

M. H.

(2001). On the estimation and inference of a cointegrated regression in panel data. In Nonstationary panels, panel cointegration, and dynamic panels, 179–222

28.

Levin

Lin

C. F.

Chu

C. S. J.

(2002). Unit root tests in panel data: Asymptotic and finite-sample properties. Journal of Econometrics, 108(1), 1–24.

29.

Liew

V. K.

Nathan

T. M.

Wong

(2012) Are sectoral outputs in Pakistan led by energy consumption? Economic Bulletin, 32(3), 2326–2331

30.

Maroof

Hussain

Jawad

Naz

(2018). Determinants of industrial development: A panel analysis of South Asian economies. Quality & Quantity, 53(3), 1391–1419.

31.

Mawejje

D. N.

(2016). Electricity consumption and sectoral output in Uganda: An empirical investigation. Journal of Economic Structures, 5(1), 1–16.

32.

Morris

Fessehaie

(2014). The industrialisation challenge for Africa: Towards a commodities based industrialisation path. Journal of African Trade, 1(1–2), 25–36.

33.

Musa

Sanusi

J. A.

(2013). Industrial output response to inflation and exchange rate in Nigeria: An empirical analysis. Journal of Economics and Sustainable Development, 4(20), 74–81.

34.

Nwosa

Akinbobola

(2012). Aggregate energy consumption and sectoral output in Nigeria. African Research Review, 6(4), 206–215.

35.

Ologbenla

(2020). Institutional quality, human capital and industrial sector growth in ECOWAS. Studia Universitatis Babeà-Bolyai Oeconomica, 65(3), 1–13.

36.

Omosebi

Aladejana

S. A.

Ajetunmobi

K. T.

Asagunla

T. M.

(2019). Disaggregated energy consumption and industrial output in Nigeria. International Journal of Academic Research in Economics and Management Sciences, 8(4), 15–26.

37.

Otim

Mutumba

Watundu

Mubiinzi

Kaddu

(2022). The effects of gross domestic product and energy consumption on carbon dioxide emission in Uganda (1986-2018). International Journal of Energy Economics and Policy, 12(1), 427–435. https://doi.org/10.32479/ijeep.12552

38.

Oyerinde

A. A.

(2019). Industrial Sector Growth, Macroeconomic Performance, and Corruption in the Sub Sahara Africa. Studia Universitatis Babes Bolyai-Oeconomica, 64(2), 72–83.

39.

Ozcan

(2013). The nexus between carbon emissions, energy consumption and economic growth in Middle East countries: A panel data analysis. Energy Policy, 62, 1138–1147.

40.

Pedroni

(1999). Critical values for cointegration tests in heterogeneous panels with multiple regressors. Oxford Bulletin of Economics and Statistics, 61(s1), 653–670.

41.

Pedroni

(2001). Fully modified OLS for heterogeneous cointegrated panels. In Baltagi

B. H.

Fomby

T. B.

Hill

R. Carter

(Ed.), Nonstationary panels, panel cointegration, and dynamic panels, 15, (93–130). Emerald Group Publishing Limited.

42.

Phillips

P. C.

Hansen

B. E.

(1990). Statistical inference in instrumental variables regression with I (1) processes. The review of economic studies, 57(1), 99–125.

43.

Phillips

P. C.

Moon

H. R.

(1999). Linear regression limit theory for nonstationary panel data. Econometrica, 67(5), 1057–1111.

44.

Polat

(2021). The impact of renewable and nonrenewable energy consumption on economic growth: A dynamic panel data approach. Asia-Pacific Journal of Accounting & Economics, 28(5), 592–603. https://doi.org/10.1080/16081625.2018.1540936

45.

Qazi

A. Q.

Ahmed

Mudassar

(2012). Disaggregate energy consumption and Industrial output in Pakistan: An empirical analysis. Economics, 2012-2029

46.

Rahman

M. M.

Kashem

M. A.

(2017). Carbon emissions, energy consumption and industrial growth in Bangladesh: Empirical evidence from ARDL cointegration and Granger causality analysis. Energy policy, 110, 600–608.

47.

Raihan

Tuspekova

(2022). Role of economic growth, renewable energy, and technological innovation to achieve environmental sustainability in Kazakhstan. Current Research in Environmental Sustainability, 4, 1–10. https://doi.org/10.1016/j.crsust.2022.100165

48.

Ranis

(1973). Industrial Sector Labor Absorption. Economic Development and Cultural Change, 21(3), 387–408.

49.

Stern

D. I.

(2019). Energy and economic growth. In Soytas

Sari

(Eds.), Routledge handbook of energy economics (pp. 28–46). Routledge.

50.

United Nations. (2022). World economic situation and prospects 2022. United Nations. https://www.un.org/en/desa/world-economic-situation-and-prospects-wesp-2022

51.

United Nations Energy Statistics Pocketbook (2022). https://unstats.un.org/unsd/energystats/pubs/documents/2022pb-web.pdf

52.

United Nations Industrial Development Organization (2020). https://www.unido.org/resources-publications-industrial-development-report-series/idr2020

53.

World Bank Development Indicators (WDI) (2023). The World Bank. http://databank.worldbank.org

54.

Yildirim

H. H.

(2017). Economic growth and energy consumption for OECD countries [Conference session]. Regional Studies on Economic Growth, Financial Economics and Management: Proceedings of the 19th Eurasia Business and Economics Society Conference (pp. 245–255). Springer International Publishing.

55.

Zehra

Aurmaghan

Ismail

(2020). Energy consumption and economic growth in Pakistan. Journal of Contemporary Macroeconomic Issues, 1(1), 55–66.

56.

Ziramba

(2009). Disaggregate energy consumption and industrial production in South Africa. Energy Policy, 37(6), 2214–2220.