Reinvestigation of Relationship Between Macroeconomic Indexes and Energy Consumption in Iran
In this study attempted to investigation the relationship
between energy consumption in economic sectors and macroeconomic indexes
of Iran for 1970-2000 by using Vector Error Correction Model (VECM). Result
showed that a long run relationship existence between total energy consumption,
price index and gross national product. With respect to results stabilization
energy price policy in economical growth conditions will encourage energy
demand dimension. So, government must change energy price policy towards
variable pricing based on amount of consumption especially in peak and
Although energy have an important role in economical development and countries
welfare, but energy crisis in 1970 decade and high level and unforeseen of energy
conveyer prices specially about oil, cause execute limitation policies and saving
in energy consumption, as many of industrial countries have leaded toward gradual
management of increasing energy consumption (Emadzadeh et
al., 2002). In the end of 1970 decade and early of 1980 decade relationship
between energy consumption and economical growth located in focus point of economical
analyzer attention. Therefore in this period many studies have been accomplished
about effect of increasing energy price and consequently its consumption limitation
on economical growth and study it from different views (Abrishami
and Mostafaie, 2000). Marginal energy consumption in year 1958 was equal
to 53.4 million oil barrels and in year 1968 this measure increased to 206.9
million barrels oil as annual average having growth equivalent to 14.6%. After
Iran revolution and political and economical changes, specially imposed war,
energy consumption gently flows its increasing trend and from 199.7 million
barrels in year 1969 increased to 331.4 million barrels in year 1979. After
obtain liberation oil products consumption in year 1979, again energy consumption
become increase fast as in years 1979 to 1983 (across first development program)
having a growth equivalent to 5.88%. This growth in 1980-2000 decreased fewly
and equal to 3.1%. Masih and Masih (1997) showed that
energy consumption for Malaysia, Singapore and Philippines is none related with
income but there is an aside relation from energy consumption to GNP for India.
Aqeel and Butt (2001) with utilized Granger causality
test showed that economical growth is that cause of energy consumption in Pakistan
and economical growth cause consumption growth of oil products. About gas part,
there isnt a causality relationship between gas consumption and economical
growth. In part of power, electronic consumption cause economical growth. But
there isnt a reflected effect from gas consumption to economical growth. Chiou-Wei
et al. (2008) found evidence supporting a neutrality hypothesis for
the United States, Thailand and South Korea. However, empirical evidence on
Philippines and Singapore revealed a unidirectional causality running from economic
growth to energy consumption while energy consumption may have affected economic
growth for Taiwan, Hong Kong, Malaysia and Indonesia. In the low income group,
there existed no causal relationship between energy consumption and economic
growth; in the middle income groups (lower and upper middle income groups),
economic growth leads energy consumption positively; in the high income group
countries, economic growth leads energy consumption negatively.
Jinke et al. (2008) discovered that unidirectional causality running
from GDP to coal consumption exists in Japan and China and no causality relationship
exist between coal consumption and GDP in India, South Korea and South Africa
while the series are not cointegrated in USA. The major OECD or non-OECD countries
especially China, India and South Africa should reduce their CO2
emissions in coal consumption to reach sustainable development. Although economic
growth and energy consumption lack short-run causality, there is long-run unidirectional
causality running from energy consumption to economic growth i.e., reducing
energy consumption does not adversely affect GDP in the short-run but would
in the long-run. Zamani (2007) discovered that there
is A long-run unidirectional relationship from GDP to total energy and bidirectional
relationship between GDP and gas as well as GDP and petroleum products consumption
for the whole economy. Causality is running from value added to total energy,
electricity, gas and petroleum products consumption and from gas consumption
to value added in industrial sector. The long-run bidirectional relations hold
between value added and total energy, electricity and petroleum products consumption
in the agricultural sector. The short-run causality runs from GDP to total energy
and petroleum products consumption and also industrial value added to total
energy and petroleum products consumption in this sector.
Fatia et al. (2004) showed that in Newzland there isnt a causality
relationship between oil, gas and coal consumption and real gross national product
and variables are endogenous toward each other. On the other hands, there is
an aside Granger causality relationship from gross national product to total
marginal energy consumption and energy consumption in industrial department.
According to these studies reveal that their results about causality relationship
between energy consumption and economical growth isnt equal which may be resulting
of structural, constructional and political changes adopted with countries and
difference in research methodology. In addition, using of Engle
and Granger (1987) tests utilized have encountered with many criticisms.
Time series properties effect on sensitivity of tests. Also, the most of studies
suppose that time series data are stationary and because of this result havent
utilized from suitable estimation. The main objective of this study is to find
whether in Iran economical growth is a factor for energy consumption or energy
consumption can make economical growth field from direct and indirect channels
such as increasing total consumption, increasing profitability, promotion efficiency
and etc. Therefore, this study tries to find causality relationship between
energy consumption in there area: total consumption, housing consumption and
commercial consumption and industrial consumption and GNP.
MATERIALS AND METHODS
Theoretical foundation of relationship between total product and energy
consumption: Nowadays addition to labor and capital inputs, also energy
is propounded as one of important inputs in macroeconomic models. Therefore,
production is a function of labor, capital and energy inputs.
||Energy input (Abrishami and Mostafaie, 2000)
Energy input can provide with a collection of factors such as oil, gas, electricity
and etc which named energy carrier (Abrishami and Mostafaie,
2000). Pindyck (1979) believed that energy price effect
on economical growth depend on energy role in products structure. With his standpoint
in industries which energy utilized as an intermediate input in production,
price increase and consequently energy consumption decrease, national product
will be reduced. While a country want to fallow its growth, in spite of rising
in prices and adjusting its economical structure, must concentrate domestic
products supply on investment goods which proportionally use less energy. Therefore
inside of parts also demand for investment goods and none investment goods cause
structural changes and investigations which can reduce energy consumption reserving
suitable and lead economic to a side which reduce consumption intensity and
Causality concept: Although regression analyzing can study dependence
of one variable to other variables, but necessarily dont give causality means.
In detail of causality means this question is propounding whether can find a
statistically causality relationship between two variables that have regency
and primacy. In the answer of this question a test which inclusive estimate
below regression defined by Granger (1986, 1988):
||If sum of estimated lagged coefficient yt
in Eq. 2 statistically opposed zero (Σαi
= 0) and sum of estimated lagged coefficient xt in Eq.
3 equal zero (Σδi = 0) there is a aside causality
from yt to xt
||If inverse of above condition happened there is causality from xt
||If sum of coefficient xt and yt in both of
regression is statistically significant and opposed zero then two
variables are independent
Data of this study for 1970-2000 collected from power ministry statistic yearbook
of Iran and to improve results, variables utilized with logarithm form. In this
study LTEC means total energy consumption, LREC is housing and commercial energy
consumption, LIEC is industrial energy consumption, LGDP is gross national product
and LCPI is price index. In this study first with utilized Augment Dickey Fuller
(ADF) test for unit root, stationary of variables have studied (Dickey
and Fuller, 1979, 1981). Then with utilized Johansen
cointegration test (Johansen, 1988, 1992),
number of cointegration vectors between variable national product, energy consumption
in different parts and price level have studied and side of causality between
variables have studied from Granger opinion with utilized three vector error
correction models that each of them inclusive one of the energy consumption
parts with two variables: national product and price level and according to
it, exogenous and endogenous of each variables have estimated. This study is
done in Iran in 2007.
RESULTS AND DISCUSSION
According to stationary augment Dickey Fuller test (Table
1) obtained that all variables in their level are non-stationary and
in type of first difference are stationary. This means those variables
are I (1). In the next step by using of Johansen-Julius, cointegration
test (according to λmax, λtrace indexes)
have done. In this study three model have studied where each of them include
three variables. This test has made for each of pattern. First model including
LCPI, LGDP, LTEC, second model including LREC, LGDP, LCPI and third model
including LIEC, LGDP, LCPI.
According to Table 2 information, tests results for
each pattern show that each of them will have a cointegration vector.
After determine number of cointegration vectors for each model, can estimate
error correction equations. Error correction equations include short run
and error correction component processes. Now, by using Wald test can
determine causality relationship between variable separately in each of
model. Estimated results of this test showed in Table 3.
Three first column (from left side) show χ2 statistic about
significantly test for summation of lags which actually can say this explain
short run causality.
|| Estimated results of stationary test
|| Results of determining number of cointegration vectors
|| Results of Wald test for determining of long and short
|*Significant at 5% level, **Significant at 10% level
Fourth column shows error correction coefficient and its
t-test. Last three columns contemporary consider short run relationship and
cointegration vector and its test which can consider as long run causality.
According to results revealed that in total energy consumption equation, Null
hypothesis based on being zero sum of LPCI and lagged LGDP coefficient has reject
in 5% significant level. This matter show that in short run price index and
gross national income effect on total energy consumption in Iran and their changes
can influence total consumption. According to positive relationship between
total energy consumption and these two variables can say that in short run with
increasing price indexes or gross national products, total energy consumption
increase too. Also, it shows that in long run, price index and gross domestic
product are effective factors on total energy consumption. In fact, GDP increase
directly through increasing society income and therefore increasing on energy
demand. With increasing investigation and therefore increase demand for energy
can effect on energy consumption. In equation GDP and CPI in model one observes
that none of variables have significant effect on them either short or long
run. Therefore, these two variables are weak exogenous for first pattern. This
subject show that error correction term is not being significant of each equation.
In second model which studied relationship between housing energy consumption,
total product and price index, although total product doesnt influence housing
and commercial consumption but in long run have effect on housing and commercial
consumption. This subject is completely compatible with reality because increasing
incomes in long run demand will be increase for energy directly and indirectly.
In long run housing and commercial consumption and also gross domestic production
measure having significant influence one price index. This event can be consider
as a result of inflation pressure of increasing total demand which makes of
economical growth and increasing energy consumption in long run. Also prices
index in short and long run influence on housing and commercial consumption.
In third pattern industrial energy consumption has studied with gross domestic
production and prices index. Also in this model total product, either in short
or long run takes account as an effective factor on energy consumption in industrial
level. This matter show that with increasing domestic product, needed to energy,
increased specially on industrial sector. Generally results of this study show
that energy consumption in three fields: total consumption, housing, commercial
consumption and industrial consumption mustnt receive exogenous from gross
domestic production and inflation rate. Energy consumption in Iran havent key
role on gross domestic production and domestic production in Iran mainly depend
on other factors which energy consumption is the less important of it rather
than other. In fact, these results show to us that about Iran there is aside
causality from domestic product to energy consumption. Results of this study
confirm results of the most studies. With respect to results, government must
do optimum pricing in energy sector in sub sectors of commercial, housing and
industrial in parallel to economic growth and increasing of gross national domestic
product. This strategy cause optimum management in this area. In other words,
energy consumption must not consider exogenous of GDP and inflation rate. In
fact, impose stabilization energy price policy in economical growth conditions
are not good phenomena and will encourage energy demand dimension. So, government
must change energy price policy towards variable pricing based on amount of
consumption especially in peak and load duration.
Abrishami, H. and A. Mostafaie, 2000. Study relationship between economical growth and oil products in Iran. Knowledge Dev. J., 11: 46-46.
Aqeel, A. and M.S. Butt, 2001. The relationship between energy consumption and economic growth in Pakistan. Asia Pacific Dev. J., 8: 101-110.
Direct Link |
Chiou-Wei, S.Z. and Ch. F. Chen and Z. Zhu, 2008. Economic growth and energy consumption revisited-Evidence from linear and nonlinear granger causality. Energy Econ., 30: 3063-3076.
CrossRef | Direct Link |
Dickey, D.A. and W.A. Fuller, 1979. Distributions of the estimators for autoregressive time series with a unit root. J. Am. Stat. Assoc., 74: 427-431.
Direct Link |
Dickey, D.A. and W.A. Fuller, 1981. Likelihood ratio statistics for autoregressive time series with a unit root. Econometrica, 49: 1057-1072.
Direct Link |
Emadzadeh, M., A. Sharifi, R. Dalaliasfahani and M. Safdari, 2002. An analysis of energy growth trend in OECD countries. Commercial Res. J., 28: 95-118.
Engle, R.F. and C.W.J. Granger, 1987. Cointegration and error-correction: Representation, estimation and testing. Econometrica, 55: 251-276.
Direct Link |
Fatia, K., O. Les and F.G. Scrimgeour, 2004. Modeling the causal relationship between energy consumption and GDP in New Zealand, Australia, India and Indonesia: The Philippines and Thailand. Math. Comput. Simulat., 64: 431-445.
Direct Link |
Granger, C.W.J., 1986. Developments in the study of cointegrated economic variables. Oxford Bull. Econ. Stat., 48: 213-228.
CrossRef | Direct Link |
Granger, C.W.J., 1988. Developments in a concept of causality. J. Econometrics, 39: 199-211.
Direct Link |
Jinke, L., S. Hualing and G. Dianming, 2008. Causality relationship between coal consumption and GDP: Difference of major OECD and non-OECD countries. Applied Energy, 85: 421-429.
Johansen, S., 1988. Statistical and hypothesis testing of cointegration vectors. J. Econ. Dyn. Control, 12: 231-254.
Johansen, S., 1992. Cointegration in partial systems and the efficiency of single equation analysis. J. Econometrics, 52: 389-402.
Direct Link |
Masih, A.M.M. and R. Masih, 1997. On the temporal causal relationship between energy consumption, real income and prices: Some new evidence from Asian energy dependent NICs based on multivariate cointegration/vector error correction approach. J. Policy Model., 19: 417-440.
Direct Link |
Pindyck, R.S., 1979. Structure of World Energy Demand. MIT Press, Cambridge, MA., ISBN-10:0-262-66177-2.
Zamani, M., 2007. Energy consumption and economic activities in Iran. Energy Econ., 29: 1135-1140.