Volume 8 Number 1, 2023 Volume 8 Number 1, 2023 Faculty of Social Sciences, Enugu State University of Science And Technology EDITOR-IN-CHIEF Prof. Oby Omeje MANAGING EDITOR Prof. Barnabas Nwankwo PUBLISHED BY ENUGU STATE UNIVERSITY OF SCIENCE & TECHNOLOGY JOURNAL OF SOCIAL SCIENCES & HUMANITIES 29 Energy Sector Performance and Real Sector Activity in Nigeria: An Empirical Re-examination Agu, Sunday Virtus1*, Azike Lawrence Chike1 Kalu, U. Chris1,2 1Department of Economics, Enugu State University of Science and Technology, Enugu. 2Department of Economics, Nnamdi Azkiwe University, Awka *Corresponding Author: sunday.agu@esut.edu.ng Abstract In spite of the remarkable achievements in the Nigerian economy, the supply of electricity has remained the poor same and this has led to a shift in other alternative sources of power which require burning of fossil fuels and has consequently increased toxic emission. This paper examined the relationship between energy sector performance and real sector activity in Nigeria between 1986 to 2021 using the dynamic autoregressive distributed lag and Granger causality approaches. The model included variables are manufacturing value added (the dependent variable) and electricity consumption, the real gross domestic product, foreign direct investment, population growth and private sector credit as explanatory variables. The finding showed that electricity consumption (performance) has a positive but insignificant impact on manufacturing value added (productivity), while both foreign direct investment and real gross domestic product have positive and significant impacts on manufacturing value added. Furthermore, the Granger causality result showed bidirectional causality between electricity consumption and manufacturing value added and unidirectional causality between manufacturing value added and private sector credit. This paper recommended among others the need for monetary and fiscal policy interactions promote real sector- manufacturing productivity, thus, the Central Bank of Nigeria and Bank of Industry can provide credit incentives at lower cost for industrialist and manufacturers. Keywords: Energy sector performance, electricity consumption, manufacturing productivity, manufacturing value added, econometric applications, Nigeria JEL Codes: F14, Q54, L6, L16 Introduction The real sector measured by manufacturing sub-sector has played an important role in national development especially that of developing economies (Naude, Szirimai & Haraguchi, 2016). The stories of emerging economies-Korea, Singapore and Malaysia have continued to show a positive connection between national economic growth and the growth of manufacturing sector (Banjoko, Iwuji& Bagshaw, 2012). Meanwhile, the sub-sector has not developed to its full capacity in utilization. This has resulted to a lot of challenges including capital flight, poverty, mailto:sunday.agu@esut.edu.ng 30 and unemployment among others. In this paper, manufacturing value added was used to measure manufacturing productivity. In spite of the remarkable achievements in the Nigerian economy, the supply of electricity has remained the poor same. This has led to a shift in other alternative source of power which require burning of fossil fuels and has consequently increased toxic emission (Shuaibu & Oyinlola, 2014). Globally, carbon dioxide (Co2) emissions account for more than 75 percent of greenhouse gas emissions with about 80 percent of it generated by energy sector (Akpan & Akpan, 2012) . This causes a lot concern given the enormous detrimental effects of pollutants emission on the environment. The energy consumption index of Nigeria over the years has being dipping even in the face of reforms. Considerable efforts have been made by several studies to examine the relationship between manufacturing sector performance and electricity consumption for example (Binh, 2011; Omotor, 2008; Omisakin, 2008; Odulara & Okonkwo, 2009; Dantama et al., 2012; Olusanya, 2012; Shauibu & Onyiliola, 2014). Most of these studies focused on economic growth, neglecting the persistence of energy depletion on the manufacturing sector. Other studies include (Ezeh & Nnadi(2016); Aseleye et al.,(2021); Effiong & Inyang(2020); Ugwoke et al.,(2016); Quadiri & Bukola, 2022; Ismail & Hassan, 2016; Kassim & Isik, 2020; Ene et al; 2022, Ume et al., 2019) . With the exception of Quadiri and Bukola (2022), these other studies never considered estimating the link in terms of causality between the energy sector and the manufacturing sector. This becomes imperative for energy and manufacturing policy issues. Therefore, the objective of this paper is to examine and estimate the relationship between both sectors in Nigeria as well as the long-run and short-run relationship between electricity consumption and manufacturing value added in Nigeria from 1986 to 2021. The rest of this paper is adumbrated as follows: Preceding this section is section two, which presents the literature review; section three is the methodology, model specification and data sources. Section four presents the results and discussion, while section five presents the conclusion and policy implications. Empirical Literature Review This section presents the extant empirical studies on the related studies. For example, Abiola and Chisaa (2021) investigated the long-run impact of electricity consumption on manufacturing sector performance proxy by manufacturing value added to GDP using canonical co-integration regression for the period of 1981-2019. The result shows that electricity consumption and credit to manufacturing sector have a negative relationship with output. In the employment equation, consumption in electricity and interest rate has negative effects on employment. In conclusion, effects of electricity consumption as input on the manufacturing sector have not improved the performance in the sector. To improve the situation, the study recommended, among others, the need to create a framework to promote energy efficiency by maximizing output from the power sector to minimize wastage. Ubong and Nora (2020) investigated the relationship between electricity consumption and manufacturing sector performance in Nigeria over the period 1981 to 2018. Electricity 31 consumption per capita was used to represent electricity consumption, while manufacturing sector performance was captured by manufacturing value added. The paper utilized the Augmented Dickey-Fuller unit toot technique, Granger causality test, and the vector autoregressive (VAR) approach. The Granger causality test indicated that a unidirectional causality flows between electricity consumption per capita and manufacturing value added. Meanwhile, the VAR result indicated that both electricity consumption and manufacturing value added were strongly endogenous. That is, they significantly predict themselves. However, electricity consumptions weakly exogenous in predicting manufacturing value added rather, manufacturing value added is strongly exogenous in predicting electricity consumption in Nigeria. This therefore points to the prevalence of the growth driven electricity consumption thesis in Nigeria. This thesis is in support of the conservation of electricity. The variance decomposition and impulse response indicated that variations in manufacturing value added responded more to shocks in itself than from shocks from electricity consumption. Allcott (2019) conducted a study estimating the effect of electricity shortages on Indian manufacturing sector using Cobb-Douglas production function model, the variables used are electricity in kilo watt and manufacturing output in percentage of GDP. Annual time series data from 1992 to 2018 on weather, power sector and manufacturing production were used. The results obtained revealed that power shortages slowed down production in the manufacturing sector. This resulted in revenue reduction of 5.6 to 8.6 percent for the average plant in a short run. The results have also shown that producer’s surplus dropped 9.5 percent for the average plant, of which 3.9 percent was due to capital costs incurred for backup generators. It was also discovered that in the short run plants reduced their inputs in response to electricity shortages and that led to a decrease in total production. Ezeh and Nnadi (2018) examined electricity supply and the output of the Nigerian manufacturing sector. The major objective is to critically determine the impact of electricity supply on the manufacturing output in Nigeria. Numerous literatures only revealed the relationship between economic growth and electricity supply, with little empirical attention on the effect of electricity on the various sectors of the economy. This could lead to fallacy of decomposition because economic growth is a function of the performance of different sectors which certainly differ in their need for electricity. In response to this perceived gap, this study explores the relationship between electricity supply and manufacturing sector’s output in Nigeria. Time series data spanning the period between 1981 and 2017 were analyzed using Johansen cointegration and vector autoregression tests. The results revealed that there exists a long run relationship between electricity and manufacturing output in Nigeria. It also identified that electricity supply has an insignificant relationship with the manufacturing sector in Nigeria. Ganiyu and Odewale (2018) analyzed the performance of Nigerian power sector so as to suggest possible means of ensuring improvements of the sector. Specification and estimation techniques were used for a period of eleven years. Secondary data were sourced from Central Bank of Nigeria (CBN) Statistical Bulletin of 2018 and Nigerian Electricity Regulatory Commission (NERC). The performance of the power generated was evaluated using the overall efficiency and thermal efficiency. The results showed that the average value of the overall efficiency for the ten years period of study was 15.68% while the thermal efficiency had the average value of 15.37%. The result confirmed that deregulation of power sector has no effect on the efficiency of Nigerian power sector when the results were compared with the international best practice 32 standards which are 30% and above for overall efficiency and 45% and above for thermal efficiency. The study therefore suggested possible strategies for efficient power sector improvement. Ibrahim (2017) examined the relationship between electricity consumption, manufacturing output and financial development in Nigeria. Time series data for 1981 until 2005 was used to examine the symmetric relationship between the electric consumption, manufacturing output and financial development in Nigeria. The result indicates the co-movement in the variable over long time horizon, meaning that any inefficiency in electricity supply would impedes industrial output. Moreover, the Granger causality test based on vector error correction framework shows the presence of causality between power utilization of manufacturing firms and economic growth without feedback. In this sense it can be stress that stable electricity consumption is important factor for Nigeria’s manufacturing sector. The result of variance decomposition further indicates that the variation in the industrial output responds more to shocks in the electricity supply than its own shock. Adebusuyi and Obamuyi (2016) examined how electricity demand impacts the performance of the Nigerian manufacturing sector. The study was conducted over 1970 to 2014 period, and the autoregressive distributive lag (ARDL) bounds test for co integration along with the OLS approach were utilized in the study. The bound test for co integration presented an evidence of a long run relationship existing between electricity demand and manufacturing sector performance; while the OLS result of the translog function indicated that electricity is a weak substitute for both capital and labour. Hussain and Lean (2015) investigated the relationship between electricity consumption, output, and price in the manufacturing sector in Malaysia using regression analysis. The result revealed that electricity consumption, output, and price are co integrated in the long run. In addition, it has been found that the relationship between electricity consumption and output is positive. In the long run, they find a unidirectional causality from manufacturing output to electricity consumption. This result indicates that the development of manufacturing sector stimulates greater demand for electricity. Government needs to make sure that the planning of electric supply in the future is in line with the economic development planning to avoid shortage in electricity supply. In the short run, a unidirectional relationship runs from electricity consumption to output is found. A decrease of energy usage in production might reduce the output growth in short run. Aslant (2014) carried out a study in Turkey from 1971 to 2007 using autoregressive distributed lag bound test and Granger causality test. The findings show that there is long-run correlation connecting electricity use and economic growth. Accordingly, the Granger causality test supported the neutrality hypothesis in the short run. However, there is bidirectional Granger causality between electricity use and economic growth in the long-run supporting the feedback hypothesis in Turkey. Okoh and Ebi (2013) using the linear regression technique, invested the impact of the nexus between the investment in infrastructure and institutional quality-captured using corruption and the enforcement of contracts, and economic growth in Nigeria. The study found that corruption had a negative and significant growth effect, while investment in infrastructure had a positive 33 and significant growth impact. On the other hand, the institutional quality – infrastructural investment nexus had an insignificant growth impact. Onakoya (2013) found that the consumption of gas has a positive but insignificant impact on economic growth in Nigeria, while the consumption of electricity, petroleum and aggregate energy consumption have a positive and significant effect on economic growth. The consumption of coal was found to have a negative and significant growth impact. The results of similar study by Olusanya (2012) which covered the period 1985 – 2010 indicated that the consumption of petroleum and electricity have positive growth effects. On the other hand, the consumption of coal and gas were found to have negative growth effects. Akpan and Akpan (2012) utilized a multivariate vector error correction model (VECM) in their investigation of the relation between electricity consumption, carbon emissions and economic growth in Nigeria between 1970 and 2008. The study found no causality between economic growth and electricity utilization in Nigeria. Abu (2012) investigated the link between power supply and business industrial development by examining the influence of government policies on power supply and industrial development in Nigeria. The Johansen co-integration technique was adopted to determine the long run relationship among some macroeconomic variables that includes the industrial component of real gross domestic product (RGDP), explicitly chosen using explanatory variables. The independent variables includes electricity consumption, electricity production (Kwh), growth rate of labour force, real gross fixed capital formation and telephone lines per hundred population and their impact on industrial component of real GDP. Annual time series data on these variables from 1981 to 2010 were collected from the Central Bank of Nigeria Statistical Bulletin, the World Bank and United Nations Statistics. Augumented Dickey Fuller (ADF) and Phillips-Perron (PP) tests are employed to test the order of integration of the variables. The study also performed a Vector Error Correction Model – VECM to correct possible disequilibrium caused in the short- run relationships. The study concluded that electricity condition which is a result of existing government policies exerts a negative impact on industrial output in the long run affects the business viability. Theoretical Framework, Model Specification and Data Sources Theoretical Framework/Model Specification The theoretical framework of this study is the Kaldor growth model. Kaldor’s growth theory suggests a vital significance to the manufacturing industry for economic growth. The Kaldor’s theory is of the assumption that a close association exists between increasing manufacturing output and increasing gross domestic product made possible by the application/usage of the electricity consumption per kilowatt, as such based on the above literature and following the framework of the adopted growth models, the relationship we want to estimate can be written as: MVA = α0 + β1LnDCRE + β2DINVEST + β3LnELCON + β4LnFDI + β5LnPGR + β6LnOILPRICE + β7RGDP + μt 3.1 Where MVA is manufacturing value added as percentage of gross domestic product (GDP), DCRE is domestic credit, DINVEST is domestic investment, ELCON is electricity consumption per kilowatt; FDI is foreign direct investment, PGR is population growth, proxy for labour force; OILPRICE is oil price; and RGDP is real gross domestic product; α0 is the intercept, which is a 34 constant term. Ln is the natural logarithm of the variables that are not in rates or percentages, for proportional effects. These variables are described and justified according to theoretical and empirical reasoning. Following Pesaran et al (2001), the Error Correction Model (ECM) of the unrestricted Autoregressive Distributed Lag (ARDL) equation based on the model specified in (3.2) is specified as follows: ∆ MVAt = α0 + α1MVAt – 1 + α2DCREt – 1 + α3 DINVEST t – 1 + α4ELCONt – 1 + α5FDIt – 1 + α6PGRt – 1 + α7OILPRICEt – 1 + α8RGDPt – 1 + 1 MVAt – i + i∆ DCREt-i+ I ∆DINVESTt-i+ i ∆ELCONt-i+ i∆FDIt-I + i∆PGRt-i+ + i∆OILPRICEt-I + i∆RGDPt-I + i∆MVAt-I + Ut 3.2 Where: Ut is the white noise error term. To establish the short-run relationship among the variables, the following ARDL-ECM- error correction model was estimated. ∆MVAt = α0 + 1 ∆MVAt – i + i ∆ DCREt-i+ I ∆DINVESTt-i+ i ∆ELCONt-i+ i ∆FDIt-i+ ∆PGRt-i+ i∆OILPRICEt-I + i∆RGDP + i ∆ ECMt-i+ Ut 3.3 Procedurally, the descriptive statistic properties of the variables were examined. The unit root, co-integration and diagnostic test were also examined. Table 3.1 presented the data set sources Table 3.1: Summary of Relevant Data Variables Description/Measurement Source (s) MVA Manufacturing value added – proxy for industrial productivity Central Bank of Nigeria Statistical Bulletin (CBN, 2020) DCRE Domestic credit CBN (2020), NBS (2021) DINVEST Domestic investment CBN (2020), NBS (2021) ELCONS Electricity consumption measured in kilo watt CBN (2020), NBS (2021) PGR Population growth proxy for labour force. NBS (2020) OILPRICE oilprice CBN (2020) FDI Foreign direct investment, proxy for investment CBN (2020) RGDP Real gross domestic product CBN (2020) Where: NBS = National Bureau of Statistics (2020); WDI = World Bank Development Indicator (2021). Source: Authors’ Compilation (2022) Result Presentation and Analysis Descriptive Statistics This study commences its empirical analysis by examining the characteristics of the variables of estimate. The results are presented in Table 4.1 35 Table 4.1: Summary of Descriptive Statistics Variables MVA DCRE DINVEST ELCON FDI PGR RGDP Mean 13.52 9.73 30.68 106.16 1.66 1.41 40.361 Std. Dev. 4.83 3.56 12.89 25.19 1.25 38.670 19.660 Skewness 0.19 0.99 0.31 0.55 1.66 0.33 0.41 Kurtosis 1.49 3.64 1.86 1.97 5.73 1.88 1.59 Jarque-Bera 3.61 6.31 2.53 2.74 27.11 2.52 4.03 Prob 0.16 0.04 0.28 0.25 0.00 0.28 0.3 Observation 36 35 36 29 35 36 36 Note: MVA = Manufacturing value added; DCRE = Domestic credit; DINVEST = Domestic credit; ELCON = Electricity consumption; FDI = Foreign direct investment; PGR = Population growth; RGDP = Real gross domestic product. Source: Authors’ computation using EVIEW-10 Software version From the estimates reported in Table 4.1, it was shown that the mean value for manufacturing value added was 13.52; that of domestic credit were 9.73; that of domestic investment was 30.68 while that of electricity consumption was 106. The foreign direct investment was 1.66; 1.41 for population growth. The mean value of the real gross domestic product was stood at 40.361. The kurtosis coefficient of the variables ranges from 1.49 to 5.73 which reflect a leptokurtic distribution for only FDI. This implies that the data was characterized by the presence of outlier. The probability values for Jaque-Bera suggested that the included variables were normally distributed except for DCRE and FDI. Table 4.2 presents the correlation matrix which detects the presence of multicollinearity in the data series. Table 4.2: Correlation Matrix Result Variables MVA DCRE DINVEST ELCON FDI PGR RGDP MVA 1.000 DCRE 0.75 1.00 DINVEST 0.92 0.74 1.00 ELCON 0.82 0.62 0.83 1.00 FDI 0.02 0.09 0.40 0.80 1.00 OILPRICE 0.87 0.72 0.84 0.91 0.03 PGR 0.93 0.74 0.97 0.83 0.02 1.00 RGDP 0.89 0.73 0.94 0.91 0.05 0.97 1.00 Sources: Authors’ computation using EView 10 Software version The results presented in Table 4.2 show that the variables are positively correlated with one another. However, the correlation coefficient of RGDP shows high degree of correlation with domestic investment. This followed by the unit root test to examine the stochastic properties of the variables. The results are presented in Table 4.3 36 Table 4.3: Unit Root Test Results Variables ADF Test Statistic Philips-Perron Test Statistics Level [I(0)] 1st Diff. Order of Integration I(1) Level [I(0)] 1st Diff. Order of Integration I(2) MVA -1.367 -9.171*** I(1) -1.372 -21.861*** I(1) DCRE -1.630 -5.136*** I(1) 0.42 -2.955*** I(1) DINVEST -0.646 -5.345*** I(1) -1.004 -5.157*** I(1) ELCON -1.180 -5.454*** I(1) -1.180 -5.971*** I(1) FDI -1.661 -3.679*** I(1) 1.960 -3.543*** I(1) OILPRICE -1.653 -8.060*** I(1) 1.646 -8.085*** I(1) PGR -0.643 -5.859*** I(1) -1.466 -5.859*** I(1) RGDP -2.653 -7.122*** I(1) -1.954 -7.333*** I(1) Note: *** and * indicate significant at 1% and 10% levels, respectively or a rejection of the null hypothesis of no unit root. Note: Variables previously defined. Source: Authors’ computation using EView 10 Software Version. The result of the unit root tests as presented in Table 4.3 showed that ADF test result was a mixture of order zero and order one at the 1% significant level. Again, the PP test showed that the variables were integrated at order 1, which is the highest order of integration. The Johansen co-integration test was used to complement the examination of the long-run relationship between the co-integrated variables. The co-integration test is presented in Table 4.4 Table 4.4: Summary of the Co-integration Estimate Trace Test Maximum Eigen Value Test Null Alternative Statistic 0.05 Critical Values Null Alternative Statistics 0.05 Critical Values r = 0 r = 0 446.91 159.53 r = 0 r = 0 172.68 52.36 r ≤ 1 r ≥ 1 274.23 125.62 r ≤ 1 r ≥ 1 121.30 46.23 r ≤ 2 r ≥ 2 152.94 95.75 r ≤ 2 r ≥ 2 57.06 40.08 r ≤ 3 r ≥ 3 96.87 69.82 r ≤ 3 r ≥ 3 37.24 33.88 r ≤ 4 r ≥ 4 58.63 47.86 r ≤ 4 r ≥ 4 20.87 27.58 r ≤ 5 r ≥ 5 37.76 29.80 r ≤ 5 r ≥ 5 17.70 21.13 r ≤ 6 r ≥ 6 20.06 15.50 r ≤ 6 r ≥ 6 15.40 14.26 r ≤ 7 r ≥ 7 4.67 3.84 r ≤ 7 r ≥ 7 4.67 3.84 Source: Authors’ computation using EVEW-10 Software From the Table 4.4, it was observed that there 8 co-integrating equations at the 0.05 level of significance using the Trace Statistics. Again, from the Maximum Eigen value, there are 4 co- integrating relationship between the dependent variable (MVA) and the covariates. of DCRE, electricity consumption, foreign direct investment, population growth and real gross domestic product. Thus, the trace and maxi-eigen statistic assert the existence of a long-run relationship 37 among the variables. The bound test to co-integration is the major test of this study since we are using the ARDL approach. The results are presented in Table 4.5. Table 4.5: Results of the ARDL Bounds Test Unrestricted Intercept Critical Values F-Statistics 4.26 5% Critical Bound Value Lower 2.79 Upper 3.67 10% Critical Bound Value Lower 2.37 Upper 3.2 1% Critical Bound Value Lower 3.65 Upper 4.66 Source: Authors’ computation using EVIEW-10 (Software) From the results as presented in Table 4.5, and using the standard 5% level of critical value, the E-statistics lies ahead of the lower and upper bounds. Therefore, we accept the alternate hypothesis and reject the null hypothesis and concluded that there is a long-run relationship between MVA and the covariates in line with the Johansen co-integration test of long-run relationship. Having established the long-run relationship among the variables, we proceed to estimate the long-run coefficients based on the ARDL approach is reported in Table 4.6. Table 4.6: Estimated Long-run Coefficients using the ARDL Approach Dependent Variable: LNMVA Variables Coefficient t-Statistic Prob. C -42.80 -4.02 0.0069 LNELCONt-1 0.08 1.809 0.12 LNFDIt-1 1.54 2.51 0.05 LNRGDPt-1 0.00 3.32 0.01 LN MVA 1.25 3.51 0.01 Source: Authors’ computation using EVIEW-10 (Software). The long-run estimates as shown in Table 4.6 indicates that electricity consumption, foreign direct investment and real gross domestic product all have positive long run relationship with manufacturing value added in Nigeria within the reviewing period. This implies that a 1 percent increase in electricity consumption, foreign direct investment and real gross domestic product will lead to 0.08, 1.54 and 0.00 percent increase in manufacturing value added. The results of the short-run dynamics show that the error correction factor is correctly signed and also statistically significant as expected. This showed a rapid rate of adjustment from the short- run disequilibrium to the long-run equilibrium. As can be seen from the result, about 25 percent of the deviation from equilibrium was corrected within one year. The coefficient of determinations shows that the estimated variables with the exception of domestic credit, domestic investment, oil price and population growth removed from the final estimation due to their collinear nature accounted for 0.91 percentage, implying that these variable explained that 91 percent of the variations in the dependent variable (MVA). The adjusted R-Square (0.79) 38 shows that the explanatory variables have explained about 80 percent of the total variations in manufacturing value added in Nigeria. This indicates a high explanatory power of the short-run dynamic model. The Durbin Watson statistic value of 2.62 showed that there’s no first order autocorrelation in the model. The analysis of the short-run estimates show that changes in the previous (one-lagged) period of foreign direct investment and real GDP have positive impact on manufacturing value added while foreign direct investment has a negative impact on manufacturing value added. From the results, foreign direct investment and real GDP were statistically significant at the 1 percent significant. Electricity consumption was not statistically significant in influencing manufacturing value added in the long-run; however, it may be significant impact in the short-run. The short- run dynamic estimates are reported in Table 4.7 using the parsimonious error correction model based in the short-run ARDL version of the estimating equation. Table 4.7: Estimated Coefficients of the Short-run Dynamic Error Correction Model Dependent Variable: D(LNMVA) Variables Coefficient t-Statistic Prob. C. 0.176 3.19 0.0040 D(LN MVA (-1)) -1.81 -5.02 0.002 D(LN ELCON) 0.03 2.19 0.07 D(LN ELCON)-1)) 0.002 0.187641 0.8573 D(LNFDI) 0.750 5.33 0.00 D(LNFDI(-1)) -0.56 -3.36 0.01 D(LN RGDP) -3.26E-05 -0.22 0.83 D(LN RGDP(-1)) 0.000 -3.21 0.01 Cointeq(-1) -0.25 5.95 0.00 R-Square = 0.91 Adjusted R-square = 0.79 Durbin Watson = 2.62 Source: Authors’ computation using EVIEW-10 (Software) This means that a1percentage increase in the previous period of electricity consumption and real GDP leads to 0.002 and 0.00 percent increase in manufacturing value added respectively, in the short-run. While a 1 decrease in foreign direct investment decrease manufacturing value added by 0.56 percent. Further examination of the short-run model shows that changes in electricity consumption and foreign direct investment are positively but insignificantly related to changes in manufacturing value added in the short-run. The implication of this is that a 1 percent increase in electricity consumption and foreign direct investment bring about 0.03 and 0.75 increase in manufacturing value respectively. These results are, however, in conformity with theoretical expectations. It is noteworthy that the positive impact of electricity consumption and foreign direct investment reflects the reforms embarked upon in the energy sector in the most recent times. However, the negative impact of the lagged real GDP could be as a result of the shocks on the Nigerian economy including 2008 financial sector crisis, the 2016 recession, the 2020 induced COVID-19 pandemic and recession which affected the global economy including the Nigeria economy. The implication of this finding especially with reference to real GDP is that in the short-run, shocks on the economy may have adverse effect on manufacturing value added. 39 From empirical analysis, the results of Asaleye (2021) are on the contrary. This study reported a negative relationship with manufacturing value added. In summary, the study concluded that effects of electricity consumption as input in the manufacturing sector have not improved the performance in the sector. This is in line with our insignificant relationship between electricity consumption and manufacturing value added. In the case of foreign direct investment and manufacturing value added, Ido and Taiga (2016) reported a positive but minimal effect of FDI on manufacturing value added, using the vector autoregressive approach. The authors went ahead to recommend infrastructure/power development to improve on the gains of FDI on the manufacturing sector. The positive relationship between FDI and manufacturing value added as revealed by Idoko and Taiga (2016) is in line with our findings. In the case of real GDP and manufacturing value added, Karami, Naser and Karami (2019) found a significant positive relationship between the variables. The authors suggested for economic policies that promotes growth of the manufacturing sector by increases of manufacturing productivity and increases in the manufacturing employment share to create job opportunities. The Granger causality test results are presented in Table 4.8 Table 4.8: Granger Causality Test Results Null Hypothesis Obs F-Statistic Prob. Remark DCRE does not Granger cause MVA MVA does not Granger cause DCRE 33 1.10646 7.77866 0.34 0.00* Reject Accept DINVEST does not Granger cause MVA MVA does not Granger cause DINVEST 34 1.00582 3.69952 0.3781 0.0371 Reject Accept ELCON does not Granger cause MVA MVA does not Granger cause ELCON 37 4.11 6.92 0.0304** 0.046** Accept Accept FDI does not Granger cause MVA MVA does not Granger cause FDI 33 0.0160 0.336 0.9841 0.7171 Reject Reject OILPRICE does not Granger cause MVA MVA does not Granger cause OILPRICE 34 0.747 5.312 0.482 0.01 Reject Accept PGR does not Granger cause MVA MVA does not Granger cause PGR 34 1.714 8.680 0.1978 0.0011 Reject Accept RGDP does not Granger cause MVA MVA does not Granger cause RGDP 34 3.244 1.678 0.05 0.20 Accept Reject Note: Accept at 5% significance. * = Unidirectional causality, ** = Bidirectional causality. Source: EView-10. (Software) Version The Granger causality result as presented above showe4d that there is a unidirectional causality between MVA and domestic credit. In order words, manufacturing value added has the capacity for promoting private sector credit. Ekundayo, Ndubuisi and Ismaila (2018) suggested that credit to the private sector exerts positive impact on manufactured output in the long-run. Furthermore, manufacturing value added showed unidirectional causality with investment at a P-value of 0.04. It is expected that investment whether domestic is expected to improve investment, technological capacities and acceleration of industrial performance in domestic firms (Adeboye, Ojo & Ifeoluwa, 2016). 40 Bidirectional causality was shown between electricity consumption and manufacturing value added and a feedback effect from manufacturing value added to electricity consumption. The result is in contrast to earlier studies (Kassim & Isik, 2020) who showed a unidirectional causality between electricity consumption and manufacturing value added. In addition, manufacturing value added showed causality with population growth and as expected, population growth had no causality with manufacturing value added. Finally, real gross domestic product Granger causes manufacturing value added and not the other way round. This implies a unidirectional causality between real gross domestic product and manufacturing value added. Industrial development has had an important role in the economic growth of countries like China, the Republic of Korea, Taiwan Province of China (Taiwan) and Indonesia. To ensure the reliability of the estimates, model diagnostic tests were carried out to determine the robustness and stability of the model. The model was tested for normality, serial correlation, auto regressive conditional heteroskedasticity and stability using the cumulative sum residual. Table 4.9: Model Diagnostic Tests Test F-Statistic Probability 1. Normality: Jarque-Bera Statistics 0.635086 0.727935 2. Serial Correlation: Breusch-Godfrey Serial Correlation 3.33 0.2131 3. Heteroskedasticity 0.851003 0.6376 4. Stability: CUSUM CUSUM Sq --- 0.05 Source: Authors’ computation using EVEW-10 (Software) version This battery of tests showed that the null hypothesis of no serial correlation is not rejected; the coefficient of the variables are unbiased and could be used for forecasting and that the variables are normally distributed. The graph of the CUSUM tests within the 0.05 percent showed that the variables are stable and that the model exhibits stability over time and can be used for forecasting. Figure 4.1 shows the stability of the model using the CUSUM and the CUSUM SQ. 41 Figure 4.1: The stability diagnostics using the CUSUM and the CUSUM SQ. -8 -6 -4 -2 0 2 4 6 8 2009 2010 2011 2012 2013 2014 CUSUM 5% Significance -0.4 0.0 0.4 0.8 1.2 1.6 2009 2010 2011 2012 2013 2014 CUSUM of Squares 5% Significance Policy Implication of Findings The policy implication of the findings can be espoused as follows: (i) Energy sector performance has positive but insignificant impact on manufacturing productivity in Nigeria. This requires some policy actions for improvement. (ii) External influence through the flow of foreign direct investment has positive and significant impact on manufacturing productivity in Nigeria. Hence, attracted FDI should be sustained since it impacts positively on manufacturing productivity in Nigeria in the immediate to the long term. (iii) Real gross domestic product has positive and significant relationship with manufacturing value added. This implies that manufacturing productivity has the capacity of promoting national income. (iv) Since there’s a bidirectional causality between energy sector performance and manufacturing sector productivity any policy to promote the energy sector will have ripple effect on the manufacturing productivity and vice versa. Conclusion and Policy Recommendations Conclusion This paper examined the relationship and the causality between energy sector performance and manufacturing productivity in Nigeria between the period 1986-2021, using the autoregressive distributed lag and the Granger causality approaches in line with the overarching objectives of this paper. This paper concluded that the Nigerian energy sector performance has significant impact on manufacturing productivity in Nigeria. Furthermore, it was concluded that there is a long-run relationship between energy sector performance and manufacturing productivity. However, bidirectional causality was suggested between energy sector performance and manufacturing productivity. Policy Recommendations This paper suggested the following policies aimed at promoting the energy sector performance and manufacturing productivity in Nigeria. 42 i) The electricity power sector reform needs to be appraised or evaluated for policy action. This will enable the identification of policy bottlenecks. The improvement of the energy sector will also impact positively on manufacturing productivity. ii) Monetary and fiscal policy measures can be used to promote manufacturing productivity. The Central Bank of Nigeria and Bank of Industry can provide credit incentives to industrialist and manufacturers. References Abdelhamid, A.M. & Ahmad, H.E. (2016). The effect of oil price shock on the Saudi Manufacturing sector, Topics in Middle Eastern and African Economics, 18 (2), 185-200. Adeboye, F. B., Ojo, J.A.T. &Ifeoluwa, I.O. (2016). Foreign direct investment and industrial performance in Africa. The Social Sciences 11 (24), 5830-5837. Afolabi, A. & Laseinde, O. T. (2019) Manufacturing sector performance and economic growth in Nigeria. Journal of Physics, 1378: 03206 doi:10.1088/1742-6596/1378/3/032067. Anyanwu, J. C. (2016) The role of human capital in manufacturing value added development in Africa, keynote presentation at the PhD prize award dinner of the 57th Annual Conference of the Nigerian Economic Society, Abuja, 27, September, 2016. Anyanwu, J. C. (2018). Manufacturing value added development in North Africa: Analysis of key drivers. Macroeconomic Policy, Forecasting and Research Department. African Development Bank, Tunisia. Asaleye, A.J, Lawal, A.I., Inegbedion, H.E., Oladipo, A.O., Akinyomade, O.O., Olayemi, M.S. & Igbolekwu, C.O. (2021). Electricity Consumption and Manufacturing Performance: Evidence from Nigeria, International Journal of Energy Economics and Policy, 11(4), 195-201. Asaleye, K. (2021). Electricity consumption and manufacturing sector performance: Evidence from Nigeria, International Journal of Energy Economics and Policy, 11(4), 195 – 201. Banjoko, S. A., Iwuji, I. I. & Bagshaw, K. (202). The performance of the Nigerian manufacturing sector: A 52-year analysis of growth and retrogression (1960 – 2012). Journal of Asian Business Strategy, 2(1), 177 – 191. Dheret, C. &Morossi, M. (2014). Towards a new industrial policy for Europe. EPC Issue Paper No 78, November 2014. Ekundayo, M., Ndubuisi, O. &Ismaila, Y. (2018). Financial development and manufacturing performance: The Nigerian Case, Studies in Business and Economics 13 (1), 97-111. Idoko, C. U. & Taiga, U.U. (2018). Effects of Foreign Direct Investment on Manufacturing Output in Nigeria (1981-2016), Advances in Social Sciences Research Journal, 5 (5), 181-197. Iiyas, M., Ahmed, H. K., Afzah, M. & Mahmood, T.(2010). Determinants of manufacturing value added in Pakistan: An application of bounds testing approach to co-integration Pakistan Economic and Social Review 48(2): 209 – 223. Iwayemi, A. (2018). Nigeria’s dual energy problems: policy issues and challenges. International Association for Energy Economics (6), 17 – 21. Kaldor, N. (1966). Marginal productivity and the macroeconomic theories of distribution: Comment on Samuelson and Modigliani. The Review of Economic Studies, 22(4), 309 – 319. 43 Karami, M. Naser, E. &Karami, Sh. (2019). The effect of manufacturing value added on economic growth: empirical evidence from Europe, Journal of Business Economics and Finance, 8(2), 133-146, Doi: 10.17261/pressacademia.2019.1044. Karami, M., Elahinia, N., & Karami, S. (2019). The effect of manufacturing value added on economic growth: Empirical evidence from Europe. Journal of Business Economics and Finance (JBEF), 8(2): 133 – 146. Doi.org/10.17261/pressacademia.2019.1044. Kassim, F.A. &Isik, A. (2020). Impact of energy consumption on industrial growth in a transition economy: Evidence from Nigeria, MPRA Paper No 101757. Lawal, A. I., Ozturk, I., Olanipekun, I. O. &Asaloye, A. J. (2020) Examining the linkage between electricity consumption and economic growth in African economies, Energy Economics 208: 118363. Lee, D. & Wolpin, K. I. (2006). Intersectorallabour mobility and the growth of the service sector, Econometrica 24(1), 1-46 Li, K. & Lin, B. (2015). The efficiency improvement potential for coal, oil and electricity in China’s manufacturing sector. Energy Economics86: 403 – 413. Liu, D., Ruan, L., Liu, J., Huan, H., Zhang, G., Feng, Y. & Liu, J. (2018). Electricity consumption and economic growth Nexus in Beijing: A causal analysis of quarterly sectoral data. Renewable and Sustainable Energy Reviews, 82: 2498 – 2503. Matsuyama, K. (2008). Structural change in New Palgrave Dictionary of Economics, Second Edition, Durlauf, S. N. &Blume, L. E. (Eds), Palgrave Macmillan, New York. National Bureau of Statistics Report 2021. Naude, N., Szirmai, A., &Haragochi, N. (2016) Structural transformation in Brazil, Russia, India and South Africa (BRICS) (No 016). United Nations University, Maastricht Economic and Social Science Research Institute on Innovation and Technology (MERIT). Nnadozie, O. & Raifu, I. A. (2020). Infrastructure and manufacturing value added in sub- Saharan Africa. Research square. Available at https://doi.org/10.21203/rs.3-rs- 21562/VI. Nnaji, C. E., Ukwueze, E. R. & Chukwu, J. O. (2010). Determinants of household energy choices for cooking in rural areas: Evidence from Enugu State, Nigeria. Continental Journal of Social Sciences 5(2), 1-11. Oyali, E. (2010). The relevance, prospects and the challenges of the manufacturing sector in Nigeria. Department of Civil Technology, Auchi Polytechnic. Oyedopo, S. O. (2012). On energy for sustainable development in Nigeria. Renewable and Sustainable Energy Reviews, 16(5): 2583 – 2598. Pesaran, M. H., Shin, Y., & Smith, R. J. (2001). Bound testing approache to the analysis of level relationships. Journal of Applied Econometrics, 16:289 – 326. Solow, R. M. (1957). A contribution to the theory of economic growth. Quarterly Journal of Economics 70: 65 – 94. Su, D. & Yao, Y. (2017). Manufacturing as the key engine of economic growth for middle- income economics. Journal of the Asia Pacific Economy, 22(1), 47 – 70. Szirmai, A. (2022). Is manufacturing still the main engine of growth in Developing countries? WIDER Angle newsletter, May. http://epsusa.org/publications/newsletter/2009/july 2009 mcdougal.pdf. Szirmai, A., & Verspajen, B. (2015). Manufacturing and economic growth in developing countries, 1950 – 2005. Statistical Changes and Economic Dynamic, 34: 46 – 59. https://doi.org/10.21203/rs.3-rs-21562/VI https://doi.org/10.21203/rs.3-rs-21562/VI http://epsusa.org/publications/newsletter/2009/july%202009%20mcdougal.pdf http://epsusa.org/publications/newsletter/2009/july%202009%20mcdougal.pdf 44 Tkalee, M. & Vizek, M. (2009). The impact of macroeconomic policies on manufacturing production nin Croatia, Privredna Kretanja Ekonomska Politika 121/2009,61 – 92. Udoh, E. & Nsikak, J. (2014). Energy requirement and industrial growth in Nigeria: An ARDL approach in energy resource management in federal system: Challenges, constraints & strategies. Proceedings of the 2013 NAEE/IAEE Conference. UNCTAD (2011). Economic Development in Africa Report 2011: Fostering industrial development in Africa in the new global environment, UNCTAD, available at http://www.unctad.org/en/docs/aldcafrica2011_en.pdf UNCTAD (2013). The state of industrial development in Africa: Unexploited opportunities amidst growing challenges, Policy Brief No 27, UNCTAD, Genera, August. United Nations Industrial Development Organization (UNIDO) (2015) Industrial Development Report 2016: The role of technology and innovation in inclusive and sustainable. Upadhyaya, S. & Kepplinger, D. (2014). However industrial development matters to the well- being of the population-some statistical evidence. Working Paper. Vienna: United Nations Industrial Development Organization. WDI (2019) World Bank Development Indicator 2019 World Bank Development Report, 2020. http://www.unctad.org/en/docs/aldcafrica2011_en.pdf Page 7