• 沒有找到結果。

The causal relationship between energy resources and economic growth in Brazil

N/A
N/A
Protected

Academic year: 2021

Share "The causal relationship between energy resources and economic growth in Brazil"

Copied!
9
0
0

加載中.... (立即查看全文)

全文

(1)

The causal relationship between energy resources and economic

growth in Brazil

Hsiao-Tien Pao

a,n

, Hsin-Chia Fu

b a

Department of Management Science, National Chiao Tung University, Taiwan, ROC

bCollege of Engineering, Huaqiao University, Quanzhou, China

H I G H L I G H T S

 We model three kinds of clean energy and non-clean energy consumption and real GDP.  There is fossil fuel consumption–economic growth bidirectional causality.

 There is new renewables consumption–economic growth bidirectional causality.  There is nuclear energy consumption–economic growth bidirectional causality.  Substitutability exists for new renewables–fossil fuel or new renewables–nuclear.

a r t i c l e i n f o

Article history: Received 7 April 2012 Accepted 23 June 2013 Available online 19 July 2013 Keywords:

Renewable Nuclear Granger causality

a b s t r a c t

This study investigates the causal relationship between clean and non-clean energy consumption and economic growth in Brazil over the period of 1980–2009. Clean energy consumption at aggregated level of total renewable energy consumption and disaggregated levels of hydroelectric, new renewables, and nuclear energy consumption are tested within a production function framework. A cointegration test reveals a long-term equilibrium relationship between real output, capital, labor, and renewable and non-renewable energy consumption at aggregated level, and a long-term equilibrium relationship between real output, capital, labor, and hydroelectric/new renewables/nuclear and fossil fuel energy consumption at disaggregated level. The capital, labor, and new renewables elasticities of real output are positive and statistically significant, other energy consumption item's elasticities are insignificant. The results from error correction model reveal the interdependencies between new renewables, nuclear, fossil fuel, and total non-renewable energy consumption and economic growth, the unidirectional causality from hydroelectric/total renewable consumption to economic growth, the substitutability between new renewables and fossil fuel consumption, and the substitutability between new renewables and nuclear energy consumption. Additionally, nuclear and new renewables energy consumption responds to bring the system back to equilibrium. Overall, aggregated analysis may obscure the relationship between different types of clean energy consumption and economic growth.

& 2013 Elsevier Ltd. All rights reserved.

1. Introduction

According to the 2010 International Energy Outlook released by the US Energy Information Administration (EIA), worldwide renewable energy consumption has been increasing at a rate of 2.6% per year. In 2008, approximately 19% of the global energy consumption was from renewable sources, 13% of which was from traditional biomass (mainly used for heating), 3.2% from hydro-electricity, and the remaining 2.7% from rapidly growing ‘new renewables’ (e.g., small hydro, modern biomass, wind, solar,

geothermal, and biofuels). Renewable energy power generation makes approximately 18% of the global electricity, with 15% from hydropower and 3% from other new renewable energy sources. ‘New renewables’ technologies are befitting for local electricity generation in rural and remote areas, where the transportation costs for crude oil or natural gas and the transmission costs of electricity are often prohibitively high. Globally, 3 million house-holds are estimated to receive power from small solar PV systems. Micro-hydropower systems configured into village-scale or county-scale mini-grids are emerging in many areas. More than 30 million rural households use family-sized biogas digesters for lighting and cooking. Biomass cookstoves have been used by 160 million households (Wikipedia, 2011). In addition, expected increases in oil prices, increased awareness of the environmental Contents lists available atScienceDirect

journal homepage:www.elsevier.com/locate/enpol

Energy Policy

0301-4215/$ - see front matter& 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.enpol.2013.06.113

nCorresponding author. Tel.:+886 3 5131578; fax: +886 3 5710102.

(2)

damage caused by fossil fuel consumptions, and government incentives for new renewables energy development will continue to foster the global usage of new renewables energy. These‘new renewables’ can provide approximately 6% of worldwide electri-city by 2030.

Although environmentalists have warned that catastrophic climate change is a real and imminent danger, we still need a large-scale source of around-the-clock electricity to meet our energy needs. Nuclear energy can generate electricity with no carbon dioxide or other greenhouse gas emissions, and it is the only effective option in order to supply the large demand for clean electricity on a global scale. Currently, nuclear power plants supply approximately 6% of the global energy and 14% of the global electricity needs. Nuclear power and new renewables will be urgently needed as partners if the world's enormous demand for clean energy is to be met (World Nuclear Association (WNA), 2011). The use of new renewable energy and nuclear energy is critical to the development of global clean energy economy in the future, due to the continuous depletion of reserve of earth's fossil fuel as well as global warming.

The topic of causal relationship between energy consumption and economic growth has been well-studied in energy economics literature such asOzturk (2010)andPayne (2010a,2010b). How-ever, the empirical results may be varied and even conflicted, due to the difference in country′ characteristic, time period, econo-metric methodology, or proxy variables for energy consumption and income. The causality between energy consumption and economic growth in different directions may have different policy implications. Under the assumption of positive correlation between energy consumption and economic growth, the presence of unidirectional causality from energy consumption to economic growth or bidirectional causality between them would suggest that energy conservation policies that reduce energy consumption may lead to decline in economic growth. In contrast, unidirec-tional causality from economic growth to energy consumption or no causality in either direction suggests that energy conservation policies will have little or no impact on economic growth (Apergis and Payne, 2013).

Among the literature of energy consumption and economic growth nexus, some studies also examined the relationships between different types of clean energy consumption (renewable or nuclear) and economic growth (Pao and Fu, 2013). Recent researches byApergis and Payne (2011b,2011c)focus on the link between both renewable and non-renewable energy consumption and economic growth for sustainable economic development. This paper extends recent works on the energy consumption–growth nexus to analyze the relationships between both the clean and non-clean energy consumption and economic growth in sustain-able countries such as Brazil. Brazil is one of the fastest-growing major economies in the world, with an annual growth rate of GDP of approximately 5%, and is expected to become one of the world's topfive economies in the future. In the past 2 decades, Brazil has achieved a development model that combines social inclusion with sustained economic growth and balanced use of natural resources. This model can maintain high levels of renewable energy to stimulate economic growth and lift millions of people out of poverty, while protecting the country's forests and biodi-versity (Secretariat of Social Communication (SECOM), 2012). According to the 2009 EIA, Brazil's renewable energy consumption reached 97% of its total domestic electricity generation, and the growth rate of different types of energy consumption are varies. During the 1980–2009 period, new renewable energy consump-tion (i.e., non-hydroelectric renewable energy consumpconsump-tion, NHREC) with a very high annual average growth rate of 8.72% accounted for 2.89% of the total renewable energy consumption (TREC), while hydroelectric energy consumption (HEC) with an

annual average growth rate of 3.66% accounted for 97.11% of TREC. Additionally, nuclear energy consumption (NUCEC), with the high-est annual average growth rate of 19.65%, accounted for 1.20% of the total non-renewable energy consumption (TNREC), while fossil fuel consumption (FFC) with the lowest annual average growth rate of 2.85% accounted for 96.67% of TNREC. Currently, Brazil is one of the world's cleanest energy matrices. A country with high growth rate or high proportion of clean energy consumption may imply an interdependent relationship between economic growth and clean energy consumption or substitutability between the clean and non-clean energy sources to achieve sustainable economy.

Due to the greatly different growth rates of the various types of energy sources, this study focuses on the disaggregated analysis of the causal relationship between clean energy (hydroelectric, new renewables, and nuclear) consumption and economic growth in Brazil over the period of 1980–2009 since aggregated analysis may well mask the differential impact of hydroelectric, new renew-ables, and nuclear energy consumption on economic growth. The results are compared with the aggregated analysis of the causal relationship between total renewable energy consumption and economic growth. The simultaneous use of clean and non-clean energy consumption in the production function framework intends to distinguish the relative influence of each type on economic growth and to analyze the substitutability between the different types of energy sources. The neo-classical one-sector aggregated production model is adopted where capital, labor, clean energy consumption, and non-clean energy consumption are treated as separate inputs. Within this framework, a vector error-correction model (VECM) is employed to test for multivariate cointegration and Granger causality.

This study is organized as follows. Section 2provides a brief literature review. Section 3 describes the analytical model and econometric methodology.Section 4presents relevant energy and economic data and also presents the cointegration and Granger causality results.Section 5presents the conclusions.

2. Literature review

Since the dawn of the 21st century, the relationships between non-renewable, renewable, or nuclear energy consumption and economic growth have been researched upon (Payne, 2010a,

2010b). Common methodologies include the forecast error var-iance decomposition analysis model, the bivariate error correction model, the Toda–Yamamoto procedure within a production func-tion framework, and the multivariate error correcfunc-tion model within a production function framework. Using a generalized forecast error variance decomposition analysis, Sari and Soytas (2004) found that different energy consumption items have different effects on real output, where lignite, waste, oil, and hydraulic power are the top four alternative energy sources in Turkey. For the US,Ewing et al. (2007)found that coal, natural gas, and fossil fuel energy sources unexpectedly have the largest impacts on the variation of real output, while several renewable energy sources also exhibit considerable impacts. Using a bivariate panel error correction model,Sadorsky (2009)presented evidence of bidirectional causality between non-hydroelectric renewable energy consumption and economic growth for a panel of 18 emerging economies. Using the Toda–Yamamoto procedure within a production function framework for analyzing data of the US,

Payne (2009)found no evidence of a causal relationship between total renewable and non-renewable energy consumption and real GDP; Payne (2011b) provided a disaggregated analysis of the causal relationship between fossil fuel (coal, natural gas, and petroleum) consumption and real GDP, and their results showed

(3)

that different energy consumption items have different effects on real output; Payne (2011a) found unidirectional causality from biomass energy consumption to real GDP; Bowden and Payne (2010) explored the causal relationship between renewable and non-renewable energy consumption by sector (commercial, indus-trial, and residential) and real GDP in the US, and their results were inconsistent. Using a multivariate panel error correction model within a production function framework, Apergis and Payne (2010a) found evidence of bidirectional short- and long-run causality between non-hydroelectric renewable energy consump-tion and economic growth for a panel of 20 OECD countries;

Apergis and Payne (2010c)found both short- and long-run uni-directional causality from energy consumption to economic growth for a panel of nine South American countries; Apergis and Payne (2010b,2011a,2012c)discovered evidence of bidirec-tional short- and long-run causalities between total renewable energy consumption and economic growth for a panel of 13 Eurasian countries, a panel of six Central American countries, and a panel of 80 countries.Menegaki (2011)found no causality between renewable energy consumption and economic growth for a panel of 27 European countries. Recently, renewable and non-renewable energy consumptions are considered simultaneously in the production model framework in order to differentiate the relative impact of each type in the economic growth process.

Apergis and Payne (2011b, 2012b) revealed bidirectional short-and long-run causalities between renewable short-and non-renewable energy consumption and economic growth for a panel of 25 developed countries, a panel of 55 developing countries, and a panel of 80 countries.Apergis and Payne (2011c)found unidirec-tional causality from economic growth to renewable electricity consumption in the short-run and bidirectional causality in the long-run, and bidirectional short- and long-run causalities between non-renewable electricity consumption and economic growth for a panel of 16 emerging market economies.Apergis and Payne (2012a) found unidirectional causality from renewable electricity consumption to economic growth in the short-run and bidirectional causality in the long-run, and bidirectional short- and long-run causalities between non-renewable electricity consumption and economic growth for a panel of six Central American countries.

For nuclear power,Apergis and Payne (2010d)found unidirec-tional long-run causality from nuclear energy consumption to economic growth and bidirectional causality in the short-run for a panel of 16 countries that currently produce nuclear energy.

Apergis et al. (2010)found unidirectional short-run causality from nuclear energy consumption to economic growth and bidirectional causality in the long-run for a panel of 19 developed and devel-oping countries. Payne and Taylor (2010) found an absence of Granger causality between nuclear energy consumption growth and economic growth in the US.Wolde-Rufael and Menyah (2010)

explored the causal relationship between nuclear energy con-sumption and real GDP for nine developed countries, and their results are inconsistent using the production model framework.

Heo et al. (2011) found unidirectional causality from nuclear energy consumption to economic growth without any feedback effect in India. Taking into account the interrelationship among nuclear energy consumption, oil consumption, oil price, and real income in six highly industrialized countries,Lee and Chiu (2011a)

found unidirectional causality from real GDP to nuclear energy consumption in Japan, bidirectional causality in Canada, Germany, and the United Kingdom, and no causality in France and the United States. They also found unidirectional causality from oil price to nuclear energy consumption except in the United States, and unidirectional causality from oil consumption to nuclear energy consumption in Canada, Japan, and the United Kingdom. They suggested that real GDP growth, oil price increases, or oil

supply shortages have significant impacts on the development of nuclear energy in highly industrialized countries; Lee and Chiu (2011b)found evidence of unidirectional long-run causality from oil price and economic growth respectively to nuclear energy consumption, no short-run causality between nuclear energy consumption and economic growth, and substitutability between nuclear energy and oil for a panel of six highly industrialized countries. They suggested that the imported-energy-dependent countries should set up long-term income and energy policies for stimulating their nuclear energy development. However, no stu-dies in the literature have explored the causal relationship between clean (e.g., hydroelectric, new renewables, nuclear) and non-clean (e.g., fossil fuel) energy consumption and economic growth in developing countries such as Brazil.

3. Model and methodology

3.1. Model

To investigate the relationships between the different types of energy sources and economic growth and to analyze the sub-stitutability between clean and non-clean energy sources, based on the production function framework, this paper extends the recent research ofApergis and Payne (2011b,2011c,2012a,2012b)

as follows:

Yt¼ f ðKt; Lt; CESt; FFCtÞ ð1Þ

where the subscript t is time. Y denotes the real GDP; K represents real gross fixed capital formation; L is total labor force; CE represents the consumption of clean energy by type S: total renewable, hydroelectric, new renewables, and nuclear; FFC repre-sents the non-clean fossil fuel energy consumption. Capital and labor variables were included to avoid estimation bias (Lütkepohl, 1982). All variables are in natural logarithms.

According to the 2009 EIA, Brazil boasts high growth rates of 12.53% and 10.94% respectively in NUCEC and NHREC for the 10-year period. However, in the past five years, NUCEC shows the lowest growth rate of 1.14% while NHREC has the highest growth rate of 12.76%. Thus, the production function framework, which investigates the substitutability between new renewables and nuclear energy consumption, is given as follows:

Yt¼ f ðKt; Lt; NHRECt; NUCECtÞ: ð2Þ

Analyses of the long-run relationship and Granger causality between variables in Eqs.(1)and(2)are crucial for a clean energy economy.

3.2. Econometric methodology

In this paper, empirical analysis tests the existence of a long-run relationship among the variables in Eqs. (1) and (2) while using the vector error-correction model to capture the Granger causality between variables. A three-step procedure is performed byfirst checking the integration order of each variable. The three standard unit root tests Augmented Dickey–Fuller (ADF) (Dickey and Fuller, 1981), Phillips–Perron (PP) (Phillips and Perron, 1988), and Kwiatkowski–Phillips–Schmidt–Shin (KPSS) (Kwiatkowski et al., 1992) are employed to investigate the stationarity and the integration order of the variables. The null hypothesis of ADF or PP is that a series is I (1), while KPSS is that a series is I (0). To obtain robust results, this paper uses KPSS to complement the widely used ADF and PP tests. However,Perron (1989)argued that in the presence of a structure break, the standard unit root tests are biased toward the non-rejection of the null hypothesis.Zivot and Andrews (1992)(hereafter referred to as ZA) developed a unit root

(4)

test with an endogenous structure break, which allows a break at an unknown point in either the intercept, the linear trend, or in both. Taking the structure break into account, ZA unit root tests are also employed for each time series.

Next, when all of the series in Eqs.(1) or (2) have the same integration order, the Johansen maximum likelihood method (Johansen and Juselius, 1990) is used to test the co-integration relationship among the variables in Eqs.(3) or(4) as follows: Yt¼ β10þ β11Ktþ β12Ltþ β13CEStþ β14FFCtþ u1t; ð3Þ

Yt¼ β20þ β21Ktþ β22Ltþ β23NHRECtþ β24NUCECtþ u2t ð4Þ

The existence of cointegration indicates that (1) the parameters estimated by OLS are super-consistent (Alves and Bueno, 2003); (2) there are long-run equilibrium relationships among the vari-ables; and (3) Granger causality exists among the variables in at least one direction (Engle and Granger, 1987;Oxley and Greasley, 2008). The coefficient βjkcan be interpreted as elasticity estimate.

For the last step, if all of the variables are I (1) and cointegrated, the vector error correction model (VECM) is used to explore the long- and short-run causality between variables. The VECM for Eq.

(2)is specified as follows: ΔYt¼ γ10þ ∑

p

k¼ 1ðγ11kΔYtkþ γ12kΔKtkþ γ13kΔLtk

þγ14kΔNHRECtkþ γ15kΔFFCtkÞ þ δ1ECTt1þ μ1t ð5aÞ

ΔKt¼ γ20þ ∑ p k¼ 1ðγ21kΔYtkþ γ22kΔKtkþ γ23kΔLtk þγ24kΔNHRECtkþ γ25kΔFFCtkÞ þ δ2ECTt1þ μ2t ð5bÞ ΔLt¼ γ30þ ∑ p k¼ 1ðγ31kΔYtkþ γ32kΔKtkþ γ33kΔLtk þγ34kΔNHRECtkþ γ35kΔFFCtkÞ þ δ3ECTt1þ μ3t ð5cÞ ΔNHRECt¼ γ40þ ∑ p k¼ 1ðγ41kΔYtkþ γ42kΔKtkþ γ43kΔLtk þγ44kΔNHRECtkþ γ45kΔFFCtkÞ þ δ4ECTt1þ μ4t ð5dÞ ΔFFCt¼ γ50þ ∑ p k¼ 1ðγ51kΔYtkþ γ52kΔKtkþ γ53kΔLtk

þγ54kΔNHRECtkþ γ55kΔFFCtkÞ þ δ5ECTt1þ μ5t ð5eÞ

where

ECTt1¼ Yt1α0α1Kt1α2Lt1α3NHRECt1α4FFCt1; ð5fÞ

Δ is the first-difference operator; k is the lag lengths determined by the Akaike's information criteria (AIC); and mjt(j¼1,2,3,4,5) is

the serially uncorrelated error terms. Note that the time series in thefirst difference of the natural logarithm can be interpreted as a growth rate of this variable. Short-run Granger causality is examined by testing H0:γlmk¼0, ∀k, using χ2-statistics, where l,

m¼1,…,5, with l≠m. Long-run Granger causality is examined by testing H0:δj¼0, using t-statistics, where j¼1,2,3,4,5. The

adjust-ment parameter δ1 measures the speed at which Y returns to

the long-term equilibrium levels after violating the long-term equilibrium relationship. Similar tests can be examined to test the short- and long-run Granger causalities in Eqs.(5b)–(5e).

In Eq. (5), the NHREC variable is replaced by TREC, HEC, or NUCEC to investigate the causal relationship between total renew-able, hydroelectric, or nuclear energy consumption and economic growth as well as to analyze the substitutability between clean energy consumption and non-clean consumption. Furthermore, the FFC variable in Eq. (5) is replaced by NUCEC to analyze the substitutability between new renewables and nuclear energy consumption.

4. Data and empiricalfindings 4.1. Data analysis

Annual data for Brazil's real GDP, real gross fixed capital formation (K), and labor force (L) from 1980 to 2009 were obtained from the World Development Indicators (WDI). Data for HEC, NHREC (sum of consumption of wind, biomass, and waste in Brazil), TREC (sum of NHREC and HEC), NUCEC, FFC (sum of consumption of petroleum, coal, and nature gas), and TNREC were extracted from the Energy Information Administration (EIA), where HEC and NHREC accounted for respectively 97.11% and 2.89% of TREC and NUCEC and FFC accounted for respectively 1.20% and 96.67% of TNREC. Note that the net consumption does not include the energy consumption of generating units. All different types of energy sources are measured in quadrillion BTU. Both real GDP and the real grossfixed capital formation are measured in US dollars at the year 2000 prices. The total labor force is measured in millions. Table 1 displays the summary statistics associated with the nine variables mentioned above for the actual data.

Fig. 1shows the time series of Brazilian data in natural logarithms, which have all confirmed growth. As shown inTable 1, nuclear and new renewables energy consumption demonstrate the largest coef fi-cient of variation (120% and 75%).Table 2shows the average growth rate of the nine variables, where the 15-year (1994–2009), 10-year (1999–2009), and five-year (2004–2009) growth rates are calculated to respectively demonstrate the long-term, medium-term, and short-term growth trends. NUCEC has the highest growth rates, which are more than 12% and 38% respectively for the long-term and medium-term periods, while it has the lowest short-medium-term growth rate of 1.14%, higher than the world short-term growth rate of0.36%. As shown inTable 2, the 10-year growth rate of 12.53% in NUCEC is nearly 6.92 times higher than the growth rate of 1.81% in FFC. In 2009, nuclear energy accounts for about 3% of Brazil's electricity.

Brazil is the third-largest country for electricity consumption in the Western Hemisphere because of an abundant supply of inexpensive electricity. In 2009, over 90% of Brazil's total electricity generation came from hydroelectricity, which was hypothesized to impact Brazil's economic growth; the share of hydroelectricity in Brazil's renewable electricity generation was approximately 94.35% and the remaining 5.65% was from new renewables (wind,

Table 1

Summary statistics for actual data, 1980–2009.

Y K L TREC HEC NHREC NUCEC TNREC FFC

(US$ Billions) (Million) (quadrillion BTU)

Mean 586.58 (132.63) 102.27 (19.26) 73.18 (16.90) 2.60 (0.76) 2.52 (0.71) 0.08 (0.06) 0.05 (0.06) 4.29 (1.39) 4.15 (1.30)

CV (%) 22.61 18.83 23.09 29.31 28.18 75.00 120.00 32.45 31.33

Notes: Figures in parentheses are standard deviations. CV represents the coefficient of variation. K and L are real gross fixed capital formation and labor force, respectively. TREC, HEC, NHREC, NUCEC, TNREC, and FFC are, respectively, total renewable, hydroelectric, non-hydroelectric renewable, nuclear, total non-renewable, and fossil fuel energy consumptions.

(5)

biomass, and waste) electricity generation. After a drought crisis in early 2000, Brazil's government actively diversified in order to reduce the country's reliance on hydropower. In low rainfall/dry seasons, Brazil's wind power is used to supplement water power. As shown inTable 2, the 10-year growth rate of 10.94% in NHREC is nearly 4.5 times higher than the growth rate of 2.45% in HEC.

Additionally, Brazil's real GDP enjoys a growth rate of 3.52%, which is higher than the world growth rate of 3.43% for the short-term period and its growth trend is expected to continue in the next decade. This study empirically investigates the causal relationship between clean energy use and economic growth in Brazil over the period of 1980–2009. Aggregated level of total renewable energy consumption as well as disaggregated levels of hydroelectric, new renewables, and nuclear energy consumption are tested.

4.2. Cointegration test results

This analysis begins with the unit root tests.Table 3 for the standard unit root tests andTable 4for the ZA unit root tests show that all of the time series are integrated at order one (i.e., I (1)). Next, the Johansen cointegration test is used for the variables in Eqs.(2) and (3). The results of Panels A–E inTable 5indicate the existence of at least one cointegrating vector for respectively the (Y, K, L, TREC, TNREC), (Y, K, L, HEC, FFC), (Y, K, L, NHREC, FFC), (Y, K, L, NUCEC, FFC), and (Y, K, L, NHREC, NUCEC) combinations at a 5% significant level. Thus, the OLS estimated coefficients shown in

Table 6for Eqs.(3)and(4)are non-spurious regression results. The values of both R2and the Jarque and Bera (JB) statistics (Jarque and Bera, 1980), shown inTable 6, indicate that Eqs. (3) and(4) are adequate. Therefore, real GDP, real grossfixed capital formation, labor force, FFC, and each of the three clean energy consumption items share common long-term trends. The cointegrating vectors shown inTable 6are (1, 0.40, 0.49, 0.20,0.04), (1, 0.36, 0.54, 0.08, 0.03), (1, 0.41, 0.54, 0.16, 0.04), and (1, 0.41, 0.74, 0.0004, 0.04) for Eq.(3), and (1, 0.35, 0.51, 0.08, 0.0008) for Eq.(4). The long-term coefficients for K, L, and NHREC are positive

and statistically significant at a 1% level, while TREC, TNREC, and the other different energy consumption items are statistically insignificant. The findings suggest that new renewables energy consumption is sensitive to real output, while the impacts caused by total renewable, total non-renewable, hydroelectric, nuclear, and fossil fuel energy consumption on real output are insignificant.

3.6 4.0 4.4 4.8 5.2 5.6 6.0 6.4 6.8 1980 1985 1990 1995 2000 2005 GDP K L -7 -6 -5 -4 -3 -2 -1 0 1 2 1980 1985 1990 1995 2000 2005 NUCEC NHREC HEC FFC

Fig. 1. Time series plots in natural logarithms of real GDP, real grossfixed capital formation, labor force, hydroelectric, non-hydroelectric renewable, nuclear, and fossil fuel energy consumption from 1980 to 2009.

Table 2

Average growth rates (%) in 2009 over 5-, 10-, and 15-year periods.

Y K L TREC HEC NHREC NUCEC TNREC FFC

5-year growth 3.52 5.70 1.67 3.90 3.49 12.76 1.14 2.01 2.05

10-year growth 3.27 3.22 1.95 2.77 2.45 10.94 12.53 1.90 1.81

15-year growth 2.85 2.68 2.04 3.10 2.85 10.15 38.12 2.91 2.81

Table 3

Results of unit roots tests, 1980–2009.

ADF PP KPSS Level 1st difference Level 1st difference Level 1st difference Y 0.6960 4.4384*** 0.5399 5.0891*** 0.7025** 0.1329 K 1.7697 3.7613*** 0.6826 4.0333*** 0.5541** 0.1907 L 2.6173 4.0425*** 2.5225 4.0425*** 0.7061** 0.0998 TREC 1.57 5.38*** 1.92 5.38*** 0.71** 0.22 HEC 1.7582 5.3878*** 2.4711 5.3867*** 0.7080** 0.2533 NHREC 0.6177 5.5776*** 1.3994 5.5900*** 0.7140** 0.2221 NUCEC 0.6076 6.1653*** 0.9744 6.9152*** 1.1324*** 0.1607 TNREC 0.75 7.49*** 0.48 14.10*** 0.69** 0.36 FFC 0.7193 7.6040*** 0.5054 14.2658*** 0.6836 ** 0.36

nnnSignificance at the 1% level. nnSignificance at the 5% level.

Table 4

Results of the ZA unit root tests with a structural break.

Level Break First difference Break

Y 4.43 (C) 1990 5.75*** (A) 2004 K 4.26 (C) 1990 5.18**(C) 2001 L 3.10 (B) 1994 4.84***(B) 1992 TREC 4.09 (C) 2001 7.15*** (C) 2003 HEC 4.10 (C) 2001 7.23*** (C) 2003 NHREC 3.31 (B) 1993 7.17*** (B) 1986 NUCEC 4.86 (C) 1995 7.45*** (C) 1995 TNREC 4.68** (B) 2001 8.21*** (C) 1994 FFC 4.68**(B) 2001 7.67***(B) 1986

Notes: The letters in parentheses indicate the Models A, B, and C ofZivot and Andrews (1992). Model A allows for a change in the level of a series; Model B allows for a change in the slope of the trend of a series, while Model C combines both of the changes.

nnnSignificance at the 1% level. nnSignificance at the 5% level.

(6)

Thus, disaggregated levels such as HEC, NHREC, NUCEC, and FFC are appropriate for exploring the relationship between clean energy use and economic growth. However, aggregated total renewable energy consumption may very well mask the differential impact of hydroelectric and new renewables energy consumption on economic growth. Table 6 also shows that Brazil's economy is highly dependent on investment and labor. A 1% increase in real gross fixed capital formation increases real GDP by more than 0.35%, a 1% increase in the labor force increases real GDP by more than 0.49%, and a 1% increase in new renewables energy con-sumption increases real GDP by 0.08%.

4.3. Granger causality results

The VECM-based causality tests are performed in this last section. The causal relationships between economic growth and both clean and non-clean energy consumptions are presented in

Table 7from estimating Eq. (5). Clean energy consumption at both aggregated level of total renewable energy consumption and disaggregated levels of hydroelectric, non-hydroelectric renewable (new renewables), and nuclear energy consumptions are tested. The relationships between total renewable and total non-renewable energy consumption and economic growth are shown

in Panel A ofTable 7. Panels B–D show the relationships between clean and fossil fuel energy consumption and economic growth, while three disaggregated levels of clean energy consumption are being tested. The substitutability between new renewables and nuclear energy consumption is presented inTable 8. A significance level of 5% is in general assumed in this study.

The results in Panels A and B of Table 7 show that the causal relationships between GDP, TREC, TNREC, K, and L are similar to the causal relationships between GDP, HEC, FFC, K, and L, because HEC and FFC account for 94.35% and 96.67% respectively of TREC and TNREC from 1980 to 2009. The short-run dynamics indicates unidirectional causality from real grossfixed capital formation to non-renewable energy consumption, and bidirectional causality between economic growth and labor force. The statistically significant coefficients of ECT in GDP, TNREC, K, and L equations indicate bidirectional long-run causality between economic growth, non-renewable energy consump-tion, real grossfixed capital formation, and labor force. Note that no statistically significant coefficient in TREC and HEC equations in Panels A and B indicates that all variables, except for both TREC and HEC, are confirmed of Granger endogeneity and that there are unidirectional causalities from total renewable energy consumption and hydro-electric consumption, respectively, to other variables. Thus, based on the aggregated analysis, the Granger causality results indicate bidirec-tional causality between non-renewable energy consumption and economic growth and unidirectional causality from renewable energy consumption to economic growth in the long-run. Thesefindings are partially similar to those reported byApergis and Payne (2011b,2011c,

2012a,2012b)for developed and developing countries, 16 emerging market economies, Central America, and 80 countries, respectively. Their studies reveal consistent results of the bidirectional causality between both renewable and non-renewable energy consumption and economic growth in the long-run, while the relationship between renewable energy consumption and economic growth is inconsistent in the short-run. The positive unidirectional causality from total renewable or hydroelectric energy consumption to economic growth is to be expected because over 97% of Brazil's total electricity generation comes from total renewable electricity generation, as well as the fact that the share of hydroelectricity consumption in Brazil's renewable electricity generation was approximately 94.35% in 2009.

For the new renewables energy consumption, the results in Panel C ofTable 7show that economic growth has negative and statistically significant impact on both non-hydroelectric renewable and fossil fuel energy consumption, and new renewables energy consumption and fossil fuel consumption are negatively affected in the short-run. The statistically significant coefficients of ECT in GDP, NHREC, FFC, and K equations indicate bidirectional long-run causality between economic growth, real grossfixed capital formation, new renewables, and fossil fuel energy consumption. Overall, the results of economic growth negative impact on both new renewables and fossil fuel energy consumption in the short-run indicate that economic growth may

Table 5

Results of Johansen's cointegration test. Eigen value Trace statistics 5% Critical value Max Eigen statistics 5% Critical value No. of CEs Panel A: Panel A, TREC, TNREC, K, and L variables; no lags

0.91 117.12***

76.97 69.38***

34.81 r¼0

0.59 47.74 54.08 25.99 28.59 r≤1

Panel B: Y, HEC, FFC, K, and L variables; no lags 0.91 118.04***

76.97 69.32***

34.81 r¼0

0.60 48.72 54.08 26.30 28.59 r≤1

Panel C: Y, NHREC, FFC, K, and L variables; no lags 0.91 125.30*** 76.97 69.09*** 34.81 r¼0 0.55 56.22** 54.08 23.44 28.59 r≤1 0.47 32.78 35.19 18.13 22.30 r≤2

Panel D: Y, NUCEC, FFC, K, and L variables; no lags

0.90 127.42*** 76.97 67.11*** 34.81 r¼0

0.60 60.31**

54.08 26.49 28.59 r≤1

0.51 33.82 35.19 20.41 22.30 r≤2

Panel E: Y, NHREC, NUCEC, K, and L variables; no lags 0.91 126.02*** 76.97 68.69*** 34.81 r¼0 0.61 57.33** 54.08 27.09 28.59 r≤1 0.39 30.24 35.19 14.30 22.30 r≤2

r is the cointegration rank.

nnnRejection of a null hypothesis at the 1% level. nnRejection of a null hypothesis at the 5% level.

Table 6

Coefficients of Eqs.(3)and(4).

K L TREC TNREC HEC NHREC NUCEC FFC Intercept R2 JB

p-Value Panel A: aggregated level

Eq.(3) 0.40nnn(9.43) 0.49nnn(3.53) 0.20 (2.02) 0.04 (0.68) 2.29nnn(4.15) 0.990 1.64 0.44

Panel B: disaggregated levels

Eq.(3) 0.41nnn(9.43) 0.54nnn(4.05) 0.16 (1.65) 0.04 (0.80) 2.06nnn(3.90) 0.989 1.60 0.45

Eq.(3) 0.36nnn(8.18) 0.54nnn(6.10) 0.08nnn(2.98) 0.03 (0.51) 2.66nnn(5.12) 0.991 0.27 0.88

Eq.(3) 0.41nnn(8.86) 0.74nnn(10.89) 0.0004 (0.12) 0.04 (0.74) 1.35nnn(4.21) 0.988 1.75 0.42

Eq.(4) 0.35nnn(8.99) 0.51nnn(7.64) 0.08nnn(3.06) 0.0008 (0.30) 2.80nnn(6.55) 0.991 0.12 0.94

Notes: Figures in parentheses indicate t-statistics. JB represents the Jarque and Bera statistics.

(7)

improve the energy infrastructure and energy efficiency, and therefore lead to a lower energy consumption while reducing fossil fuel and new renewables energy consumption. Thefinding of long-run bidirectional causalities between new renewables energy consumption, fossil fuel consumption, and economic growth supports their mutual interde-pendence, suggesting that policies which enhance new renewables and fossil fuel energy consumption have a positive influence on economic growth. Thus, mixing the two types of energy sources is important although fossil fuel consumption affect the environment adversely. Thesefindings are similar to those reported by Sadorsky (2009)for emerging economies andApergis and Payne (2010a) for OECD countries. They reported short- and long-run bidirectional causalities between non-hydroelectric renewable energy consumption and economic growth. Additionally, the negative bidirectional caus-ality between new renewables and fossil fuel energy consumption reflects the potential substitutability between the two energy sources. Thefinding suggests that the development of the new renewables sector may provide relief from greenhouse gas emissions generated by fossil fuel consumption.Table 2shows that the 10-year growth rate of 10.94% in new renewables energy consumption is nearly 6.1 times higher than the growth rate of 1.81% in fossil fuel consumption. The substitutability is similar to the results reported byApergis and Payne (2010a, 2012a) for 80 countries. With respect to the long-run dynamics, the coefficient of ECT in the new renewables equation is negative and significant, implying that new renewables energy con-sumption would respond to bring the system back to equilibrium when a shock occurs. The speed of adjustment toward long-term equilibrium is rather slow, due to the fact that electricity generation

from new renewables accounted for only 5.51% of Brazil's total electricity generation in 2009.

For the nuclear energy consumption, the results in Panel D of

Table 7 show that economic growth has negative statistically significant impact on nuclear energy consumption, and nuclear energy consumption has positive statistically significant impact on economic growth in the short-run. The statistically significant coefficient of ECT in NUCEC equation indicates unidirectional long-run causality from respectively economic growth, real grossfixed capital formation, labor force, and fossil fuel consumption to nuclear energy consumption. The results inTable 8indicate that nuclear and non-hydroelectric renewable energy consumption is negatively affected on each other in the short-run. The statistically significant coefficients of ECT in GDP, NHREC, NUCEC, and K equations indicate bidirectional long-run causality between eco-nomic growth, real grossfixed capital formation, new renewables, and nuclear energy consumptions. Overall, thefinding of bidirec-tional causalities between economic growth, new renewables, and nuclear energy consumption supports their mutual interdepen-dence, suggesting that limiting nuclear energy use would hamper economic growth. Thisfinding is very similar to the results from

Wolde-Rufael and Menyah (2010)for France, Spain, the UK, and the US,Lee and Chiu (2011a)for Canada, Germany, and the UK; and is partially similar to Apergis and Payne (2010c) for 16 countries, Lee and Chiu (2011b) for a panel of six developed countries, and Heo et al. (2011) for India; but differs from the results fromPayne and Taylor (2010)for the US, where there is no causal relationship between nuclear energy consumption and real

Table 7

Causality tests for clean and fossil fuel energy consumptions and economic growth.

Dependent variable Source of causation (independent variables) short-run Long-run

Panel A: total renewable and total non-renewable energy consumptions and economic growth

ΔY ΔTREC ΔTNREC ΔK ΔL ECT

ΔY 3.732n(+) 1.311 0.250 8.411nnn 0.186nnn(4.157)

ΔTREC 1.025 0.565 2.599 0.052 0.092 (0.951)

ΔTNREC 3.875 0.114 10.706nnn 2.553 0.851nnn(3.664)

ΔK 0.325 0.964 0.792 2.615 0.631nn(2.199)

ΔL 6.617nn 0.260 2.016 3.981 0.121nn(2.775)

Panel B: hydroelectric renewable and fossil fuel energy consumptions and economic growth

ΔY ΔHEC ΔFFC ΔK ΔL ECT

ΔY 3.740n(+) 1.271 0.266 8.628nnn 0.194nnn(4.311)

ΔHEC 1.242 0.605 3.236 0.104 0.087 (0.930)

ΔFFC 5.169 0.508 14.142nnn 2.580 0.904nnn(4.199)

ΔK 0.307 0.870 0.807 2.603 0.627nn(2.231)

ΔL 6.065nn 0.292 2.120 4.450 0.117nn(2.581)

Panel C: non-hydroelectric renewable and fossil fuel energy consumptions and economic growth

ΔY ΔNHREC ΔFFC ΔK ΔL ECT

ΔY 0.434 2.957 7.93nn 3.188 0.027nnn(3.946) ΔNHREC 7.158nn() 9.528nnn() 10.272nnn 1.453 0.091nn(2.567) ΔFFC 6.030nn() 32.317nnn() 4.542 3.083 0.150nnn(7.177) ΔK 3.539 5.024c 3.618 7.752nn 0.140nnn(4.086) ΔL 1.843 1.337 1.417 1.739 0.008 (1.090)

Panel D: nuclear and fossil fuel energy consumptions and economic growth

ΔGDP ΔNUCEC ΔFFC ΔK ΔL ECT ΔGDP 9.749nnn(+) 1.792 0.616 0.652 0.106 (1.357) ΔNUCEC 6.264nn() 1.948 6.364nn 4.533 12.140nnn(7.231) ΔFFC 0.281 4.301 4.463nn 5.604 0.138 (0.593) ΔK 2.743 3.089 1.402 0.789 0.333 (1.041) ΔL 1.018 5.625nn 1.720 0.641 0.058 (1.139)

Notes: Wald chi-square tests reported with respect to short-run change in the independent variables. ECT represents the estimated coefficient on the error correction term. Figures in parentheses are t-statistics.

nnnRejection of a null hypothesis at the 1% level. nnRejection of a null hypothesis at the 5% level. nRejectin of a null hypothesis at the 10% level.

(8)

output. Additionally, the negative bidirectional causality between new renewables and nuclear energy consumption reflects the potential substitutability between the two energy sources. This finding suggests that the establishment of partnerships between these two types of clean energy sources would be an urgent need in order to meet the huge demand for clean energy economy in Brazil. In fact, in the past five years, NUCEC shows the lowest growth rate of 1.14% while NHREC has the highest growth rate of 12.76%. With respect to long-run dynamics, the coefficients of ECT in the nuclear equation is negative and significant, implying that nuclear energy consumption would respond to bring the system back to equilibrium when a shock occurs. Adjustment back to equilibrium takes as little time as 0.2 year. The overall causal relationship between energy consumption and economic growth in Brazil is shown inFig. 2.

5. Conclusions

In light of the economic and societal growing concerns over global warming caused by fossil fuel, high volatility of energy prices, and high growth of energy needs, clean energy (e.g., hydroelectric, new renew-ables, and nuclear) has become an important alternative energy source for fossil fuel. Currently, hydroelectric, new renewables, and nuclear energy supply respectively make up approximately 15%, 3%, and 14% of the global electricity needs. The aim of this study is to empirically investigate the causal relationship between clean and non-clean energy use and economic growth in Brazil over the period of 1980– 2009. Aggregated level of total renewable energy consumption as well as disaggregated levels of hydroelectric, new renewables, and nuclear energy consumption are tested. The simultaneous use of clean and non-clean (e.g., fossil fuel) energy consumption in the production function framework intends to distinguish the relative influence of each type on economic growth and to analyze the substitutability between the different types of energy sources.

The Johansen cointegration tests reveal that there is a long-term equilibrium relationship between real GDP, real grossfixed

capital formation, labor force, and renewable and non-renewable energy consumption at aggregated level, as well as a long-term equilibrium relationship between real GDP, real grossfixed capital formation, labor force, hydroelectric/new renewables/nuclear and fossil fuel energy consumption at disaggregated level. The real grossfixed capital formation, labor force, and new renewables energy consumption elasticities of real GDP are statistically sig-nificant at a 1% level and higher than 0.35, 0.48, and 0.08, respectively, while total renewable and total non-renewable energy consumption at aggregated level and hydroelectric, nuclear, and fossil fuel energy consumption at disaggregated level are statistically insignificant at a significance level of 5%. This suggests that in the long term, increases in real grossfixed capital formation and labor force are major drivers behind real GDP and that new renewables energy consumption has a lesser and positive impact on real GDP, while total renewable, total non-renewable, hydroelectric, nuclear, and fossil fuel energy consump-tion do not seem to have a very strong impact on real GDP. Aggregated analysis may obscure the relationship between renew-able energy consumption and real GDP.

The results from the vector error correction models show that (1) for the aggregated analysis of renewable energy consumption, the result of long-run interdependence between non-renewable energy consumption and economic growth (TNREC2ΔY) affirms the importance of traditional energy sources in the design of energy policy for a more sustainable energy future, although clean energy sources are increasingly important in Brazil. The result of positive unidirectional causality from TREC to economic growth (TREC-ΔY) suggests that limiting total renewable energy use would hamper economic growth. This causal relationship is to be expected because over 97% of Brazil's total electricity genera-tion comes from total renewable electricity generagenera-tion. (2) For the disaggregated analysis of clean energy consumption, the results of bidirectional causality between new renewables, nuclear, and fossil fuel energy consumption and economic growth in the long-run support their mutual interdependence (NHREC2ΔY, NUCEC2ΔY, and FFC2ΔY), suggesting that these three kinds of energy resources are important for the future development of the clean energy economy. The result of positive unidirectional caus-ality from hydroelectric consumption to economic growth (HEC -ΔY) suggests that limiting hydroelectric use would hamper eco-nomic growth. In fact, after a drought crisis in early 2000, Brazil's government actively diversified in order to reduce the country's reliance on hydropower, such as the use of wind power to supplement water power in dry seasons and the development of the four largest nuclear reactors to be online by 2025. Further-more, the share of HEC in Brazil's TREC was approximately 94.35% in 2009. Thus, there is no reverse causality from HEC or TREC to economic growth. Yet, to facilitate the expansion of the new renewables energy sector, economic growth is vital to provide

Table 8

Results of causality tests for the non-hydroelectric renewable energy consumption model.

Dependent variable

Source of causation (independent variables) short-run Long-run

ΔY ΔNHREC ΔNUCEC ΔK ΔL ECT

ΔY 3.408 9.792nn(+) 11.816nnn 4.894 0.249nn(2.177)

ΔNHREC 13.316nnn() 16.927nnn() 15.700nnn 8.906nnn 1.547nnn(3.742)

ΔNUCEC 4.106 6.701nn() 3.950 2.046 19.536nnn(5.247)

ΔK 3.539 5.024 3.618 7.752nn 0.140nnn(4.086)

ΔL 1.843 1.337 1.417 1.739 0.008 (1.090)

Notes: Wald chi-square tests reported with respect to short-run change in the independent variables. ECT represents the estimated coefficient on the error correction term. Figures in parentheses are t-statistics.

nnnRejection of a null hypothesis at the 1% level. nnRejection of a null hypothesis at the 5% level.

(9)

the resources for further research and development of new renewables energy technologies and corresponding infrastructure. Additionally, to increase nuclear energy supply investment, the right balance should be struck between the quest for economic growth, nuclear safety, clean energy, and the drive toward making the country relatively energy independent. (3) In the short-run dynamics, nuclear energy consumption's impact on economic growth is positive and the impacts of economic growth on nuclear, new renewables, and fossil fuel energy consumption are negative. These findings suggest that limiting nuclear energy use would hamper economic growth. Economic growth may improve energy infrastructure and energy efficiency, and therefore lead to a lower energy consumption while reducing different types of energy consumption. (4) The presence of substitutability between new renewables and fossil fuel energy consumption provides an avenue for the continued use of government policies that enhance the development of the new renewable energy sector as well as encourage the effective development of carbon markets in Brazil to reduce fossil fuel use. Finally, (5) the presence of substitutability between new renewables and nuclear energy consumption sug-gests that the establishment of partnerships between these two types of clean energy sources would be an urgent need in order to meet the greener low carbon economy in Brazil. In fact, new renewables technologies are suitable for local electricity and nuclear power is the only effective option to provide for a large demand for clean electricity at a national scale.

To continue the development of new renewables energy sources and renewable energy markets, the Brazilian government should introduce more preferential policies, such as investment subsidies or tax rebates, tax incentives, sales tax, and green certificate trading, to promote the development of a clean energy economy. In order to ensure energy security and stability, minimize the impact of high oil price volatility on macroeconomics, and reduce greenhouse gas emissions, the development of nuclear power is a means, but economic necessity should not outweigh the risk involved. After all, nuclear safety is a global concern that calls for a global solution.

Acknowledgments

We are very grateful to two anonymous referees whose con-structive comments and suggestions have helped to improve upon the quality of the paper.

References

Alves, D.C.O., Bueno, R.D., 2003. Short-run, long-run and cross elasticities of gasoline demand in Brazil. Energy Economics 25, 191–199.

Apergis, N., Menyah, K., Payne, J.E., Wolde-Rufael, Y., 2010. On the causal dynamics between emissions, nuclear energy, renewable energy, and economic growth: a panel investigation. Ecological Economics 69, 2255–2260.

Apergis, N., Payne, J.E., 2010a. Renewable energy consumption and economic growth: evidence from a panel of OECD countries. Energy policy 38, 656–660. Apergis, N., Payne, J.E., 2010b. Renewable energy consumption and growth in

Eurasia. Energy Economics 32, 1392–1397.

Apergis, N., Payne, J.E., 2010c. Energy consumption and economic growth in South America: evidence from a panel error correction model. Energy Economics 32, 1421–1426.

Apergis, N., Payne, J.E., 2010d. A panel study of nuclear energy consumption and economic growth. Energy Economics 32, 545–549.

Apergis, N., Payne, J.E., 2011a. The renewable energy consumption–growth nexus in Central America. Applied Energy 88, 343–347.

Apergis, N., Payne, J.E., 2011b. On the causal dynamics between renewable and non-renewable energy consumption and economic growth in developed and developing countries. Energy System 2, 299–312.

Apergis, N., Payne, J.E., 2011c. Renewable and non-renewable electricity consump-tion–growth nexus: evidence from emerging market economies. Applied Energy 88, 5226–5230.

Apergis, N., Payne, J.E., 2012a. The electricity consumption–growth nexus: renew-able versus non-renewrenew-able electricity in Central America. Energy Sources, Part B: Economics, Planning, and Policy 7, 423–431.

Apergis, N., Payne, J.E., 2012b. Renewable and non-renewable energy consumption– growth nexus: evidence from a panel error correction model. Energy Econom-ics 34, 733–738.

Apergis, N., Payne, J.E., 2012c. A global perspective on the renewable energy consumption–growth nexus. Energy Sources, Part B: Economics, Planning, and Policy 7, 314–322.

Apergis, N., Payne, J.E., 2013. Another look at the electricity consumption–growth nexus in the South America. Energy Sources, Part B: Economics, Planning, and Policy 8, 171–178.

Bowden, N., Payne, J.E., 2010. Sectoral analysis of the causal relationship between renewable and non-renewable energy consumption and real output in the US. Energy Sources, Part B: Economics Planning, and Policy 5, 400–408. Dickey, D.A., Fuller, W.A., 1981. Likelihood ratio statistics for autoregressive time

series with a unit root. Econometrica 49, 1057–1072.

Engle, R.F., Granger, C.W.J., 1987. Co-integration and error correction: representa-tion, estimarepresenta-tion, and testing. Econometrica 55, 251–276.

Ewing, B.T., Sari, R., Soytas, U., 2007. Disaggregate energy consumption and industrial output in the United States. Energy Policy 35, 1274–1281. Heo, J.Y., Yoo, S.H., Kwak, S.J., 2011. The causal relationship between nuclear energy

consumption and economic growth in India. Energy Sources, Part B 6, 111–117. Jarque, C.M., Bera, A.K., 1980. Efficient tests for normality, homoskedasticity and

serial independence of regression residuals. Economics Letters 6, 255–259. Johansen, S., Juselius, K., 1990. Maximum likelihood estimation and inferences on

co-integration with approach. Oxford Bulletin of Economics and Statistics 52, 169–209.

Kwiatkowski, D., Phillips, P., Schmidt, P., Shin, Y., 1992. Testing the null hypothesis of stationarity against the alternative of a unit root: how sure are we that economic time series have a unit root? Journal of Econometrics 54, 159–178. Lee, C.C., Chiu, Y.B., 2011a. Nuclear energy consumption, oil price, and economic

growth: evidence from highly industrialized countries. Energy Economics 33, 236–248.

Lee, C.C., Chiu, Y.B., 2011b. Oil price, nuclear energy consumption, and economic growth: new evidence using a heterogeneous panel analysis. Energy Policy 39, 2111–2120.

Lütkepohl, H., 1982. Non-causality due to omitted variables. Journal of Econo-metrics 19, 267–378.

Menegaki, A.N., 2011. Growth and renewable energy in Europe: a random effect model with evidence for neutrality hypothesis. Energy Economics 33, 257–263. Oxley, L., Greasley, D., 2008. Vector auto-regression, co-integration and causality: testing for causes of the British industrial revolution. Applied Economics 30, 1387–1397.

Ozturk, I., 2010. A literature survey on energy–growth nexus. Energy Policy 38, 340–349.

Payne, J.E., 2009. On the dynamics of energy consumption and output in the US. Applied Energy 86, 575–577.

Payne, J.E., 2010a. Survey of the international evidence on the causal relationship between energy consumption and growth. Journal of Economic Studies 37, 53–95.

Payne, J.E., 2010b. A survey of the electricity consumption–growth literature. Applied Energy 87, 723–731.

Payne, J.E., 2011a. On biomass energy consumption and real output in the US. Energy Sources, Part B: Economics, Planning, and Policy 6, 47–52.

Payne, J.E., 2011b. US disaggregate fossil fuel consumption and real GDP: an empirical note. Energy Sources, Part B: Economics, Planning, and Policy 6, 63–68.

Payne, J.E., Taylor, J.P., 2010. Nuclear energy consumption and economic growth in the US: an empirical note. Energy Sources, Part B: Economics, Planning, and Policy 5, 301–307.

Pao, H.T., Fu, H.C., 2013. Renewable energy, non-renewable energy and economic growth in Brazil. Renewable and Sustainable Energy Reviews 25, 381–392. Perron, P., 1989. The great crash, the oil price shock and the unit root hypothesis.

Econometrica 57, 1361–1401.

Phillips, P.C., Perron, P., 1988. Testing for a unit root in time series regression. Biometrika 75, 335–346.

Sadorsky, P., 2009. Renewable energy consumption and income in emerging economies. Energy Policy 37, 4021–4028.

Sari, R., Soytas, U., 2004. Disaggregate energy consumption, employment, and income in Turkey. Energy Economics 26, 335–344.

Secretariat of Social Communication (SECOM), 2012. Sustainable development in Brazil. In: Proceedings of the Rio+20 United Nations Conference on Sustainable development.

Wikipedia, 2011. Renewable energy in Brazil.

Wolde-Rufael, Y., Menyah, K., 2010. Nuclear energy consumption and economic growth in nine developed countries. Energy Economics 32, 550–556. World Nuclear Association (WNA), 2011. Nuclear is part of the clean energy

solution.

Zivot, E., Andrews, D.W.K., 1992. Further evidence on the great crash, the oil-price shock, and the unit-root hypothesis. Journal of Business & Economic Statistics 3, 251–270.

數據

Fig. 1 shows the time series of Brazilian data in natural logarithms, which have all con firmed growth
Table 6 for Eqs. (3) and (4) are non-spurious regression results. The values of both R 2 and the Jarque and Bera (JB) statistics ( Jarque and Bera, 1980 ), shown in Table 6 , indicate that Eqs
Table 7 from estimating Eq. (5). Clean energy consumption at both aggregated level of total renewable energy consumption and disaggregated levels of hydroelectric, non-hydroelectric renewable (new renewables), and nuclear energy consumptions are tested
Table 7 show that economic growth has negative statistically signi ficant impact on nuclear energy consumption, and nuclear energy consumption has positive statistically signi ficant impact on economic growth in the short-run
+2

參考文獻

相關文件

• One technique for determining empirical formulas in the laboratory is combustion analysis, commonly used for compounds containing principally carbon and

You are given the wavelength and total energy of a light pulse and asked to find the number of photons it

Teachers may consider the school’s aims and conditions or even the language environment to select the most appropriate approach according to students’ need and ability; or develop

Wang, Solving pseudomonotone variational inequalities and pseudocon- vex optimization problems using the projection neural network, IEEE Transactions on Neural Networks 17

Hope theory: A member of the positive psychology family. Lopez (Eds.), Handbook of positive

volume suppressed mass: (TeV) 2 /M P ∼ 10 −4 eV → mm range can be experimentally tested for any number of extra dimensions - Light U(1) gauge bosons: no derivative couplings. =>

Define instead the imaginary.. potential, magnetic field, lattice…) Dirac-BdG Hamiltonian:. with small, and matrix

incapable to extract any quantities from QCD, nor to tackle the most interesting physics, namely, the spontaneously chiral symmetry breaking and the color confinement.. 