Dynamic impact of urbanization, economic growth, energy ...

Dynamic impact of urbanization, economic growth, energy consumption, and trade openness on CO 2 emissions in Nigeria Hamisu Sadi Ali, Siong Hook Law & Talha Ibrahim Zannah

Environmental Science and Pollution Research ISSN 0944-1344 Environ Sci Pollut Res DOI 10.1007/s11356-016-6437-3

1 23

Your article is protected by copyright and all rights are held exclusively by SpringerVerlag Berlin Heidelberg. This e-offprint is for personal use only and shall not be selfarchived in electronic repositories. If you wish to self-archive your article, please use the accepted manuscript version for posting on your own website. You may further deposit the accepted manuscript version in any repository, provided it is only made publicly available 12 months after official publication or later and provided acknowledgement is given to the original source of publication and a link is inserted to the published article on Springer's website. The link must be accompanied by the following text: "The final publication is available at link.springer.com”.

1 23

Author's personal copy Environ Sci Pollut Res DOI 10.1007/s11356-016-6437-3

RESEARCH ARTICLE

Dynamic impact of urbanization, economic growth, energy consumption, and trade openness on CO2 emissions in Nigeria Hamisu Sadi Ali 1 & Siong Hook Law 1 & Talha Ibrahim Zannah 2,3

Received: 26 August 2015 / Accepted: 7 March 2016 # Springer-Verlag Berlin Heidelberg 2016

Abstract The objective of this paper is to examine the dynamic impact of urbanization, economic growth, energy consumption, and trade openness on CO2 emissions in Nigeria based on autoregressive distributed lags (ARDL) approach for the period of 1971–2011. The result shows that variables were cointegrated as null hypothesis was rejected at 1 % level of significance. The coefficients of long-run result reveal that urbanization does not have any significant impact on CO2 emissions in Nigeria, economic growth, and energy consumption has a positive and significant impact on CO2 emissions. However, trade openness has negative and significant impact on CO2 emissions. Consumption of energy is among the main determinant of CO2 emissions which is directly linked to the level of income. Despite the high level of urbanization in the country, consumption of energy still remains low due to lower income of the majority populace and this might be among the reasons why urbanization does not influence emissions of CO2 in the country. Initiating more open economy policies will be welcoming in the Nigerian economy as the openness leads to the reduction of pollutants from the environment particularly CO2 emissions which is the major gases that deteriorate physical environment. Responsible editor: Philippe Garrigues * Hamisu Sadi Ali [email protected]

1

Department of Economics, Faculty of Economics and Management Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia

2

Department of Land Management, Faculty of Agriculture, Universiti Putra Malaysia, 43400 Serdang, Selangor, Malaysia

3

Department of Agricultural Technology, Ramat Polytechnic, P.M.B 1070 Maiduguri, Borno State, Nigeria

Keywords CO2 emissions . Urbanization . Energy consumption . Autoregressive distributed lags (ARDL) . Economic growth . Nigeria

Introduction Man-made activities seriously affected the immediate environment which mostly becomes disastrous considering its negative implications which sometimes lead to total destruction of the natural environment. One of the serious problems facing the world presently is global climate change that mainly resulted from combustion of fossil fuels (coal, oil, and natural gas), which ultimately increases CO2 emissions as well as other form of greenhouse gases that pollute the atmosphere (Intergovernmental Panel on Climate Change 2007). Urbanization is one of the most rapid social phenomena in the present world; this is because large numbers of people are trooping to cities daily for various economic needs and social protection. Under normal circumstances urbanization enhances productivity, open up more economic prospects, create more wealth, and lead to more innovation that reshape arts, science, politics, and other professions (Stewart and Lee 1986; Bloom et al. 2008; Glaeser 2011). Conversely, urbanization accelerates spread of diseases and aggravates other vices of exclusion, crime, poverty, and environmental degradation (Bloom et al. 2008). Based on the United Nations estimate of 2014, it reported that more than 50 % of the world’s population is living in the urban areas and the projection is that the figure will reach 70 % by 2050. It is also estimated that rapid upsurge of urbanization will be concentrated in Asia and Africa as the two regions are expected to record 90 % of the global urbanization growth. India, China, and Nigeria will lead the global urban growth based on 2014 review of the World Urbanization Prospects by United Nations (UN)

Author's personal copy Environ Sci Pollut Res

DESA’s Population Division. The urban growth of these three countries is expected to reach 37 % in the world’s urban population between 2014 and 2050. By 2050, urban population growth in India is projected to reach 404 million, China 292 million, and Nigeria 212 million. Percentage of urban population in Nigeria based on 2013 estimation of the World Bank is 50.84 %. Urbanization is recently recognized as an important determinant of emissions in the existing literature; see for example Hossain (2011), Sharma (2011), and Farhani et al. (2013) among others. They opined that emissions is considered as independent exogenous variable and would be determined by the consumption of energy, trade, income, and urbanization. Increase in the number of people that are living in the urban areas could increase energy consumption which will ultimately enhance the growth of the economy and by that increase the trade with the rest of the world which may eventually increases the level of CO2 emissions in the economy. There are only few studies that examined the dynamic impact of urbanization, economic growth, energy consumption and trade openness on CO2 emissions in the existing literature and no single study attempted to investigate this relationship in Nigeria; therefore, our paper contributed to the existing literature by examining the causal linkage between urbanization, economic growth, energy consumption, and trade openness on CO2 emissions in Nigeria over the period of 1971–2011.

Literature review Studies conducted in this area came up with different findings, for example, recent finding of Farhani and Ozturk (2015) showed that financial development positively influence CO2 emissions and monotonic nexus exist between GDP and CO2 emissions which deviated from environmental Kuznets curve (EKC) hypothesis. It also emphasize that increase in urbanization could lead to more emissions of CO2 in Tunisia. Liu (2009) investigated the relationship among energy consumption, population growth, economic growth, and urbanization in China for the period of 1978–2008 based on ARDL bound testing approach and factor decomposition model. The result suggests that there is single unidirectional causality running from urbanization to overall energy consumption for short and long-run cases. Poumanyvong and Kaneko (2010) applied STIRPAT model and examined the impact of urbanization on energy consumption and CO2 emissions across 99 countries for the period of 1975–2005. The result reveals that impact of urbanization on energy consumption and CO2 emissions differ with the level of development. In case of low income countries, when urbanization increases it leads to the reduction of energy consumption, while it increases energy consumption in middle and high income countries. Urbanization positively influences CO2 emissions for all the income groups, though is more prominent in the middle income countries. Martinez-Zarzoso and

Maruotti (2011) found an inverted U-shaped relationship between urbanization and CO2 emissions as the effect is positive for low urbanization levels. When the countries were grouped based on threshold analysis the result showed that for the two groups a point is identified after which emission-urbanization elasticity become negative and increase in urbanization beyond such point do not increase emissions. For the other group, also it shows that urbanization does not contribute to more emissions rather than affluence and population. Sharma (2011) used a panel data set of 69 countries and applied dynamic panel model and investigated the determinants of CO2 emissions for the period of 1985–2005 by dividing the panel into sub-panels of high income, middle income, and low income. The main result exhibited that trade openness, per capita GDP, and energy consumption positively influences CO2 emissions, while, urbanization negatively affected it for all the sub-panels. For the overall panel, the result shows that urbanization, trade openness, and per capita electric power consumption have negative impacts on CO2 emissions, whereas GDP per-capita and per-capita total primary energy consumption are statistically significant on CO2 emissions. Poumanyvong et al. (2012) applied Stochastic Impacts by Regression on Population, Affluence, and Technology (STIRPAT) technique and examined the impact of urbanization on national residential energy consumption and CO2 emissions across 88 high, middle, and low income countries for the period of 1975–2005. The finding reveals that increase in the urban population decreases residential energy consumption in low income countries, whereas it increases the use of energy in high income countries. In the middle income countries the result indicated that household energy consumption initially falls and then rises with urbanization up to a given threshold at about 70 %. On the other hand, when the sample reduced to 80 countries the result shows that urbanization raises residential emissions in low and middle income countries. The residential emission of high income countries rises at the initial stage and consequently declined with urbanization up to 66 % turning point. Hossain (2011) investigated the dynamic causal relationship among CO2 emissions, energy consumption, economic growth, trade openness, and urbanization across newly industrialized countries for the period of 1971−2007. The causality result reveals no element of longrun relationship, but unidirectional short-run connection exist that run from economic growth and trade openness to CO2 emissions, from economic growth to energy consumption, from trade openness to economic growth, from urbanization to economic growth, and from trade openness to urbanization. It is also found that the elasticity of CO2 emissions in relation to energy consumption is greater than the short-run elasticity, this means that with time when energy consumption increases in the countries under study, the level of emissions will also raises which brings about more environmental pollution. However, with regard to economic growth, trade openness,


and urbanization the environmental quality is established to be normal good in the long-run. Zhang and Lin (2012) applied STIRPAT model across national and regional levels in China and examined the interactions among urbanization, energy consumption, and CO2 emissions for the period of 1995−2010. The outcome shows that urbanization increases energy use and CO2 emissions at national level. At regional level, however, its effect on energy consumption and CO 2 emissions differ through regions as the impact of urbanization on CO 2 emissions in the central region is higher than that of the eastern region. However, the impact of urbanization on energy consumption is greater than that of CO2 emissions in case of eastern region. Hossain (2012) examined the causal relationship among CO2 emissions, energy consumption, economic growth, foreign trade, and urbanization for the period of 1960−2009 in Japan. The result shows that more consumption of energy increase environmental pollution, but economic growth, trade openness, and urbanization does not affect environmental quality in the long-run. Based on his study that examined the impact of income, urbanization, and industrialization on energy intensity across 76 developing countries, Sadorsky (2013) applied mean group estimators and common correlated effects estimators and found that in the long-run, when income increases by 1 %, energy intensity declines by 0.45 to 0.35 %. The long-run urbanization elasticities range from 0.07 % to 0.12 %, while the impact of urbanization on the intensity of energy is varied. Ponce de Leon Barido and Marshall (2014) empirically examined how urbanization affected CO2 emissions and environmental policy across 80 countries for the period of 1983−2005 based on fixed and random effect techniques. The result shows that averagely, the elasticity of urbanization-emission relationship is 0.95 % which means 1 % increase in urbanization could lead to 0.95 % increase of CO2 emissions. Sadorsky (2014) examined the impact of urbanization and industrialization on energy consumption across number of emerging economies. The outcome indicated that increase in income stimulate energy consumption for both short-run and long-run. In the long-run, urbanization decrease energy use while industrialization raises it. Zhao and Wang (2015) empirically examined the relationship among economic growth and energy consumption in China for the period of 1980–2012 based on VECM and granger causality techniques. The result shows that the effect of energy use on urbanization and the effect of urbanization on economic growth appear to be relatively negligible. Furthermore, the granger causality results exhibited bidirectional relationship between energy consumption and economic growth, while unidirectional causality exist running from urbanization to energy consumption and economic growth to urbanization. Akpan and Akpan (2012) investigated the relationship between electricity consumption, carbon emissions and

economic growth in Nigeria for the period of 1970–2008 based on multivariate vector error correction model. The result reveals that in the long-run, increase in the overall economic performance increases emission of CO2 and more consumption of electricity could lead to more CO2 emission. The Granger causality result reveals unidirectional causality that run from economic growth to CO2 emissions. Al-Mulali et al. (2015a, b) based on 129 countries divided by income levels, applied heterogeneous panel for the period of 1980–2011 and investigated the dynamic impact of urbanization, economic growth, trade openness, petroleum consumption, and financial development on CO2 emission. The Pedroni cointegration shows that variables have long-run relationship, the DOLS and Granger causality results reveal that development of the financial sector enhance quality of the environment in both short and long-runs as it reduces CO2 emission. Petroleum consumption is among the main source of environmental degradation in most income groups. Acaravcı et al. (2015) in their study on the relationship between electricity consumption per capita, real gross domestic product (GDP), per capita, trade openness, and per capita foreign direct investment inflows in Turkey for the period of 1974–2013 based on ARDL framework and augmented Granger causality model. The bound test result shows the element of long-run relationship among the variables. Granger causality error correction result shows that unidirectional causality running from electricity consumption per capita to real GDP per capita exist in both short and long-run. The impact of disaggregated renewable electricity production on CO2 emissions was investigated by Al-Mulali et al. (2015a, b) across 23 European countries for the period of 1990–2013 based on panel cointegration technique. Pedroni test shows that CO 2 emission, GDP growth, urbanization, financial development, and renewable electricity production by source have long-run relationship. The FMOLS result reveals that financial development, economic growth, and urbanization increase CO2 emission in the long-run, whereas trade openness reduces it. In addition, renewable electricity that is sourced from combustible renewables and waste, nuclear power and hydroelectricity affected CO 2 emission in the long-run, while renewable energy sourced from solar and wind powers is insignificant. The VECM Granger causality reported that only economic growth caused CO2 emission in the long-run in the entire models, while the other variables only caused CO2 emission only in few models. Almulali and Ozturk (2015) used data of 14 MENA countries and investigated the causes of environmental degradation in the region based on panel cointegration techniques for the period of 1996–2012. Pedroni cointegration result reveals an element of long-run relationship among ecological footprint, energy consumption, urbanization, trade openness, industrial development, and political stability. Furthermore, fully modified ordinary least square


result reported that energy consumption, urbanization, trade openness, and industrial development lead to environmental degradation, whereas stability in the political system reduces it. The Granger causality result shows that the variables have both short and long-run causal relationships with the ecological footprint. Ben Jebli et al. (2015) applied panel cointegration technique across 24 sub-Saharan Africa countries for the period of 1980–2010. The short-run Granger causality shows the existence of bidirectional causal relationship between CO2 emissions and economic growth; bidirectional causality between CO2 emissions, and real exports; unidirectional causality that run from real imports to emissions and one way causality running from trade openness to renewable energy consumption. In general, the long-run result does not support EKC hypothesis in the countries investigated. Ozturk (2015) examined the dynamic links between energy consumption, air pollution, and climate change across six panel of industrialized and transition economies classified by United Nations Framework Convention on Climate Change (UNFCC) for the period of 1990–2012. The result based on pooled square regression reveals that energy consumption and air quality variables positively influence climate change. The finding based on fixed and random effect regression confirmed the earlier finding by pooled square regression, while the result did not show any substantial impact between energy consumption and climate change when controlling for countryspecific and time variant shocks for the study period. Shahbaz et al. (2014) examined the relationship between economic growth, electricity consumption, urbanization, and environmental degradation for the period of 1975–2011 based on ARDL technique for United Arab Emirates (UAE). The result shows that variables have long-run relationship, the relationship between economic growth and CO2 emissions is found to be inverted U-shaped, at the initial stage growth of the economy increase energy consumption up to certain threshold level of per-capita income where it reduce it. Consumption of electricity reduces CO2 emissions and urbanization increase CO2 emissions; export enhances quality of environment through reduction of CO2 emissions. The Granger causality result reconfirmed feedback effect between CO2 emission and consumption of electricity and economic growth and urbanization causes CO2 emission. Balibey (2015) investigated the dynamic impact of economic growth, CO2 emissions and foreign direct investment by evaluating environmental Kuznets curve hypothesis for Turkey in 1974–2011 based on Johansen cointegration, Granger Causality tests and impulse-response and variance decomposition analysis of vector autoregression model (VAR) models. The result based on causality reveals that FDI and economic growth have positive significant effect on CO2 emissions. Impulse response function and variance-decompositions results model backed the relationship among GDP, FDI, and CO2.

Theoretical framework In this paper, we considered both Stochastic Impacts by Regression on Population, Affluence, and Technology (STIRPAT) and economic theories that envisage Environmental Kuznets Curve (EKC) with regard to income and the sociological theories that related EKC to urbanization not only on economic development. The earlier works of Dietz and Rosa (1994, 1997) brought the idea of formulation stochastic type of the IPAT equation which contain quantitative variables of population size (P), affluence per-capita (A), as well as weight of the industry in economic dealings as a measure of environmentally damaging technology (T). The single year model specification is given by the following equation; β

I i ¼ αPi 0 Aβi 1 T βi 2 ei

ð1Þ

Where; Ii Pi Ai and Ti represent the effect of population and technology in country i, α and β’s are estimated parameters and ei refers to random error term. STIRPAT is being extensively applied to examine the factors that affect environment (see for example Dietz and Rosa 1997; York et al. 2003; Cole and Neumayer 2004). The hypothesis is that production of CO 2 is centered on demographics but can change when effectiveness of living in urban areas is attained. Specifically, the economic activities in urban areas may have two distinct impacts, those related to higher incomes and higher consumption intensities of industrialization. Initially, urbanization enhances the shift to current fuels that modify the patterns of resources used.

The data Data for this study was obtained from World Development Indicators (WDI CD ROM, 2015), World Bank for the period of 1971–2011. CO2 emissions is measured by the CO2 emissions in kilo terms (kt), urbanization is measured by urban population as a percentage of total population, economic growth (Y) is proxied by real GDP, energy consumption is proxied by fossil fuel consumption which comprised of coal, oil, petroleum, and natural gas products, while trade openness is measured by the total export and import as a percentage of GDP.

Specification of the model Following Hossain (2011) and Farhani et al. (2013), we incorporated urbanization into the emission regression as opined that CO2 emissions will be determined by trade, income,


energy consumption, and urbanization, our base line model is therefore specified below; CO2 ¼ A α1 UB α2 Y α3 EC α4 T O

ð2Þ

Taking the natural logarithm of Eq. (2) above lead us to the following equation; lnCO2 ¼ α0 þ α1 lnUB þ α2 lnY þ α3 lnEC þ α4 lnTO

ð3Þ

Based on the above two equations as urbanization variables introduced into the emissions equation we proceed to formulate our autoregressive distributed lags (ARDL) that will be estimated in order to find the links among the variables under investigation as shown in Eq. (4) below; ΔlnCO2t ¼ α0 þ α1 lnCO2t−1 þ α2 lnUBt−1 þ α3 lnYt−1 þ α4 lnEC t−1 þ α5 lnT Ot−1 Xp Xq þ γ ΔlnCO2t−1 þ δ ΔlnUBt− j i i j j þ þ

Xp

φ ΔlnY t−l þ l l

Xq

η ΔlnTOt−n m m

lnCO2t ¼ α0 þ

Xp

þ

ω ΔlnEC t−m l k

þ εt

ð4Þ

Where; lnCO2 refers to log of CO2 emissions, lnUB is the log of urbanization, lnY is the log of economic growth, Table 1 ADF and PP unit root tests results from 1971 to 2011

lnEC is the log of energy consumption, and lnTO is the log of trade openness, subscript t denote time period. Initially, we need to estimate Eq. (4) based on ordinary least square (OLS) and test the joint significance of the coefficients of lagged variables based on Wald test of Ftest with the aim of observing the long-run relationship among the variables. The null hypothesis of no cointegration (H0 = α1 = α2 = α3 = α4 = α5 = 0) will be tested against alternate hypothesis of cointegration (Ha ≠ α1 ≠ α2 ≠ α3 ≠ α4 ≠ α5 ≠ 0). The decision rule as suggested by Pesaran et al. (2001) is that if the estimated F-test value is higher than the upper bound critical value, null hypothesis will be rejected which signifies the presence of long-run relationship. However, if the estimated value is less than the lower critical value, null hypothesis cannot be rejected which means no element of cointegration exist among the variables, while if the value falls within the lower and upper bound critical bounds the result is inconclusive (Pesaran & Pesaran 1997). We use Eq. (5) below in order to test the long-run coefficients of ARDL;

Variables

Level lnCO2t

−2.438 −1.177 lnUBt 1.605 lnYt −5.145* lnECt −2.617* lnTOt First difference −6.694* lnCO2t lnYt lnECt lnTOt

−2.685** −5.223* −4.430* −7.996*

γ lnCO2t−1 i¼1 i

X q2

φ lnY t−l þ i¼0 l

X q4

X q1

δ lnUBt− j j¼0 j

X q3

η lnEC t−m m¼1 m

∪ lnT Ot−m þ εt

ð5Þ

m¼0 n

To select lag length of the model, Schwarz Bayesian Criterion (SBC) is chosen and applied error correction

ADF Constant without trend

lnUBt

þ

Xp

PP Constant with trend

Constant without trend

Constant with trend

−2.583 −2.652 −0.099 −3.613** −2.908

−2.438 −1.147 −1.587 −5.023* −2.617

−2.583 −1.225 −0.129 −3.754** −2.908

−6.866*

−7.111*

−6.992*

−2.043** −5.745* −5.075* −8.090*

−1.759** −5.324* −4.346* −7.944*

−2.408** −5.730* −5.075* −8.032*

The ADF and PP test equations include both constant and trend terms. The Schwarz information criterion (SIC) is used to select the optimal lag order in the ADF test equation The values in brackets are corresponding p values *denote significance level at 1 %, **5 %, and ***10 %, respectively

Author's personal copy Environ Sci Pollut Res Table 2

ARDL bound test estimation result

Model for estimation

Lag Flength statistics

4

F CO2 ðUBjYjECjTOÞ

7.45

Table 4 Estimated short-run coefficients from error correction model based on SBC Dependent variable (ΔlnCO2t)

Significance Critical level bound F-statistics I(0) I(1) 1%

4.42 6.25

5% 10 %

3.20 4.54 2.66 3.83

Regressors

Coefficients

T-ratio (p values)

ΔlnCO2t

0.279 0.486 0.017 0.439

1.515 (0.141) 2.952 (0.006)* 0.116 (0.908) 2.237 (0.033)**

1.356 −0.257

2.851 (0.008)* −1.955 (0.061)***

−0.649

−3.933 (0.001)

ΔlnCO2t − 1 ΔlnUBt ΔlnYt ΔlnECt

Critical values are obtained from Narayan (2005) (Table Case III: Unrestricted intercept and no trend; pg. 1988)

ΔlnTOt ECM(-1)

*, **, and *** denote significant at 1 %, 5 %, and 10 % levels, respectively.

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

model with a view to determine short-run dynamics of the variables; ΔlnCO2t ¼ α0 þ þ þ

Xp

Xq

i

γ i ΔlnCO2t−1

δ ΔlnU Bt− j þ j j

Xq

η ΔlnEC t−m m m

þ ϑecmt−1 þ εt

Xq i

þ

φl ΔlnY t−l

Xq

∪ m n

at first differences which means are I(1) variables. Therefore, with the order of integration of the variables, we are at liberty to run ARDL as Pesaran et al. (2001) suggested that ARDL is suitable with the combination of I(0) and I(1) variables.

Result of ARDL cointegration test

ΔlnT Ot−n ð6Þ

Following Pesaran (1997) the stability of the long-run coefficient together with short-run dynamics is tested based on cumulative sum of recursive residuals (CUSUM) and cumulative sum of squares of recursive residuals (CUSUMSQ).

Estimation results In order to test the order of integration of the variables and observe if the possibility of running ARDL is feasible or not, we tested Augmented Dickey Fuller (ADF) and PhillipsPerron (PP) unit root tests. The results indicated that energy consumption and trade openness are stationary at level means are I(0) variables, while CO2 emissions, urbanization, and economic growth are non-stationary at levels, but stationary Table 3 Estimated long-run coefficients based on Schwarz Bayesian Criterion (SBC) Dependent variable (lnCO2t) Regressors

Coefficients

T-ratio (p values)

lnUBt lnYt lnECt lnTOt Constant

0.026 0.677 0.808 −0.395 −7.138

0.115 (0.909) 2.898 (0.007)* 1.912 (0.067)*** −0.052 (0.050)*** −1.819 (0.080)

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

The works of Pesaran and Shin (1999) as well as Pesaran et al. (2001) makes ARDL one of the most widely accepted and used methodology considering its numerous advantages over other cointegration techniques e.g., Engle and Granger (1987), Johansen (1988), and Johansen and Juselius (1990) techniques. Some of the advantages of ARDL includes (a) order of integration do not necessarily be the same, (b) ARDL is possible to run with small sample, and (c) multiple equations is not necessary because reduced-form equation may provide the same outcome (Ozturk and Acaravci 2010, 2013). Despite its shortcomings ARDL is still among the best time series technique widely used in economics and econometrics analysis (Table 1). Table 2 below reported the ARDL cointegration result which indicated that the estimated F-statistic (7.45) is greater than upper critical bound as obtained in Narayan (2005) table of hypothesis rejection for ARDL small sample. This means that variables were cointegrated as null hypothesis is rejected Table 5

ARDL diagnostic test results

Test statistics

LM version

F-version

1: Serial correlation 2: Functional form 3: Normality 4: Heteroscedasticity

CHSQ(1) = 0.635[0.426] CHSQ(1) = 1.116[0.291] CHSQ(2) = 3.682[0.159] CHSQ(1) = 2.272[0.132]

F(1, 26) = 0.467[0.500] F(1, 26) = 0.832[0.370] N/A F(1, 34) = 2.291[0.139]

* denotesx significance level at 1 %, *5 %, and ***10 %, respectively 1: Langrange multiplier test of residual serial correlation 2: Ramsey’s misspecification test using squarer of the fitted values 3: Jacque-Bera test for normality 4: Autoregressive conditional heteroskedasticity test

Author's personal copy Environ Sci Pollut Res Fig. 1 Plot of cumulative sum of recursive residuals

15 10 5 0 -5 -10 -15 1976

1981

1986

1991

1996

2001

2006

2011

The straight lines represent critical bounds at 5% significance level

at all significance levels but we stick on 1 % since is the best significance level statistically. The long-run relationship among the variables is established and gave us a chance to estimate our Eq. (5) in order to obtain long-run coefficients as shown in the Tables 3 and 4 below. The long-run result as reported in the above Table 3 shows that urbanization has positive insignificant impact on CO2 emissions which means urbanization does not have any significant impact on CO2 emissions in Nigeria, this result reconfirmed the findings of Martinez-Zarzoso and Maruotti (2011) based on threshold analysis of developing countries which shows that in certain level, CO 2 emissionurbanization elasticity is negative and when urbanization increases beyond such level it does not stimulate higher CO2 emissions, and the finding also reconfirmed that of Hossain (2012) who claimed that urbanization does not have any significant impact on CO2 emissions in Japan. But this result contradict the recent finding of Zhang and Lin (2012) and Farhani and Ozturk (2015) whose claimed that increase in urbanization leads to more CO2 emissions in China and Tunisia, respectively. One of the probable reasons that urbanization does not have impacts on CO2 emissions in Nigeria might be related to income distribution in the country as well as poverty level. For example, the Nigeria’s National Bureau of Statistics (2012) reported that 65 % of the population is poor in 2010, therefore might not have adequate resources to consume substantial amount of energy, hence energy consumption is low and CO2 emissions is also low, thus growth of urban population will not have significant impact on CO2 emissions. Economic growth has positive significant impact on CO2 emissions as 1 % increase in economic growth could Fig. 2 Plot of cumulative sum of squares of recursive residuals

lead to 0.6 % increase in CO 2 emissions; the result reconfirmed the findings of Akpan and Akpan (2012) and Al-Mulali et al. (2015a, b). Energy consumption also positively influenced CO2 emissions because based on the result 1 % increase in energy consumption could lead to 0.8 % increase in CO2 emissions; this reconfirmed the outcome found by Hossain (2012). However, trade openness negatively affected CO2 emissions as 1 % increase in trade openness could reduce CO2 emissions by 0.3 % in Nigeria, this finding therefore corroborated that of Al-Mulali et al. (2015a, b). The short-run dynamics reveals that urbanization does not have any substantial effect on CO2 emissions as the case of the long-run, economic growth, and energy consumption positively increases CO2 emissions as in the long-run, while trade openness also maintain its stand of reducing CO2 emissions also as in the long-run. As theoretically expected our ECM is negative, statistically significant at 1 % and less than one, the values explain that any deviation from the long-run equilibrium among the variables will be corrected and converge to long-run equilibrium level at 65 % annually. To justify the consistency and efficiency of our model, we conducted diagnostic tests as reported in the Table 5 above which shows that our model passed all the four (4) time series problems of serial correlation, functional form, normality, and heteroscedasticity. For all the four (4) tests results, null hypothesis cannot be rejected which established the fact that the modelling problems are all overcome by our model. The stability of the model is shown by cumulative sum of recursive residuals (CUSUM) and cumulative sum of squares of recursive residuals (CUSUM of squares) in Figs. 1 and 2, respectively, as the blue lines lies within the

1.5 1.0 0.5 0.0 -0.5 1976 1981 1986 1991 1996 2001 2006 2011 The straight lines represent critical bounds at 5% significance level


critical bounds and is significant at 5 % which confirmed that our model is highly stable over the sample period.

Conclusion and policy recommendations This article examined whether increase in the number of people that are living in urban areas in Nigeria could stimulate more emissions of CO2 for the period of 1971–2011. The cointegration result signified the existence of long-run relationship among the variables and we rejected null hypothesis at 1 % significance level. The main finding in the long-run reveals that urbanization does not have any significant impact on CO2 emissions in Nigeria for the sample period. While energy consumptions and economic growth positively influence CO2 emissions, trade openness negatively affected it. The trend of this relationship remains the same even in the short-run dynamics of the variables. The policy implication is that policy makers should provide more eligible policies that will make the economy more open and integrated to the rest of the world in order to reduce environmental pollution in the country. Therefore, based on this research outcome the policy makers should not give more attention to urbanization when considering policies to overcome pollution problems as the increase in urban population has nothing to do with increase in CO2 emissions maybe due to lower income levels of most peoples in the country. Implementing favorable policies that will reduce the level of carbon emissions is paramount considering the enormous dangers caused by incessant emission due to the consumption of energy, renewable sources of energy could among the best alternative considering its prices as well less environmental damages.

References Acaravcı A, Erdogan S, Akalın G (2015) The electricity consumption, real income, trade openness and FDI: the empirical evidence from Turkey. Int J Energy Econ Policy 5(4):1050–1057 Akpan GE, Akpan UF (2012) Electricity consumption, carbon emissions and economic growth in Nigeria. Int J Energy Econ Policy 2(4):292–306 Al-Mulali U, Ozturk I (2015) The effect of energy consumption, urbanization, trade openness, industrial output, and the political stability on the environmental degradation in the MENA (Middle East and North African) region. Energy 84:382–389 Al-Mulali U, Ozturk I, Lean HH (2015a) The influence of economic growth, urbanization, trade openness, financial development, and renewable energy on pollution in Europe. Nat Hazards 79(1):621–644 Al-mulali U, Tang CF & Ozturk I (2015) Does financial development reduce environmental degradation? Evidence from a panel study of 129 countries. Environ Sci Pollut Res 1–10 Balibey M (2015) Relationships among CO2 emissions, economic growth and foreign direct investment and the EKC hypothesis in Turkey. Int J Energy Econ Policy 5(4):1042–1049

Ben Jebli M, Ben Youssef S, Ozturk I (2015) The role of renewable energy consumption and trade: environmental Kuznets curve analysis for sub-Saharan Africa countries. Afr Dev Rev 27(3):288–300 Bloom DE, Canning D, Fink G (2008) Urbanization and the wealth of nations. Science 319(5864):772–775 Cole MA, Neumayer E (2004) Examining the impact of demographic factors on air pollution. Popul Dev Rev 2(1):5–21 Dietz T, Rosa EA (1994) Rethinking the environmental impact of population, affluence and technology. Hum Ecol Rev 1:277–300 Dietz T, Rosa EA (1997) Effects of population and affluence on CO2 emissions. Proc Natl Acad Sci USA 94(1):175–179 Engle RF, Granger CWJ (1987) Co-integration and error correction: representation, estimation, and testing. Econometrica 55:251–276 Farhani S, & Ozturk I (2015) Causal relationship between CO2 emissions, real GDP, energy consumption, financial development, trade openness, and urbanization in Tunisia. Environ Sci Pollut Res 1–14 Farhani S, Shahbaz M, Arouri M (2013) Panel analysis of CO2 emissions, GDP, energy consumption, trade openness and urbanization for MENA countries. MPRA Paper No. 4925 Glaeser E (2011) Cities, productivity, and quality of life. Science 333:592–594 Hossain MS (2011) Panel estimation for CO2 emissions, energy consumption, economic growth, trade openness and urbanization of newly industrialized countries. Energy Policy 39:6991–6999 Hossain S (2012) An econometric analysis for CO2 emissions, energy consumption, economic growth, foreign trade and urbanization of Japan. Low Carbon Econ 2012(3):92–105 Intergovernmental Panel on Climate Change (IPCC) (2007) Climate Change 2007 Synthesis report. IPCC, Geneva Johansen S (1988) Statistical analysis of cointegrating vectors. J Econ Dyn Control 12:231–254 Johansen S, Juselius K (1990) Maximum likelihood estimation and inference on cointegration—with application to the demand for money. Oxf Bull Econ Stat 52:169–210 Liu Y (2009) Exploring the relationship between urbanization and energy consumption in China using ARDL (autoregressive distributed lag) and FDM (factor decomposition model). Energy 34(11):1846–1854 Martinez-Zarzoso I, Maruotti A (2011) The impact of urbanization on CO2 emissions: evidence from developing countries. Ecol Econ 70:1344–1353 Narayan PK (2005) The saving and investment nexus for China: evidence from cointegration tests. Appl Econ 37:1979–1990 National Bureau of Statistics—NBS (2012) Annual abstract of statistics. NBS Press, Abuja Ozturk I (2015) Measuring the impact of energy consumption and air quality indicators on climate change: evidence from the panel of UNFCC classified countries. Environ Sci Pollut Res 22(20): 15459–15468 Ozturk I, Acaravci A (2010) CO2 emissions, energy consumption and economic growth in Turkey. Renew Sustain Energy Rev 14:3220–3225 Ozturk I, Acaravci A (2013) The long-run and causal analysis of energy, growth, openness and financial development on carbon emissions in Turkey. Energy Econ 36:262–267 Pesaran MH (1997) The role of economic theory in modelling the long run. Econ J 178–191 Pesaran MH, Pesaran B (1997) Working with microfit 4.0: interactive econometric analysis; [Windows version]. Oxford University Press, Oxford Pesaran HM, Shin Y (1999) Autoregressive distributed lag modelling approach to cointegration analysis. In: Storm S (ed) Econometrics and economic theory in the 20th century: the Ragnar Frisch centennial symposium. Cambridge University Press, Cambridge Pesaran MH, Shin Y, Smith RJ (2001) Bounds testing approaches to the analysis of level relationships. J Appl Econ 16:289–326

Author's personal copy Environ Sci Pollut Res Ponce de Leon Barido D, Marshall JD (2014) Relationship between urbanization and CO2 emissions depends on income level and policy. Environ Sci Technol 48(7):3632–3639 Poumanyvong P, Kaneko S (2010) Does urbanization lead to less energy use and lower CO2 emissions? A cross-country analysis. Ecol Econ 70(2):434–444 Poumanyvong P, Kaneko S & Dhakal S (2012) Impacts of urbanization on national residential energy use and CO2 emissions: evidence from low-, middle- and high-income countries (No. 2-5). Hiroshima University, Graduate School for International Development and Cooperation (IDEC) Sadorsky P (2013) Do urbanization and industrialization affect energy intensity in developing countries? Energy Econ 37:52–59 Sadorsky P (2014) The effect of urbanization and industrialization on energy use in emerging economies: implications for sustainable development. Am J Econ Sociol 73(2):392–409 Shahbaz M, Sbia R, Hamdi H, Ozturk I (2014) Economic growth, electricity consumption, urbanization and environmental degradation relationship in United Arab Emirates. Ecol Indic 45:622–631

Sharma SS (2011) Determinants of carbon dioxide emissions: empirical evidence from 69 countries. Appl Energy 88(1):376–382 Stewart Jr. CT, Lee JH (1986) Urban concentration and sectoral income distribution. J Developing Areas 357–368 United Nations (2014): http://www.un.org/en/development/desa/news/ population/world- urbanization-prospects-2014.html United Nations, Department of Economic and Social Affairs, Population Division (2014).World Urbanization Prospects: The 2014 Revision, Highlights. (available at http://esa.un.org/unpd/wup/Highlights/ WUP2014-Highlights.pdf) York R, Rosa EA, Dietz T (2003) STIRPAT, IPAT and IMPACT: analytic tools for unpacking the driving forces of environmental impacts. Ecol Econ 46(3):351–365 Zhang C, Lin Y (2012) Panel estimation for urbanization, energy consumption and CO 2 emissions: a regional analysis in China. Energy Policy 49:488–498 Zhao Y, Wang S (2015) The relationship between urbanization, economic growth and energy consumption in China: an econometric perspective analysis. Sustainability 7(5):5609–5627