Renewable and Sustainable Energy Reviews ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews

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

Economic development and the energy intensity of human well-being:

Evidence from the GCC countries
Osama D. Sweidan a,n, Ahmed A. Alwaked b
a
College of Business and Economics, Department of Economics and Finance, United Arab Emirates University, P.O. Box 15551, Al-Ain, UAE
b
Faculty of Economics and Administrative Sciences, Department of Economics, The Hashemite University, P.O. Box 150459, Al-Zarqa 13115, Jordan

art ic l e i nf o a b s t r a c t

Article history: This paper empirically investigates the effect of economic development on the energy intensity of
Received 12 November 2014 human well-being (EIWB) in the Gulf Cooperation Council (GCC) during the period (1995–2012). We
Received in revised form employ a time-series cross-sectional Prais–Winsten regression model with panel-corrected standard
30 April 2015
errors (PCSE). We conclude that economic development has had a significant positive impact on EIWB
Accepted 1 June 2015
during the period (1995–2012). The results indicate that the stress on the environment had increased in
a noticeable manner from 1995 until 2006. However, over the period 2007–2012, the stress went back to
Keywords: the level observed in 1995. This observation is unfortunate from the perspective sustainable economics.
Sustainability economics & 2015 Elsevier Ltd. All rights reserved.
Energy consumption
Human well-being
Panel analysis
GCC countries

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2. Literature review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
3. Some facts about the GCC countries. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
4. Data and methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
4.1. Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
4.2. Control variables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
4.3. Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
5. The results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
6. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

1. Introduction limited supplies of coal it enjoyed. Likewise, in 1972, Meadows and

others looked into the sustainability of the entire industrial civiliza-
Historically, sustainable economics is a fairly old field engaging tion, given the ultimate finiteness of the planet's capacities to provide
many economists. In 1798, Robert Malthus expressed concerns material inputs to modern economies [38]. With recent global
regarding the unavoidable rise in Britain's population to a level warming, global pollution, global water crisis, and many such aspects
beyond what could be sustained by the finite amount of land the of global environmental threat or stress, concern is growing regard-
country had. In 1865, Stanley Jevons questioned how Britain's ever- ing the issue of sustainability. The term ‘sustainability’ refers to the
increasing energy consumption could be sustained on the basis of the standard behavior with respect to the way humans should act
towards environment, and how they are responsible towards one
n
Corresponding author.
another and generations to come. Baumgartner and Quaas [5]
E-mail addresses: [email protected] (O.D. Sweidan), defined sustainability economics as being “ethically founded in the
[email protected] (A.A. Alwaked). idea of efficiency, that is non-wastefulness, in the use of scarce

http://dx.doi.org/10.1016/j.rser.2015.06.001
1364-0321/& 2015 Elsevier Ltd. All rights reserved.

Please cite this article as: Sweidan OD, Alwaked AA. Economic development and the energy intensity of human well-being: Evidence
from the GCC countries. Renewable and Sustainable Energy Reviews (2015), http://dx.doi.org/10.1016/j.rser.2015.06.001i
2 O.D. Sweidan, A.A. Alwaked / Renewable and Sustainable Energy Reviews ∎ (∎∎∎∎) ∎∎∎–∎∎∎

resources for achieving the two normative goals of (1) the satisfac- The drivers could be categorized into three types: economic (GDP/
tion of the needs and wants of individual humans, and (2) justice, capita, share of exports in GDP), demographic (population growth,
including justice between humans of present and future generations urbanization) and geographic (climate, population density). Contrary
and justice towards nature, in the setting of human–nature relation- to many previous studies, Lamb et al. [30] focused on the differences
ships over the long-term and inherently uncertain future”. Logically, between consumption-based and production-based emissions. They
the main goal of economic development ought to be the improve- showed that the drivers of consumption-based carbon emission are
ment human well-being. At the same time, the mechanics of similar but not the same as the drivers of production-based carbon
economics development, i.e., using fossil fuels, will generate a threat emissions. Moreover, they showed that certain countries are able to
to the environment, which will affect human well-being negatively. achieve higher levels of human well-being while imposing relatively
Implicitly, economic sustainability imposes a challenge on the little stress on the environment.
relationship between the environment and economic development Second, Kuznets [29] analyzed the relationship between eco-
[55,26,42]. Ehrlich and Holden [14] came up with a formula to nomic growth and income inequality. He argue that, as gross
describe the effect of human economic activity of the environment. domestic product per capita (GDPPC) increases to higher levels,
This formula is known in the literature as IPAT or ðI ¼ PAT Þ, where income inequality first increases until it reaches a maximum and
the impact on environment ðI Þ equals the product of Population ðP Þ, declines thereafter. This implies that the relationship between those
Affluence ðAÞ, and Technology ðT Þ. They believe that population two economic variables is quadratic; specifically, an inverted U-
growth and rise in affluence are the primary forces driving negative shaped function. In the early 1990s, a group of economists
environmental impacts. They consider technology to be a neutral or [18,46,44,17,51,50] claimed that Kuznets's idea could be applied
even mildly beneficial element. On the contrary, Commoner [11] on the environment3. They suggested that as GDPPC of a country
views technology as the main destructive force1. increases, the environmental stress placed by the country increases
The present paper focuses on understating the effect of eco- initially until reaches a turning point after which any further
nomic development on the amount of energy used per unit of growth decreases environmental stress. This is known in the
human well-being (EIWB) in the Gulf Cooperation Council (GCC), literature as the environmental Kuznets curve (EKC).
whose members are Bahrain (BA), Kuwait (KW), Oman, (OM), Qatar This dramatic change in the relationship is due to many factors.
(QA), Saudi Arabia (SA), and United Arab Emirates (UAE). The paper Stern [50] argued that there are many factors leading to such a
contributes to the literature from two vital dimensions. First, it behavior, changes in the scale of the economy, changes in economic
highlights the relationship between economic development and structure, changes in technology, changes in the input mix, envir-
EIWB in the GCC countries who rely heavily on the environment to onmental regulation, awareness, and education. Dietz et al. [13]
generate economic growth. A unique feature of these countries is added two dimensions: shifts in preferences as basic needs are met
that they are major producers of energy in the world, and thus the and ecological modernization theory, which suggests that the
price of energy is relatively cheap compare to other countries. As a process of modernization leads to a broadening of the scope of
result, the finding of the current paper gives an indicator if rationality. In other words, the scope moves from achieving
possessing an energy natural resources turns to be a curse on instantaneous economic goals to achieving broader or sustainability
economic sustainability. Second, based on our knowledge, it appears targets which encompasses environmental externalities. Conse-
that such a study focusing on GCC countries has so far been missing. quently, the environmental impacts will be reduced.
Technically, we are trying to check if the economic growth in the Originally, the EKC framework was applied both theoretically
GCC countries is good to the environment or not. By contrast, if and empirically on emission of pollutions ([32,21,45,22,36,52,31];
economic growth in these countries is harmful to the environment, and [50]). Thus, the main dependent variable in econometric work
they face a trade-off between the positive effect of economic used to be emissions per capita. Later, Dietz et al. [13] defined the
growth and environmental stress. Put slightly differently, if the dependent variable as ecological intensity of well-being (EIWB),
results show just a positive relationship between economic growth defined more specifically as the ratio between ecological footprint
and environmental stress, there is no change in the structure or and life expectancy (LE) at birth. In a more recent work, Jorgenson
technology of the economy. The rest of the paper is organized as et al. [24] redefined EIWB as the ratio of energy consumption per
follows. Section 2 reviews relevant literature. Section 3 presents capita (ECPC) to LE at birth. Both definitions are consistent with the
some facts about the GCC countries. Section 4 introduces the data literature, where the dependent variable is always measure an
and methodology adopted in our work. Section 5 discuss the results. undesirable outcome (EIWB¼ Stress/Human Well-Being). It mea-
Conclusions are made in Section 6. sures the amount of environmental stressor generated per unit of
well-being. More efficient nations have small ratios while less ones
have bigger ratios. For the purpose of the present paper, we adopt
2. Literature review the definition of Jorgenson et al. [24],4.

A substantial amount of work on sustainable development has

been reported in literature. We will structure our review around
two pioneering works. 3. Some facts about the GCC countries
First, in their seminal work, Mazur and Rosa [35] demonstrated
that the energy consumption across nations had gone well beyond GCC countries rely heavily on the revenues of oil and gas as a
the traditional lifestyle norms2. This work pioneered research on significant source of economic growth. In addition, they use a
how to increase human well-being while reducing the stress humans portion of these revenues to build a first-class infrastructure and
place on the environment [23,12,1,13,48,28,49]. Lamb et al. [30] used improve social indicators. Further, these countries have been
a novel clustering technique to pursue this theme of research. In this
technique, countries are grouped according to their drivers. In 3
This strand of research was originally developed by Grossman and Krueger's
particular, six drivers of national carbon emissions are considered. [18] while studying the potential impacts of North America Free Trade Agreement
(NAFTA), and Shafik and Bandyopadhyay's [46] groundbreaking study presented in
the 1992 World Development Report. For more details, see Stern [50].
1 4
For more details see Carson [9]. The reason behind this choice is that the data of ECPC and LE is available for
2
Recently, this idea has been confirmed by Mazur [34] and Steinberger and us for free. However, the data of ecological footprint has a high cost and this
Roberts [49]. research is not funded.

Please cite this article as: Sweidan OD, Alwaked AA. Economic development and the energy intensity of human well-being: Evidence
from the GCC countries. Renewable and Sustainable Energy Reviews (2015), http://dx.doi.org/10.1016/j.rser.2015.06.001i
 O.D. Sweidan, A.A. Alwaked / Renewable and Sustainable Energy Reviews ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 3

-1

-2

-3
82 84 86 88 90 92 94 96 98 00 02 04 06 08 10 12
OP UAE SA
OM QA BA Fig. 3. Live expectancy at birth.
KW

Fig. 1. Economic growth and growth of oil prices during (1980–2012).

Fig. 4. Ratio of energy consumption per capita to life expectancy.

Fig. 2. Energy consumption per capita (KG of Oil Equivalent).

putting extensive efforts into diversifying their economies5, but still

oil revenues have remained the main basis for economic growth.
(Fig. 1) illustrates the economic growth6 observed in the GCC
countries and the growth of oil prices during the period 1980–
2012. The following observations are noteworthy. First, the eco-
nomic growth and the oil prices growth are highly fluctuating.
Second, following the great recession of 2007–2008, the economic
growth rate in all the GCC countries either deteriorated or slowed-
down in 2009 when the price of oil registered a considerable drop.
Rutledge [43] argued that during boom periods oil resource
endowment in GCC countries had led to several bouts of the Dutch
disease. This disease is an economic concept explaining the rela- Fig. 5. Real GDP per Capita (In Thousand U.S. $).
tionship between an extensive exploitation of a natural resource,
such as crude oil, and the decline in the share of the manufacturing
sector or agriculture in the economy. Exploiting a natural resource he found only one relationship prevailing in long-run, which is a
generates huge revenues, so the nation's domestic currency statistically significant positive unidirectional relationship running
becomes stronger (currency appreciation) compared to the other from real output to ECPC.
nations' currencies. As a result, the exports from the manufacturing (Fig. 2) shows the panel data of ECPC for the GCC countries during
or other domestic sectors become more expensive, which will make the period 1995–2012. The line plots for each country can be classified
these sectors less competitive. One consequence of the Dutch into two categories. The first shows an upward trend in ECPC, i.e., OM
disease in the GCC countries has been that, until recently, the role and SA. The second one displays an upward drift then a downward
of the non-oil private sector was limited. trend, i.e., UAE and BA. (Fig. 3) presents the panel data of LE. Note that
As for the relationship between ECPC and real output in the the GCC countries have achieved a remarkable improvement in the
GCC countries, Al-Iriani [3] investigated the causality relationship areas of public health and human well-being during the period 1995–
between GDP and ECPC in the six GCC countries. He found a 2012. (Fig. 4) displays the panel data of the EIWB. The line plots show
unidirectional causality running from GDP to ECPC. In the same trends practically identical to those found in ECPC for the same
vein, Sweidan [54] explored the relationship between ECPC and countries. This implies that some of the GCC countries have exhibited
real output in the UAE economy. He found that, in the short-run, more stress (moving from more to less efficient) on the environment
there was a statistically significant bidirectional positive relation- during the period 1995–2012. (Fig. 5) presents the panel data of the
ship between ECPC and real output in the UAE economy. However, GDPPC for the GCC countries from 1995 to 2012. The line plots of
GDPPC for the GCC can be divided into three groups. The first displays
an upward increase in the GDPPC, i.e., SA and QA. The second exhibits
5
For more readings about economic diversification in the GCC countries see a downward decrease in the GDPPC, i.e., UAE. The third has a
Hvidt [20], Strum et al. [53], Malaeb [33], Fasano and Iqbal [16].
6
Based on our source of data, economic growth for Qatar is not available
fluctuating trend in terms of GDPPC, i.e., BA, KW, and OM.
during the period (1980–1994), likewise, the economic growth for Kuwait is
missing during the period (1980–1996).

Please cite this article as: Sweidan OD, Alwaked AA. Economic development and the energy intensity of human well-being: Evidence
from the GCC countries. Renewable and Sustainable Energy Reviews (2015), http://dx.doi.org/10.1016/j.rser.2015.06.001i
4 O.D. Sweidan, A.A. Alwaked / Renewable and Sustainable Energy Reviews ∎ (∎∎∎∎) ∎∎∎–∎∎∎

4. Data and methodology between GDP per capita and each yearly dummy variable as well
as the country-specific and time-specific dummy variables are not
4.1. Data reported in the table. The distribution of skewness and kurtosis for
all the variables were within the acceptable ranges8, so we did not
To achieve the goal of the current paper, we used a perfectly need to transform the data by using the natural logarithm.
balanced sample data for the six countries GCC countries. The data
use in our panel analyses covers the period 1995–2012. Specifically, 4.3. Methodology
we had 108 observations. The source of the data was the database
on World Bank World Development Indicators (WBWDI) via Data We employed the time-series cross-sectional Prais–Winsten
Stream. The data extracted were: ECPC (kilogram of oil equivalent), regression model with panel-corrected standard errors (PCSE). This
LE data and GDPPC measured by constant 2005 U.S. dollars. methodology has been recommended by Beck and Katz [6] because
the feasible generalized least squares (FGLS) method, first described by
Parks [39], produces dramatically incorrect standard errors. When
4.2. Control variables
computing the standard errors and the variance–covariance estimates,
PCSE within STAT 13 software assumes that, by default, the distur-
The model of the present paper, the two way fixed effect model, is
bances are heteroskedastic and contemporaneously correlated across
able to explain a high portion of the variation in the dependent
panels. Since we use 18 annual observations for the six GCC countries,
variable. Despite that, we also included 3 time-variant control variables
we corrected for first-order autocorrelation, AR(1), within panels.
to make our results sounds logical. Those variables are known in the
We estimated our model by the two-way fixed effect technique.
literature are related directly to the research are of environment and
This meant that we were controlling for two dimensions; country-
human well-being. We used the following variables:
specific and time-specific disturbances. We used the following
First, health expenditures as a % of GDP: it is expected that the
general form for our model:
increase in health expenditure relative to the size of the domestic
economy will increase LE and enhance human well-being, thus Y it ¼ θX it þui þ εt þ eit ð1Þ
reducing EWIB. The empirical results on this part are mixed. For where Y it represents the dependent variable for each country at each
example, Elola et al. [15] discovered that high level of health care time period, θX it is the vector of coefficients for explanatory variables
expenditures were associated with higher LE for females in for each country that vary over time, ui is the country-specific
Western Europe. Likewise, Kennelly et al. [27] concluded a positive disturbance term, εt stands for the time-specific disturbance term
relationship between government health expenditures and LE in a that is constant across all countries, and eit denotes the unique
sample of high income nations. On the contrary, Jorgenson et al. disturbance term to each country at each point in time. In the
[24] found statistically insignificant relationship between govern- estimation process, we utilized dummy variables to control for
ment health expenditures and EWIB in a sample of 12 Central and country-specific disturbances ðui Þ and time-specific disturbances
Eastern European (CEE) nations, this implies there is no relation- ðεt Þ. The former controls for potential unobserved characteristics
ship between government health expenditures and LE in CEE. for each individual entity that may or may not influence the
Similarly, Shandra et al. [47] found that government health explanatory variables. In literature, this is known as the time-
expenditures does not have a significant effect on infant mortality invariant characteristics within countries (or country- specific inter-
rates in developing nations. A possible theoretical explanation is cepts). The latter act as controls for any unobserved features existing
that the increase in government spending on health might not over the time-period (time-specific intercepts). More precisely, the
improve the outcomes because the public spending crowds out two-way fixed effect technique concentrates on two simultaneous
private sector provisions [7,40]. The source of this variable tasks: explaining the outcomes within-country variation, and redu-
was WBWDI. cing the likelihood of biased model estimates resulting from similar
Second, exports of goods and service as a percentage of GDP: time trend9.
this variable represents the openness and economic integration The specific form of the present paper using the two-way fixed
with the rest of the world. Theoretically, this variable has two effect technique is as follows:
effect channels. It is predicted that the increase in the level of
exports implies more energy use which should increase EIWB EIWB ¼
[26,41]. On the contrary, other economists state that export- θ1 GDP per capitait þ θ2 GDP per capitait  Year1996t
oriented production enhance economic development, raises wages þ … þ θ18 GDP per capitait
and income for the private sector and the government, improve Year 2012t þ θ19 Year 1996t þ … þ θ35 Year 2012t
life expectancy, and thus enhances the well-being of nations or þ θ36 Health Expenditureit þ θ37 Democratizationit
reduces EIWB. However, the empirical work is not decisive
þ θ38 Exportsit þui þ eit ð2Þ
[24,8,37]. This variable was also extracted from WBWDI.
Third, the nature of the political system inside the country or This model can be divided into the following parts: EIWB is the
democratization. Literature suggests that democracy plays an impor- dependent variable, GDP per capita (θ1 GDP per capitait ), the interac-
tant role in enhancing the standard of living and LE in a country tion for GDP per capita and the dummy variables for each
[10,56]. As a result, the expected effect of the political regime on year (θ2 GDP per capitait  Year 1996t þ …þ θ18 GDP per capitait 
EIWB is negative. We used POLTIY2 as a proxy for democracy in the Year 2012t ) where 1995 is the reference year, the period specific
GCC countries. This proxy ranges between -10 and þ10; -10 means a intercepts (θ19 Year 1996t þ … þ θ35 Year 2012t ). The control variables
strongly dictatorial regime and þ10 represents a strongly demo- include the following: health expenditures as a % of GDP
cratic regime. The source of POLITY2 proxy is the Integrated Net-
work for Societal Conflict Research (INSCR)7. 8
We followed the rule of thumb proposed by [19], which states that the
Table 1 presents the descriptive statistics for the variables acceptable ranges for skewness (SK) and kurtosis (KUR) is 7 3 of the standard error
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
included in the model adopted the current paper. The interactions for each. The standard
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi error for SK ¼ ð6=108Þ ¼ 0:236 and the standard error for
KUR¼ ð24=108Þ ¼ 0:471. Thus, the acceptable ranges are þ 0.708–0.708 and
þ1.413–1.413, respectively.
7 9
The website of the Network is: http://www.systemicpeace.org/inscrdata. For more details, see Jorgenson et al. [24], Allison [2], Baum [4], and
html. Wooldridge [57].

Please cite this article as: Sweidan OD, Alwaked AA. Economic development and the energy intensity of human well-being: Evidence
from the GCC countries. Renewable and Sustainable Energy Reviews (2015), http://dx.doi.org/10.1016/j.rser.2015.06.001i
 O.D. Sweidan, A.A. Alwaked / Renewable and Sustainable Energy Reviews ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 5

Table 1 command. The results show that we can reject the null hypothesis
Descriptive statistics. that “all years coefficients are jointly equal to zero”11.
The results illustrate that the influence of GDP per capita
Item Mean Standard Skewness Kurtosis
deviation (economic development ) is positive and statistically significant
(1.204), besides all interactions between economic development
EWIB 2658.87 53.27  0.03  0.96 and the yearly dummy variables are positive and statistically
GDP per Capita (in 27.90 15.27 0.67  0.93 significant with the exception of the interaction between the level
thousands)
Health Expenditures as a % 3.19 0.70 0.01  1.12
of economic development for the years 2007–2012. Thus, the
of GDP relationship between EIWB and development in GCC countries is
Democratization  8.54 1.24 0.13  1.11 positive and appears to vary in magnitude through time. Putting it
Exports as a % of GDP 59.38 13.33 0.33  0.52 more simply, economic development in the GCC countries creates
more stress on the resources, and its magnitude changes
over time.
(θ36 Health Expenditureit ), the political regime (θ37 Democrati In the interest of more detailed understanding of the effect of
zationit ), and exports as a % of GDP (θ38 Exportsit ), ui is the country- GDP per capita on EIWB, (Fig. 6) plots and reports the estimated
specific disturbance term, eit is the unique disturbance term to each influence of economic development for each year. The first
country at each point in time. It is useful to mention that the coefficient (1.204) plotted and reported in (Fig. 6) is the effect of
coefficient for GDP per capita is the unit change in the dependent economic development on EIWB in 1995 and increases continu-
variable (EIWB) in 1995 (the reference year) for each unit increase in ously until 1999, where the estimated coefficient of economic
GDP per capita for the same year. The overall effect of GDP per capita development is 2.207. Subsequently, this influence declined to 1.81
for the subsequent time points (i.e., 1996,1997, …, 2012) equals the by 2000. After that, the estimated impact of economic develop-
sum of the coefficients for GDP per capita (in the reference year, ment on EIWB had started to rise until it touched 2.841 in 2004.
1995) plus the appropriate interaction term if it is statistically Thereafter, the influence declined to the same level as observed for
significant [24,2]. 1995 (1.204).
In our data sample, the coefficient of variation (standard devia- Next, the yearly affects during (2007–2012) became statistically
tion/mean) of the ECPC was 0.50 and the range was 2829.36– insignificant differ from zero, which suggests that the effect of
23109.77. As for LE, the coefficient of variation was 0.025 and the 1995 had prevailed during this period of time.
range was 69.99–78.45. The coefficients of variation indicated that As for the control variable, the results presented in (Table 2) show
the relative variation in the ECPC, the numerator, was much larger that the effects of health expenditures as a % of GDP, exports as a % of
than variation in LE, the denominator. This meant that variation in GDP, and democratization are statistically insignificant differ from
ECPC was driving variation in the ratio. We clearly needed to resolve zero. This suggests that economic development is the sole element
this problem before using the ratio as dependent variable. We that adds more stress on the environment over the time horizon.
followed Dietz et al. [13] and Jorgenson et al. [24] in the solution Overall, our results seem to present a clear picture regarding
process. As stated by Dietz et al. [13], this problem had been the relationship between economics development and the stress
resolved earlier by a researcher from the New Economics Founda- on the environment (EIWB) in the GCC countries. Our paper has
tion. The mechanics of the solution process involved restraining the come up with two main finding. Firstly, economic development in
coefficient of variation for the numerator and denominator to be the GCC countries has been the sole component that contributed
equal by adding a constant to one of them. The process shifted the negatively to the stress on the environment. Secondly, economic
mean without changing the variance, thereby allowing for the development has had a positive impact on EIWB during the period
equalization of the coefficient of variation for the two variables. In 1995–2012. This conclusion is not encouraging from a sustain-
our sample data, the coefficients of variation for the two variables ability standpoint. The results have indicated that the stress on the
could be made equal by adding 188394.2 to the ECPC10. Hence, the environment increased in a noticeable manner from 1995 until
adjusted EWIB could be calculated as: 2006, then during the period 2007–2012 the stress returned to its
  previous level as in 1995.
ECPC þ 188394:2
EIWB ¼ n100 ð3Þ The majority of the previous empirical work on EIWB economic
LE
development concluded that economic development discourages
economic sustainability or at least do not make it better [13,28].
However, recently some studies, i.e. Jorgenson et al. [24], found
5. The results that economic development improves economic sustainability.
The result of our paper is consistent with the first group. We
The results from the estimated mode are reported in (Table 2). believe that our result is logical and consistent with the economic
Rho shows that 86% of the EIWB variance is explained by country- trend and facts on the ground in the GCC countries for the
specific disturbances ðui Þ. The estimated R-square is significantly high following three reasons. First, our measure of EIWB focuses on
(0.99), suggesting that the model has been able to explain 99 percent energy consumption. Second, during the period of our study the
of the overall variation in EIWB. Such an unusually high R-square GCC witnessed a high economic growth rate due to the noticeable
value is mainly the result of the country-specific and year-specific increase in the price of oil. At the same time, these group of
intercepts that serve as fixed effects. Put simply, the very high R- countries is working to diversify the economy to reduce the share
square is due to using the two-way fixed effect. Similar results have of the oil sector as a percentage of GDP. Third, the GCC countries
been reported by Jorgenson and Clark [25] and Jorgenson et al. [24]. have a comparative advantage in producing energy which is
To decide if the time fixed effects (the period specific intercepts) are considered relatively cheap to other nations, thus the consumption
needed while running our model, we perform a joint test to check if level of energy is high. The bottom line is that all these reasons
the dummies for all years are equal to zero by using the “testparm” contribute to a higher energy consumption.

10
The coefficient of variation is equal to the standard deviation, S, divided by
the mean M. Using subscripts EC and LE to indicate energy consumption and life
  11
expectancy, respectively, the correction factor CF ¼ ½SEC nM LE =SLE  M EC . The value of the Chi2 (17) ¼229.23, Prob4Chi2 ¼ 0.00.

Please cite this article as: Sweidan OD, Alwaked AA. Economic development and the energy intensity of human well-being: Evidence
from the GCC countries. Renewable and Sustainable Energy Reviews (2015), http://dx.doi.org/10.1016/j.rser.2015.06.001i
6 O.D. Sweidan, A.A. Alwaked / Renewable and Sustainable Energy Reviews ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Table 2 setting of human–nature relationships over the long-term and inher-

The Model Estimate (Dependent variable is EWIB). ently uncertain future”. In view of this, this paper has examined how
one could enhance human well-being while reducing the stress or
Independent variables Coefficients Panel corrected standards
errors threat humans place on the environment.
A large body of work has been reported in literature on
GDP Per Capita 1.204nnn 0.340 sustainable economics. Unfortunately, the GCC countries did
GDP Per Capita  1996 0.543nnn 0.153 receive considerable attention during explorations of the relation-
GDP Per Capita  1997 0.682nnn 0.194
GDP Per Capita  1998 0.824nnn 0.244
ship between economic development and the stress on the
GDP Per Capita  1999 1.003nnn 0.267 environment. The present paper has investigated the effect of
GDP Per Capita  2000 0.606nn 0.288 economic development on energy intensity of human well-being
GDP Per Capita  2001 1.300nnn 0.308 (EIWB) in GCC countries. A unique feature of these countries is
GDP Per Capita  2002 1.642nnn 0.323
that they are a vital producers of energy for the world.
GDP Per Capita  2003 1.542nnn 0.331
GDP Per Capita  2004 1.637nnn 0.340 To achieve the goals of the paper, we utilize a time-series cross-
GDP Per Capita  2005 1.191nnn 0.360 sectional Prais–Winsten regression model with panel-corrected stan-
GDP Per Capita  2006 0.698n 0.373 dard errors (PCSE). We use data for the six countries GCC countries for
GDP Per Capita  2007 0.617 0.408 the period 1995–2012 for panel analysis. Accordingly, we had 108
GDP Per Capita  2008 0.029 0.439
GDP Per Capita  2009  0.173 0.480
observations. The source of the data was the World Bank World
GDP Per Capita  2010  0.066 0.520 Development Indicators (WBWDI) database via Data Stream.
GDP Per Capita  2011 0.179 0.487 We believe that our paper has succeeded in presenting a fairly
GDP Per Capita  2012 0.424 0.503 clear picture of the relationship between economics development
Control variables
and the stress on the environment (EIWB) in the GCC countries. Our
Health Expenditures as a % of 0.503 3.52783
GDP paper has produced two main finding. Firstly, economic develop-
Exports as a % of GDP  0.340 0.2201953 ment in the GCC countries has been the sole factor contributing
Democratization 1.138 1.874875 negatively to the stress on the environment. Secondly, economic
R-Squared 0.99 development has been impacting EIWB significantly positively
Rho 0.86
N 108
during the period 1995–2012. This conclusion however is unfortu-
♯ of estimated coefficients 39 nate from the economic sustainability. Our results have shown that
N/♯ of estimated coefficients 2.77 the stress on the environment had increased in a noticeable manner
between 1995 and 2006, after which (2007–2012) the stress went
Notes: the coefficient of the period specific intercepts are not reported in (Table 2).
back to the level observed in 1995.
nnn
Po 0.01 (one-tailed tests),
nn
On the policy implication side, as long as economic develop-
Po 0.05 (one-tailed tests), and
n
Po 0.10 (one-tailed tests). ment increases the stress on the environment (EIWB), then the
current development policies in the GCC countries are harmful to
economic sustainability, and must be revised or well-adjusted
with new policies to counter the unplanned environmental harms
of development. Besides, we can claim that our positive relation-
ship between economic development and EIWB indicates that the
economies utilize the energy inefficiently. However, policy impli-
cation in the GCC countries should be managed in a cautious
manner if energy consumption leads to economic development.
Finally, if the new development policies do not succeed to improve
the present situation, then we can expand the conclusion and
claim that possessing an energy natural resources turns to be a
curse on economic sustainability.

Acknowledgments

We would like to thank the editor and two anonymous referees

of Renewable & Sustainable Energy Reviews for valuable and helpful
comments. We are responsible for any remaining errors.
Fig. 6. Estimated coefficients of the effect of yearly economic Development
on EIWB.
References

6. Conclusions [1] Alekseyko A, Giedraitis V. Economic growth and energy security: economic
and environmental solutions in a changing Europe. Lap Lambert Academic
Press; 2013.
Sustainability economics is a fairly mature field. Recently, the world [2] Allison P. Fixed effects regression models. Thousand Oaks: Sage; 2009.
economy has started using energy in an extensive way to enhance [3] Al-Iriani M. Energy-GDP relationship revisited: an example from GCC coun-
economic development. Unfortunately, this action has had a negative tries using panel causality. Energy Policy 2006;34:342–3350.
[4] Baum C. An introduction to modern econometrics Using Stata. College Station,
influence on the environment. Economists have observed that sustain-
TX: Stata Press; 2006.
ability economics is “ethically founded in the idea of efficiency, that is [5] Baumgartner S, Quaas M. What is sustainability economics? Ecological
non-wastefulness, in the use of scarce resources for achieving the two Economics 2010;69:445–50.
normative goals of (1) the satisfaction of the needs and wants of [6] Beck N, Katz J. What to do (and not do) with time-series cross-section data.
Am Polit Sci Rev 1995;89:634–47.
individual humans, and (2) justice, including justice between humans [7] Bokhari F, Gai Y, Gottret P. Government health expenditures and health
of present and future generations and justice towards nature, in the outcomes. Health Econ 2007;16:257–73.

Please cite this article as: Sweidan OD, Alwaked AA. Economic development and the energy intensity of human well-being: Evidence
from the GCC countries. Renewable and Sustainable Energy Reviews (2015), http://dx.doi.org/10.1016/j.rser.2015.06.001i
 O.D. Sweidan, A.A. Alwaked / Renewable and Sustainable Energy Reviews ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 7

[8] Brady D, Kaya Y, Beckfield J. Reassessing the effect of economic growth on [33] Malaeb M. Diversification of the GCC economies: analysis of the preceding
well-being in less developed countries,1980–2003. Stud Comp Int Dev debate 1993–2003. Dubai: Gulf Research Centre; 2006.
2007;42:1–35. [34] Mazur A. Does increasing energy or electricity consumption improve quality
[9] Carson TR. The environmental Kuznets curve: seeking empirical regularity and of life in industrial nations? Energy Policy 2011;39:2568–72.
theoretical structure. Rev Environ Econ Policy 2010;4(1):3–23. [35] Mazur A, Rosa E. Energy and lifestyle: massive energy consumption may not
[10] Clark R. World health inequality: convergence, divergence, and development. be necessary to maintain current living standards in America. Science
Soc Sci Med 2011;72:617–24. 1974;186:607–10.
[11] Commoner B. The environmental cost of economic development. In Popula- [36] McConnell K. Income and the demand for environmental quality. Environ.
tion, environment and resources. Washington, DC: U.S. Government Printing Dev. Econ. 1997;2:383–99.
Office; 1972. [37] Moore S, Teixeira A, Shiell A. The health of nations in a global context: trade,
[12] Dietz T, Jorgenson A. Structural human ecology: new essays in risk, energy, global stratification, and infant mortality rates. Soc. Sci. Med. 2006;63:165–78.
and sustainability. Pullman, WA: Washington State University Press; 2013. [38] Pezzey J, Toman M. The economics of sustainability: a review of journal
[13] Dietz T, Rosa E, York R. Environmentally efficient well-being: is there a articles. Resources for the Future, Discussion Paper 02–03; January2002.
Kuznets curve? Appl Geogr 2012;32:21–8. [39] Parks R. Efficient estimation of a system of regression equations when
[14] Ehrlich P, Holden J. Impact of population growth. Science 1971;171:1212–7. disturbances are both serially and contemporaneously correlated. J. Am. Stat.
[15] Elola J, Daponte A, Navarro V. Health indicators and the organization of health Assoc. 1967;62:500–9.
care systems in Western Europe. Am J Public Health 1995;85:1397–401. [40] Rajkyman A, Swaroop V. Public spending and outcomes: does governance
[16] Fasano U, Iqbal Z. GCC countries: from oil dependence to diversification, IMF matter? J Dev Econ 2007;86:96–111.
External Relations Department material. Washington, D.C.: International [41] Roberts T, Parks B. A climate of injustice: global inequality, north–south
Monetary Fund; 2003. politics, and climate policy. Cambridge, MA: MIT Press; 2007.
[17] Grossman G, Krueger A. Economic growth and the environment. Q J Econ [42] Rosa E, Diekmann A, Dietz T, Jaeger C. Human footprints on the global
1995;110:353–77. environment: threats to sustainability. MIT Press; 2010.
[18] Grossman G, Krueger A. Environmental impacts of a North American free trade [43] Rutledge E. Monetary union in the gulf: prospects for a single currency in the
agreement. National Bureau of Economic Research Working Paper 3914, NBER,
Arabian Peninsula. London: Routledge; 2009.
Cambridge MA; 1991.
[44] Selden T, Song D. Environmental quality and development: is there a Kuznets
[19] Hahs-Vaughn D, Lomax R. An Introduction to statistical concepts. third edition.
curve for air pollution? J Environ Econ Manag 1994;27:147–62.
New York: Routledge; 2013.
[45] Selden T, Song D. Neoclassical growth, the J curve for abatement and the
[20] Hvidt M. Economic diversification in GCC countries past record and future
inverted U curve for pollution. J Environ Econ Manag 1995;29:162–8.
trends, Research Paper, Kuwait Programme on Development. Governance and
[46] Shafik N, Bandyopadhyay S. Economic growth and environmental quality:
Globalisation in the Gulf States; 2013.
time series and cross country evidence. Background Paper for the World
[21] John A, Pecchenino R. An overlapping generations model of growth and the
Development Report. Washington, DC: The World Bank; 1992. p. 1992.
environment. Econ J 1994;104:1393–410.
[47] Shandra J, Nobles J, London B, Williamson J. Dependency, democracy, and
[22] John A, Pecchenino R, Schimmelpfennig D, Schreft S. Short-lived agents and
infant mortality: a quantitative cross-national analysis of less developed
the long lived environment. J Public Econ 1995;58:127–41.
[23] Jorgenson A. Economic development and the carbon intensity of human countries. Soc Sci Med 2004;59:321–33.
wellbeing. Nat Clim Change 2014;4:186–9. [48] Steinberger J, Roberts T, Peters G, Baiocchi G. Pathways of human develop-
[24] Jorgenson A, Alekseyko A, Giedraitis V. Energy consumption, human well- ment and carbon emissions embodied in trade. Nat Clim Change 2012;2:81–5.
being and economic development in central and eastern European nations: a [49] Steinberger J, Roberts T. From constraint to sufficiency: the decoupling of
cautionary tale of sustainability. Energy Policy 2014;66:419–27. energy and carbon from human needs, 1975–2005. Ecol Econ 2010;70:425–33.
[25] Jorgenson A, Clark B. The relationship between national-level carbon dioxide [50] Stern D. The rise and fall of the environmental Kuznets curve. World Dev
emissions and population size: an assessment of regional and temporal 2004;32:1419–39.
variation, 1960–2005. PLOS ONE 2013;8:e57107. [51] Stern D, Common M, Barbier E. Economic growth and environmental degra-
[26] Jorgenson A, Clark B. Are the economy and the environment decoupling? A dation: the environmental Kuznets curve and sustainable development. World
comparative international study, 1960–2005 Am J Sociol 2012;118:1–44. Dev 1996;24:1151–60.
[27] Kennelly B, O' Shea E, Garvey E. Social capital, life expectancy, and mortality: [52] Stokey N. Are there limits to growth? Int Econ Rev 1998;39:1–31.
across-national examination. Soc Sci Med 2003;56:2367–77. [53] Strum M, Strasky J, Adolf P, Peschel D. The gulf cooperation council countries
[28] Knight K, Rosa E. The environmental efficiency of wellbeing: a cross national economic structures, recent developments and role in the global economy.
analysis. Soc Sci Res 2011;40:931–49. European Central Bank, Occasional paper series No. 92; 2008.
[29] Kuznets S. Economic growth and income inequality. Am Econ Rev [54] Sweidan O. Energy consumption and real output: new evidence from the UAE.
1955;45:1–28. OPEC Energy Rev 2012;36:287–300.
[30] Lamb W, Steinberger J, Bows-Larkin A, Peters G, Roberts J, Wood F. Transitions [55] York R. Do alternative energy sources displace fossil fuels? Nat Clim Change
in pathways of human development and carbon emissions. Environ Res Lett 2012;2:441–3.
2014;9:1–10. [56] Wickrama K, Mulford C. Political democracy, economic development, disarti-
[31] Lieb C. The environmental Kuznets curve and satiation: a simple static model. culation, and social well-being in developing countries. Sociol Q
Environ Dev Econ 2001;7:429–48. 1996;37:375–90.
[32] Lopez R. The environment as a factor of production: the effects of economic [57] Wooldridge J. Introductory econometrics: a modern approach. third ed..
growth and trade liberalization. J Environ Econ Manag 1994;27:163–84. Mason: Thomson; 2005.

Please cite this article as: Sweidan OD, Alwaked AA. Economic development and the energy intensity of human well-being: Evidence
from the GCC countries. Renewable and Sustainable Energy Reviews (2015), http://dx.doi.org/10.1016/j.rser.2015.06.001i