Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
ENERGY, DEBT, AND GROWTH: TIME SERIES
EVIDENCE ON CAUSALITY IN OECD
EUROPEAN COUNTRIES
Mohamed Awada, Joanna Darwiche, AND Moustapha Badran*
1. Introduction
The interactions between economic growth and debt, already addressed in the
literature, appear more complex than the intuitive interpretation by political
leaders aiming to stimulate growth through the leverage of debt. Regardless of the
economic situation (expansion or recession), we have observed that for certain
countries, the pace of GDP growth is conditioned by their energy consumption.
*Mohamed Awada earned his Doctorate in Economic Sciences from the University of Montpellier.
In his thesis, he examined the interrelationships between energy supply, public debt, and economic
growth as applied to a group of OECD European countries. His research interests include
macroeconomics, finance, and energy economics with a focus on European countries. He has several
scientific articles published in the Journal of Energy and Development and other journals. His teaching
expertise covers several subjects including corporate finance, financial analysis, statistics,
microeconomics, macroeconomics, and financial mathematics. He is currently working as a contractual
lecturer at the Faculty of Economics in Montpellier, primarily teaching a course on private economic
calculus and leading econometrics tutorials.
Joanna Darwiche holds a Ph.D. in Economic Sciences from the University of Montpellier and is
currently an Assistant Professor at the Faculty of Business Administration and Management at Saint
Joseph University of Beirut. Her research focuses on the intersection of macroeconomics, finance, and
geopolitical risks, with an emphasis on Internet globalization, exchange rates, and monetary policy in
emerging markets. She also examines the impact of technological innovations, ESG factors, and energy
dynamics on financial development and inflation. Joanna has been invited to present at conferences in
cities such as Beirut, Montpellier, Brussels, and Istanbul. She also holds a certificate in Risk and
Financial Services from the Chartered Institute for Securities & Investment. (continued)
The Journal of Energy and Development, Vol. 49, Nos. 1-2
Copyright #2024 by the International Research Center for Energy and Economic Development
(ICEED). All rights reserved.
287
Shortages of raw materials and surging commodity prices, linked to the strong
recovery of activity following the COVID-19 crisis, are a good illustration of this.
This aspect is predominantly explored in the literature through a demand-
focused lens, emphasizing energy consumption. Furthermore, researchers typically
examine the interactions between growth and energy consumption and the relation-
ships between growth and debt separately, often neglecting the potential impacts of
energy supply. Therefore, our study underscores the importance of incorporating
energy supply into a comprehensive analysis that integrates all three elements:
debt, growth, and energy.
This research article aims to analyze the causal links between energy supply,
public debt, and economic growth for a group of European countries that are part
of the OECD, a selection of major economies, and high-risk countries. The sample
consists of nine European OECD countries: Belgium, Finland, Germany, Nether-
lands, Spain, Ireland, France, Italy, and Greece, with data collected from the
OECD website. We selected these nine countries to compare major economies
with high-risk countries, heavily indebted countries with less indebted ones, and
northern countries with southern countries. Additionally, we aimed to compare our
results with those in the literature. The availability of data on the OECD site was
also a crucial factor in our selection.
Numerous studies, initiated since the beginning of the financial crisis have
focused on the growth rate of GDP and levels of debt. Initially, these studies have
produced varying results: the increase in public debt contributing to economic
growth (Modigliani, 1961; Diamond, 1965; Saint-Paul, 1992), high levels of debt
negatively affecting economic growth (Poirson et al., 2004), no significant relation-
ship (Schclarek et al., 2004), and an inverse relationship between the variables
(Kumar, 2010). Subsequently, these studies have enabled the development of a
typology of countries (G
omez-Puig and Sosvilla-Rivero, 2015): bidirectional cau-
sality for France and Finland, a unidirectional causality from the percentage change
in sovereign debt to economic growth for Italy, the Netherlands, and Spain, and no
causal relationship concluded for Belgium, Germany, Greece, and Ireland.
In addition, the specialized studies by Giraud and Kahraman (2014) in energy
economics highlight a growing interest in the role of energy in production. This
raises questions about the feasibility of decoupling economic growth from energy
Moustapha Badran holds a Doctorate in Economic Sciences from the University of Montpellier,
with a research focus on corporate finance, capital structure, and financial development. Currently, he
serves as an adjunct lecturer in Economics and Social Sciences at the University of Grenoble Alpes. He
has several scientific articles published in the Journal of Energy and Development and other journals.
His teaching portfolio spans a range of subjects, including theoretical and time series econometrics, as
well as both macroeconomics and microeconomics.
The authors would like to thank anonymous reviewers for their valuable suggestions and helpful
comments, which have greatly enhanced the quality of this paper.
288 THE JOURNAL OF ENERGY AND DEVELOPMENT
consumption. Moreover, it prompts an examination of whether the high depen-
dence on energy imports in certain European countries poses challenges to their
autonomous development. Could the debts accumulated since the beginning of the
21st century be more closely tied to limitations in production growth, exacerbated
by these nations’energy constraints?
To address these questions, we propose an analysis of the interactions between
debt, growth, and energy supply across each selected country, categorizing them based
on their levels of indebtedness and energy independence. Our primary focus is on inte-
grating energy supply dynamics into the understanding of debt and growth relation-
ships. Through this approach, we aim to provide a comprehensive perspective on how
these variables evolve over time, delineating distinct study periods for each country.
In what follows, we will proceed with a presentation of the energy-growth and
debt-growth relationships discussed in the literature, followed by an overview of
the macro-energy balance for each European OECD country studied. After present-
ing the methodology used and the data available, we will highlight the results of
our study and their economic implications.
2. Literature Review
Relationship Between Energy and Growth: The relationship between energy
and economic growth has been a subject of extensive research. Energy is a crucial
input in the production process, influencing the output of goods and services. Sev-
eral studies have examined this relationship, revealing varying results based on dif-
ferent methodologies and country-specific factors.
The debate surrounding the relationship between energy consumption and eco-
nomic growth has given rise to two opposing views. One perspective argues that
energy consumption limits growth. It has also been argued that the potential
impacts of energy consumption on growth vary depending on the economic struc-
ture and growth cycle of the country in question. As economies develop, their
production structure is expected to shift towards services, which are less energy-
intensive activities. This viewpoint has been supported by authors such as Denison
(1985), Cheng (1995), Asafu-Adjaye (2000), and Solow (2016).
The other viewpoint suggests that energy can be both the source and the engine
of economic growth. Increased energy consumption is seen as a consequence of
economic growth. Moreover, energy is considered a key source of economic
growth because numerous consumption and production activities rely on energy as
a fundamental production factor. From a physical perspective, energy consumption
enhances economic productivity and industrial growth, making it essential for the
development of any modern economy (Sacko, 2004).
Empirical studies thus show a unidirectional causality from total energy consump-
tion to economic growth (applied to a group of African countries), as concluded by
289ENERGY, DEBT, AND GROWTH
Akinlo (2008) and Adebola (2011). Another unidirectional causality in the opposite
direction was found in the study by Kraft and Kraft (1978) applied to the United
States for the period from 1947 to 1974, and in Abaidoo’s (2011) study using quar-
terly data over 39 years. Behname et al. (2012) conducted a study that leads to a bidi-
rectional relationship between the two variables. Carminel (2015) concluded no
relationship between energy and growth, explaining that the decoupling between
energy and growth is constrained by concerns over raw material supply. For example,
technologies related to energy needed to extract certain materials are subject to geo-
political constraints. The issue of raw material supply, in turn, limits the deployment
of equipment necessary to improve energy intensity.
In their study of a group of countries, Erol and Yu (1987) attempted to establish
the cause-and-effect relationship between energy consumption and GDP. Using
Granger and Sims’causality test, they arrived at the following conclusions: there is
a unidirectional causality from energy consumption to GDP for West Germany;
bidirectional causality in Italy; and no causality between the two variables for
France, the United Kingdom, and Canada.
Masih and Masih (1996) applied their work to six Asian countries (India, Indo-
nesia, Malaysia, Pakistan, Philippines, and Singapore) using Johansen’s methodol-
ogy, vector error correction model, and variance decomposition. The differences in
results can be summarized as follows: Cointegration between energy consumption
and GDP was found in India, Indonesia, and Pakistan. No cointegration between
the two variables was observed in Malaysia, Singapore, Pakistan, and the Philip-
pines. Energy consumption leads GDP in India (more energy consumption leads to
more growth). GDP causes energy consumption in Indonesia. There is bidirectional
causality in Pakistan. These findings illustrate varying relationships between
energy consumption and GDP across different Asian countries, highlighting the
complexity and diversity of energy-economic interactions in regional contexts.
Chontanawat et al. (2008) evaluated the relationship between the two variables
across a panel of over 100 countries, including 30 OECD member countries and
78 non-OECD member countries, to detect the relationship between energy and
growth. Their findings reveal that energy consumption leads to GDP (more energy
consumption leads to more growth) in 21 out of 30 OECD countries, accounting
for 70%. For non-OECD countries, this relationship is found in 36 out of 78 coun-
tries, accounting for 46%. In summary, there is a more common causality between
energy consumption and GDP for advanced OECD countries. From this perspective,
a policy focused on limiting energy consumption would have a more pronounced
negative impact on the GDP of OECD countries compared to non-member countries.
Ozturk et al. (2010) sought to test the relationship between energy and growth
in four Eastern European countries: Albania, Bulgaria, Hungary, and Romania,
over the period from 1980 to 2006 using the Engle-Granger model. Their conclu-
sions indicate no causality for Albania, Bulgaria, and Romania. However, they
found evidence of bidirectional causality for Hungary.
290 THE JOURNAL OF ENERGY AND DEVELOPMENT
To conclude this section, various studies have explored the complex relation-
ship between energy consumption and economic growth across different countries
and regions. These investigations have revealed the links developed by Jumbe
(2004), Shiu and Lam (2004), Altinay and Karagol (2005), Chen et al. (2007),
Mozmuder and Marathe (2007), Squalli (2007), Apergis and Payne (2010), and
Oztruk et al. (2010), and categorized into four main categories, each with signifi-
cant implications for energy policy:
The Growth Hypothesis: This hypothesis asserts that there is a unidirectional
relationship between energy consumption and economic growth. It argues
that energy consumption plays a pivotal role in economic growth, both as a
direct factor in the production process and indirectly as a complement to
labor and capital. Energy is considered here as a complementary factor of
production alongside traditional factors like capital and labor. Under these
conditions, the implementation of energy policy influences the level of pro-
duction, as indicated by Yu and Choi (1985), Tsani (2010), Belke et al.
(2011), and Destek (2016).
The Conservation Hypothesis: Economic growth leads to an increase in
energy consumption. This hypothesis suggests that a restrictive energy policy
can be implemented in an economy without negative effects on growth. If
there is a unidirectional Granger causality from growth to energy consump-
tion, this hypothesis is supported. Indeed, Paul and Bhattacharya (2004),
Hatemi et al. (2005), and Gelo (2009) argue that energy-saving policies can
be implemented with little or no negative effects on economic growth.
The Neutrality Hypothesis: It implies that there is no cause-and-effect rela-
tionship between energy consumption and economic growth. These two vari-
ables are not correlated. In other words, any increase or decrease in energy
consumption has no effect on economic growth. This means that neither
energy-saving policies nor intensive energy policies influence the level of
wealth creation in an economy, as explained by Jobert and Karanfil (2007).
The Feedback Hypothesis: It asserts that there is bidirectional causality
between energy and economic growth. This means that energy and economic
policies should be implemented jointly. In this case, energy consumption pol-
icies should be developed to avoid any negative impact of energy on growth.
Studies applied to one or groups of countries (e.g., Greece, G7, OECD coun-
tries) support this view, according to Hondroyiannis (2004), Lee et al.
(2008), Mutascu (2016), and Dos Santos Gaspar et al. (2017).
The variability in how energy influences economic growth necessitates tailored
policy approaches that account for local economic conditions, energy infrastruc-
ture, and environmental considerations to foster sustainable development. Simi-
larly, the relationship between debt and economic growth is complex, with debt
serving as both a tool for financing growth and a potential risk to economic
291ENERGY, DEBT, AND GROWTH
stability. Effective management of public debt requires strategies that balance
short-term economic benefits with long-term fiscal sustainability.
Relationship Between Debt and Growth: The empirical literature on this rela-
tionship between debt and growth not only presents ambiguous results but also pri-
marily focuses on the possible impact of high debt levels on economic growth,
often overlooking the potential reverse causality from growth to debt (with rare
exceptions: Ferreira, 2009; Puente-Ajov
ın and Sanso-Navarro, 2015).
However, Bell et al. (2015) find theoretical evidence suggesting that public debt
is likely to accumulate when growth is low. In this regard, with low growth imply-
ing limited public revenues, governments may be compelled to increase their level
of indebtedness to sustain the welfare state, stimulate short-term demand, and fos-
ter long-term growth, according to Feldstein (2014).
Theoretically, neoclassical and endogenous growth models, such as those by
Modigliani (1961), Diamond (1965), and Aizenman et al. (2007), suggest that high
levels of public debt would undoubtedly reduce the rate of economic growth. Other
channels through which public debt may negatively affect long-term growth
include the hypothesis of over-indebtedness (Krugman, 1988; Roubini and Sachs,
1989), the liquidity constraint hypothesis (Moss and Chiang, 2003), the crowding-
out hypothesis (Hansen, 2004), and uncertainty effects (Codogno et al., 2003;
Cochrane, 2011). Another channel through which high indebtedness can have a
negative impact on growth is through long-term interest rates (Elmendorf and
Mankiw, 1999; Tanzi and Chalk, 2000).
Checherita and Rother (2012) analyzed the average impact of public debt on
per capita GDP growth in 12 Eurozone countries over a period of about 40 years
starting from 1970. They concluded that there is a non-linear effect of debt on
growth, with a turning point beyond which the public debt-to-GDP ratio has a neg-
ative impact on long-term growth, estimated to be around 90-100% of GDP.
Panizza and Presbitero (2014) studied the causal effect of public debt on economic
growth in a sample of OECD countries. Their findings align with existing litera-
ture, demonstrating a negative correlation between the two variables.
Mencinger et al. (2014) empirically studied the relationship between public
debt-to-GDP ratios and GDP growth across a panel of 25 EU countries. They
divided the countries into old and new member states and used panel estimation in
a generalized economic growth model augmented with a debt variable. Their find-
ings indicated a statistically significant non-linear impact of public debt ratios on
annual per capita GDP growth rates. They identified a turning point in the debt-to-
GDP ratio where the positive effect of debt accumulation shifts to a negative effect,
approximately between 80% and 94% for old member states and between 53%
and 54% for new member states. This research contributed to understanding the
implications of high public debt on economic activity within the EU.
Finally, some effects associated with financial liberalization, such as increased
bank risk-taking and significant accumulation of external debt, can make a country
292 THE JOURNAL OF ENERGY AND DEVELOPMENT
vulnerable to economic shocks that often lead to severe recessions (Eichengreen
and Leblang, 2003; Nyambuu and Bernard, 2015). Given the theoretical predic-
tions highlighted above, it is somewhat surprising that the conclusion by Reinhart
and Rogoff (2010), stating that countries’debt should not exceed 90% of GDP,
beyond which GDP growth rates decrease significantly, sparked such controversy.
Others have strongly criticized this conclusion, noting that during the period 1946-
2009, countries with a public debt-to-GDP ratio exceeding 90% actually showed
an average annual real GDP growth of 2.2%, not 20.1%.
All the previous research highlights the interest in determining the potential
short-term and long-term relationships between energy, debt, and growth, this will
be addressed in the next section by presenting a macro-energy balance of the sam-
ple consisting of nine European countries. The aim is to provide a comprehensive
overview of the strengths and challenges faced within the European region.
Macro-Energy Balance of the Nine European Countries: We present in this
section the macro-energetic state, which allows us to explain through the literature
the energy situation (energy supply) and macroeconomic situation (public debt and
economic growth) of each selected European OECD country in order to compare it
with the empirical part.
The sample of nine European countries is divided into three categories based on
their level of debt (Benassy, 2017) as outlined in Figure 1. The first category
includes Belgium and Finland, countries that do not exhibit any imbalances. The
second category consists of countries with imbalances: Germany, where public
investment is insufficient; the Netherlands, with private debt; and Spain and Ireland,
with both private and public debt. The three remaining countries form the third cate-
gory, which is characterized by excessive imbalances: France, characterized by pub-
lic debt; Italy, with unemployment and a lack of competitiveness; and Greece,
marked by its fragility following the financial crisis and the sovereign debt crisis.
Belgium: The supply of primary energy is considerable thanks to imports (96%
of energy needs in 2030), divided between oil (the most important energy source,
Figure 1
CLASSIFICATION OF THE NINE EUROPEAN COUNTRIES BY DEBT LEVEL
293ENERGY, DEBT, AND GROWTH
up to 40% of demand) and gas (Marbraix and Van Ypersele de Strihou, 2004).
Goals have also been set to reduce fossil fuel consumption so that it will represent
4% of final demand in favor of gas and electricity (Gusbin and Hoornaert, 2004).
The relationship between debt and growth has been influenced by significant
temporal events. Since 1969-1970, there has been a period marked by inflation,
deteriorating productivity, and unemployment. The two oil shocks of 1973 and
1979 led to a 70% increase in the price of oil, causing growth to stall. Keynesian
reforms resulted in a widened budget deficit, bringing the debt-to-GDP ratio to
106% in 1983. The crisis persisted through the 1980s, leading to increased indebt-
edness and resulting in stagflation. Despite an easing of problems after 2000, the
financial crisis pushed the ratio back up to 100% in 2010 and to 105.2% today.
Despite the negative impact of debt on growth, a high level of debt does not pose
an obstacle to the country’s short-term growth.
Finland: Finland’s energy policy is characterized by the diversification of sup-
ply despite its dependence on foreign countries. International interest in Finnish
energy policy has been very strong for several reasons. Firstly, electricity markets
were liberalized in the early 2000s. Secondly, the construction of a nuclear power
plant was authorized by parliament in 2002. Additionally, Finland is the only coun-
try in the world where the permanent disposal of spent nuclear fuel in bedrock has
been approved at both national and local levels. Finally, the use of renewable
energy is twice as high, with its total share being among the highest in the EU
(28.5%) (Ruostetsaari, 2009).
The Finnish economy has faced frequent difficulties since the onset of the
global crisis: in 2016, GDP fell to 4.5% below the level reached in 2008, and the
debt-to-GDP ratio nearly doubled between 2008 and 2016, reaching 64% of GDP
(75.23% in 2023). Three types of reasoning have been proposed to explain this
unfavorable outcome. Firstly, Finland was hit by the 2008 crisis, which was
beyond the control of political authorities. Secondly, macroeconomic policy was
unable to mitigate the impact. Finally, the economy lacked flexibility and resilience
in the face of shocks. Although the national economy suffered from weak growth,
it has now returned to growth thanks to an improved external environment (includ-
ing the absorption of shocks) and sound policies supported by certain underlying
strengths (Vihri€
al€
a, 2017).
Germany: Germany places great importance on environmental issues and was
one of the promoters of the Kyoto Protocol (Sievers et al., 2019). It heavily
depends on fossil fuel imports, with an energy dependency rate of 62% in 2006
(Energiebilanzen, 2018). As the largest importer of fossil fuels in Europe and the
biggest CO
2
emissions producer, Germany faces high energy prices compared to
the European average. To reduce CO
2
emissions and limit dependency, a strategy
of reducing energy consumption is necessary (Fischer, 2009). The future of energy
supply is uncertain, whether through importing gas from Russia or the massive use
of fossil fuels (Deshaies, 2007). Although perceived as costly, the energy transition
294 THE JOURNAL OF ENERGY AND DEVELOPMENT
could be advantageous due to fossil fuel scarcity and reduced imports through
renewable energy sources (Linkohr, 2013). Despite the discovery of new deposits,
the end of the oil era will not be due to a lack of oil, just as the Stone Age did not
end due to a lack of stones (Autret, 2007).
On the macroeconomic front, Germany was the most affected country by the
financial crisis in Europe due to its strong dependence on trade, but its responsive-
ness allowed it to overcome the crisis (Storm and Naastepad, 2015). There is sub-
stantial evidence of the short- and long-term link between debt and growth in
Germany, based on estimates from dynamic error correction models. The results
suggest a significantly negative relationship between regional public debt and per
capita GDP in the long term (Mitze and Matz, 2015). A restraint on debt will be
necessary to invigorate a European debate on the feasibility of consolidation.
Netherlands: Although the Netherlands is well-equipped with natural resources,
it is also dependent on foreign supply. Starting in 1850, the consumption of imported
coal increased rapidly, causing concern among politicians. In 1901, state mines were
established to exploit domestic coal, and its production gradually increased. During
the interwar period, the Netherlands became almost self-sufficient in coal, leading to
a transition towards dependence on oil. In 1959, natural gas was discovered, and the
government partnered with state mines for its exploitation. The rapid introduction of
natural gas reduced the role of coal, but dependence on imported oil remained. The
oil crisis triggered a reorientation of energy policy, emphasizing diversification of
both resources and supplier countries. The Dutch government has been heavily
involved in previous energy transitions and will also play a significant role in the
transition to renewable energy (H€
olsgens, 2019).
Regarding the relationship between debt and growth, a unidirectional relation-
ship, applying the Granger causality test, is inferred from sovereign debt to eco-
nomic growth. Furthermore, variations in public debt also began to have a negative
effect on growth starting in 2009, when the debt-to-GDP ratio reached 56%
(G
omez-Puig and Sosvilla-Rivero, 2015). More recently, the debt-to-GDP ratio
remains at 46.50% in 2023.
Spain: Spain is going through a period of growth with a focus on energy supply.
Historically, in the 1970s, it experienced high inflation and low economic growth,
unlike the last two decades, which saw a 62% increase in GDP between 1990 and
2010. Its energy model is characterized by growing energy demand, CO
2
emissions,
and heavy dependence on fossil fuels (Girard et al., 2016). Previous studies have
already confirmed the impact of oil shocks on GDP, generally reflecting the relation-
ship between energy and macroeconomics, which has been declining since the
1970s-1980s, primarily due to the reduced share of oil in the economy.
The sovereign debt crisis in Europe initially affected Iceland, Ireland, and
Greece before spreading to Spain. Spain demonstrated its ability to cope, though it
faced significant budget deficits between 2008 and 2009 and since 2012. The nega-
tive relationship between debt and GDP led to a debt-to-GDP ratio increasing from
295ENERGY, DEBT, AND GROWTH
40.2% in 2008 to 107.7% currently. Gruppe and Lange (2014) observed a conver-
gence with the German market, noting a structural break starting from early 2009.
However, Granger causality tests applied over time and space between 1980 and
2013 (G
omez-Puig and Sosvilla-Rivero, 2015) suggest evidence of a feedback
loop between low growth and high debt levels in Spain following the break. There-
fore, it is essential to reduce public debt while considering the short-term negative
effects of budget adjustments on growth prospects (Jaramillo and Cottarelli, 2012).
Ireland: In the current period of heightened awareness regarding climate
change and the foreseeable scarcity of fossil fuels, Ireland has developed a new
approach for its energy future. In this context, Ireland is heavily dependent on
imports of oil and gas, which continue to increase to meet demand. Irish policy is
guided by multi-year strategic plans that set objectives and modalities. Priority is
given to ensuring supply security at reasonable prices: €8.5 billion was allocated for
the period 2007-2013 for renovating existing energy infrastructure and businesses,
and to a lesser extent, supporting the development of renewable energy sources.
Additionally, integrating renewable energy sources into the electricity grid and ensur-
ing supply stability have been the focus of research by Saintherant et al. (2008).
During the Celtic Tiger era from around 1994/1995 to the global financial crisis
of 2007/2008, Ireland experienced a significant economic and cultural transforma-
tion, reaching its peak in development (Whelan, 2014). In recent years, Ireland has
seen one of the most remarkable economic recoveries among Eurozone countries.
By 2007, Ireland was hailed as Europe’s top-performing economy with sustained
high growth, low unemployment, and budget surpluses. However, the subsequent
crash—marked by a property market collapse, surging unemployment, and a severe
banking crisis—proved extremely challenging. Ireland entered an EU and IMF
adjustment program in 2010 and is now poised to exit, though economic conditions
remain tough with high unemployment. Despite this, Ireland is often cited as a
model for other countries facing economic difficulties (Murphy, 2016), with a
debt-to-GDP ratio of 43.70% in 2023.
France: On the European level, the goal is to achieve a unified energy market
by removing regulatory barriers, which also influences French policy towards
boosting exports (Andriosopoulos and Silvestre, 2017). During the 1970s, France,
highly dependent on energy, was severely affected by two oil crises. Ensuring
energy supply security is crucial for assessing its energy structure. Primary energy
production peaked in 2015 but has since declined to 132 million tonnes of oil
equivalent (Mtep) in 2017, primarily due to reduced nuclear and hydroelectric out-
put. Fossil fuel production is minimal (none for coal and natural gas, and 1 Mtep
for crude oil). Fossil fuel imports, particularly oil and coal, have risen by 3.3% due
to increased demand. Despite these challenges, electricity exports have grown
steadily since 1980. As of 2017, France’s energy independence rate was 53% (cur-
rently 55%), decreasing by 0.5% from 2016 due to reduced nuclear power genera-
tion (Baudry et al., 2019).
296 THE JOURNAL OF ENERGY AND DEVELOPMENT
Among the top five economies in the Eurozone (Spain, Italy, Germany, France,
and the Netherlands), only French debt has increased proportionally to GDP since
the 2008 financial crisis (
Egert et al., 2011). Although limited, several studies con-
verge on highlighting that a high level of public debt can restrain long-term
growth. Specifically, public debt exceeding 90% of GDP is associated with a GDP
growth reduction of 1 to 3 percentage points on average for developed countries,
and 0.5 percentage points for France (Reinhart and Rogoff, 2010). Furthermore, a
10-percentage-point increase in the debt-to-GDP ratio decreases annual growth by
0.15 percentage points (Woo and Kumar, 2015). Lastly, there is an impact on
potential growth of at least 1 percentage point above a debt level of 80% to 120%
of GDP (Checherita and Rother, 2010). In France, public debt has quickly
approached the 90% threshold (98.1% in 2019 and 110.6% in 2023), potentially
posing a threat to long-term growth.
Italy: On the energy front, Italy’s strong dependence on imports, primarily ori-
ented towards hydrocarbons, may intensify over time as these resources become less
available. The high demand for petroleum products, coupled with low European pro-
duction, leads to an increasing need for foreign sourcing. Therefore, achieving a cor-
relation between economic growth and sustainable development will be necessary to
meet greenhouse gas reduction targets (Schifano and Moriconi, 2009).
According to the OFCE (Antonin et al., 2019), Italy faces two major challenges:
negative growth and a heavy burden of public debt relative to GDP. The evolution
of the debt-to-GDP ratio since 1960 can be divided into four phases: moderate
debt growth (1960-1980), inflated debt due to increased deficits and interest (1980-
1994), consolidated debt with reduced financial charges (1994-2008), and a disrup-
tion in stability due to recession (2008-2019). To address current issues, the only
viable solution is to improve real GDP growth, which has been negative especially
during the periods of 2008-2009 and 2012-2013. Italy needs more support from
Europe to enhance its growth rate and subsequently reduce its debt; abandoning
the euro would lead to debt catastrophes, necessitating cooperation with other
countries to ensure successful growth recovery policies.
Greece: During the rapid growth period from 1960 to 1973, Greece experi-
enced annual increases of 12.6% in final energy consumption and robust economic
growth averaging 7.7%, alongside a 12.3% rise in total energy consumption. How-
ever, by 1980, annual growth in final energy dropped to 4.3%, coinciding with
Greece’s economic growth slowing to an average of 1.6% annually (Hondroyiannis
et al., 2002). Despite the 1970s oil shocks, Greece improved its energy intensity.
Dependent heavily on energy imports, particularly oil, which constituted 79% of
expenditures in 1996, Greece aimed to reduce economic disparities in the early
1990s, enhancing competitiveness by reducing inflation from 15.9% in 1992 to
3.7% in 1998 (Ministry of Environment, Energy and Climate). However, by the
end of 2008, CO
2
emissions had risen 26% above 1990 levels, reflecting energy
efficiency challenges (Hatzigeorgiou et al., 2011). Greece’senergyissuesstem
297ENERGY, DEBT, AND GROWTH
from poor sectoral performance and limited alternative sources, necessitating a
strategy emphasizing the development of substitute energy sources to reduce reli-
ance on imported oil. Key objectives of Greek energy policy include advancing
alternative energy sources and international collaboration on energy management
projects (Donatos and Mergos, 1989).
Extensive public spending in Greece has significantly increased government
debt, a trend highlighted in studies by Schclarek et al. (2004), Adam and Bevan
(2005), and others, which reveal a long-term negative correlation between public
debt and economic growth. Financial crises have exacerbated this relationship,
leading to reduced economic growth rates and a surge in debt levels. Furthermore,
research, including findings from Pattillo et al. (2002), underscores a persistent
negative association between external debt and economic growth. Spilioti and
Vamvoukas (2015) conducted further analysis on Greece, incorporating variables
such as fiscal policy indicators, economic openness, external competitiveness,
demographic characteristics, and indicators of the country’s capacity for invest-
ment and short-term financing. Their findings highlight key determinants of GDP
growth rate, including debt levels, GDP per capita, gross national savings, total
imports and exports, trade dynamics, unemployment, and population growth rate
(Smyth and Hsing, 1995; Nguyen et al., 2003; Reinhart and Rogoff, 2010).
After presenting the theoretical macro-energy state for each of the nine OECD
European countries, we will analyze the possible links between three variables:
energy supply, public debt, and economic growth. To ensure comparability with
existing literature, we will conduct a country-specific analysis, grouping them into
several categories. While inter-country analysis is certainly of interest, it will be
the subject of future research, as it requires causality tests involving numerous
combinations to determine whether the dynamics of one variable in a country
cause those in another country.
3. Data and Methodology
Modeling VAR and Causality Tests: To highlight the relationships between
the variables, we will construct VAR models to study Granger causality between
stationary variables (or variables made stationary in the case of non-stationary or
differenced processes). VAR(P) modeling is based on a system of lagged equations
of order P, where each variable is both explanatory and explained. The minimiza-
tion of the Akaike criterion also determines the optimal lag order P. Assuming two
stationary time series xtand yt, the VAR(P) model is therefore written as follows:
xt5
a
11X
P
j51
b
1,jxtj1X
P
j51
g
1,jytj1e1,t(1)
298 THE JOURNAL OF ENERGY AND DEVELOPMENT
yt5
a
21X
P
j51
b
2,jxtj1X
P
j51
g
2,jytj1e2,t(2)
(
a
i,
b
i,j,
g
i,j) represent the parameters of the VAR(P) model, and ei,trepresent the
innovations following an independent and identically distributed process (0,
s
2
e)
with i 51, 2 and j 51….P.
From estimating this model, it is possible to analyze the contemporaneous
impact on xtand ytof a shock at time ton the innovations ei,tand its propagation,
meaning its effects on the variables in period h,xt1hand yt1hwith h51…H.
This allows us to obtain the impulse response functions of xtand ytfollowing a
shock on ei,tfor all i 51, 2.
Since we have the series response for period h, it is possible to calculate the
forecast error variance for each variable and attribute the proportions due to shocks
on ei,t. This allows us to obtain variance decomposition tables (Hamilton, 1994).
Granger (1969-1988) defines causality as the ability of ytto predict xt, meaning
if the historical information of ytis included in xt. Therefore, a VAR model allows
testing the hypothesis that ytdoes not cause xtby testing restrictions on its para-
meters, specifically
g
1;15
g
1;25
g
1,P50 (using tests such as Fisher, Wald, or
Likelihood-Ratio tests through a likelihood ratio). The reverse relationship —
whether xtdoes not cause yt—is examined by testing if
b
1;15
b
1;25
b
1,P50.
When variables do not share the same characteristics (either all stationary or
cointegrated of the same order), the traditional Granger causality approach cannot
be applied. Toda and Yamamoto (1995) proposed a method that overcomes this
limitation by allowing for testing causality (in the Granger sense) between vari-
ables of different natures. Their methodology builds upon a standard VAR(P)
model. Once the order P is determined, they suggest estimating an augmented
VAR model (VAR(P1m)), where m is the maximum order of integration identi-
fied for any variable (in our case, m is equal to 1). The causality test is then con-
ducted using a Wald test on the parameters of the augmented VAR model.
Assuming two stationary time series xtand yt, the VAR(P11) model is written
as follows:
xt5
a
11X
P11
j51
b
1,jxtj1X
P11
j51
g
1,jytj1e1,t(3)
yt5
a
21X
P11
j51
b
2,jxtj1X
P11
j51
g
2,jytj1e2,t(4)
(
a
i,
b
i,j,
g
i,j) represent the parameters of the VAR(P11) model, and ei,trepresent
the innovations following an independent and identically distributed process pro-
cess (0,
s
2
e)withi51, 2 and j 51….P11.
299ENERGY, DEBT, AND GROWTH
Stationarity Tests of the Selected Variables: Stationarity tests of variables
are crucial to ensure that the time series used in VAR analysis are suitable.
Several commonly used tests include the Augmented Dickey-Fuller (ADF) test.
These tests assess whether a time series has unit roots, which would indicate
non-stationarity. Specifically, the ADF test is often preferred because it can detect
the presence of unit roots even in the presence of serial correlation. To conduct
these tests, the null hypothesis is specified that the series has a unit root (is non-sta-
tionary). If the test results reject this null hypothesis, it suggests that the series is
stationary. Once all series are stationary or made stationary through differencing,
they can be appropriately used in a VAR model to analyze relationships between
them.
The data were extracted from OECD Economic Outlook No. 115-May 2024
(primary energy supply, government gross financial commitments, and gross
domestic product in volume terms at market prices). To extend the series, we also
used OECD Economic Outlook No. 73-June 2003 for all selected countries, in
order to obtain the broadest possible observations (data calibration between two
sources of data).
We present three variables. Firstly, primary energy supply represents the sum
of energy production and imports, adjusted for exports, international bunkers, and
stock changes. The calculation method by the International Energy Agency (IEA)
is based on the calorific value of energy products and a common unit of account,
the tone of oil equivalent (toe), equal to 107 kilocalories (41.868 gigajoules) (Pri-
mary energy supply). Secondly, public debt, which is the gross debt of general
government, is a key indicator of public finance sustainability. Debt includes cash
and deposits, securities other than shares, loans, insurance, pensions, and standard-
ized guarantee schemes and other accounts payable. Changes in public debt over
time primarily reflect the impact of past government deficits. This variable is
expressed in euros (General government gross financial liabilities). Finally, eco-
nomic growth, indicated by annual variations in gross domestic product (GDP), is
the standard measure of value added from the production of goods and services in
a country over a specific period. Real GDP, expressed in euros, refers to GDP in
constant prices and reflects the volume level of GDP (Gross domestic product, vol-
ume, market prices). We have a sample of nine European countries (Table 1). The
observation period varies for different countries.
Before analyzing the potential relationships between these three variables, we
will examine whether they are stationary or not. Preliminary investigation into the
stationarity of variables is essential before any estimation. For this purpose, we can
use the Dickey-Fuller Simple or Augmented test (Dickey and Fuller, 1979-1981).
The Dickey-Fuller test is preferred in the case of homoscedastic variances. The
Dickey-Fuller test is based on several models that help identify the nature of
300 THE JOURNAL OF ENERGY AND DEVELOPMENT
non-stationarity depending on the structure of the process generating the time
series xt, whether it has a stochastic or deterministic trend (TS or DS with/without
a constant).
We follow the testing procedure developed by Ertur (1991-1998) and rely on
minimizing the Akaike Information Criterion (AIC) to select the optimal lag P.
This procedure for a simple DF test starts with estimating the model with trend
and constant:
xt5Ø1:xt11et1bt 1c (5)
Dxt5Ø11
ðÞ
:xt11et1bt 1c5
r
:xt11
e
t1bt 1c (6)
With tas time, Ø1as the coefficient associated with lag 1,
r
5(Ø11) and
e
trepresenting innovations following an i.i.d. process (0,
s
2
e).
The null hypothesis of non-stationarity for the process xtimplies that Ø151, or
equivalently
r
50. Under this hypothesis, the process xtis autocorrelated at lag 1,
known as a random walk (DS process), meaning the differenced series Dxtis a ran-
dom process. Conversely, under the alternative hypothesis, jØ1j,1, the process xt
is a stationary AR(1) process.
The Augmented Dickey-Fuller test evaluates the nullity of Ø1while considering
the autocorrelation of the series. A lag polynomial
u
(B) applied to the differenced
series Dxtadjusts the original test models to account for autocorrelation.
Thus, the complete model (including a trend and a constant) is written as fol-
lows:
u
B
ðÞ
1Ø1B
ðÞ
xt5c11b:t1et(7)
Table 1
REPRESENTATION OF THE NUMBER OF OBSERVATIONS PER
STUDIED EUROPEAN COUNTRY
Country Period Number of observations
Belgium 1970-2023 54
Finland 1970-2023 54
Germany 1970-2023 54
Netherlands 1970-2023 54
Spain 1970-2023 54
Ireland 1970-2023 54
France 1970-2023 54
Italy 1970-2023 54
Greece 1970-2023 54
301ENERGY, DEBT, AND GROWTH
With tas time, Ø1the coefficient associated with the first lag; Bas the lag operator;
and etthe innovations following an i.i.d. process (0,
s
2
e).
u
B
ðÞ
is the lag operator
polynomial of degree (p21).
Therefore, the null hypothesis is still Ø151or
r
5(Ø121). (12
u
1
2
u
2
–
u
p-1
)
50 and the alternative hypothesis of stationarity remains jØ1j,1.
The nullity of
r
represents the null hypothesis of the test, indicating the pres-
ence of a unit root. However, it is appropriate to test the joint nullity of
r
along
with the parameters cand b, which represent the constant and the slope of the
trend, respectively. Depending on the result obtained, it is necessary to estimate a
model without the trend and to reiterate the joint nullity tests of the parameters.
Table 2 provides an overview of the stationarity test results.
4. Results and Discussion
Henceforth, we refer to the following acronyms to characterize our variables:
energy supply, public debt, and economic growth.
PDE corresponds to the debt variable. This is stationary either by nature or
through a “stationarization”operation of TS processes.
ESU corresponds to the primary energy supply variable. This is stationary
either by nature or through a “stationarization”operation of TS processes.
EGR corresponds to the economic growth variable. This is stationary either
by nature or through a “stationarization”operation of TS processes.
We will present the empirical results defined by the results of Granger causality
tests, variance decomposition, and impulse response functions. For the analysis of
these, we interpret the results following a positive shock.
Table 2
STATIONARITY TEST RESULTS
Country Energy supply (ESU) Public debt (PDE) Economic growth (EGR)
Belgium Stationary Non-Stationary (DS) Stationary
Finland Stationary Stationary Stationary
Germany Non-Stationary (TS) Non-Stationary (TS) Non-Stationary (TS)
Netherlands Non-Stationary (TS) Non-Stationary (TS) Non-Stationary (TS)
Spain Stationary Stationary Non-Stationary (DS)
Ireland Stationary Stationary Stationary
France Non-Stationary (TS) Non-Stationary (TS) Non-Stationary (TS)
Italy Non-Stationary (TS) Non-Stationary (TS) Non-Stationary (TS)
Greece Non-Stationary (TS) Non-Stationary (TS) Non-Stationary (TS)
302 THE JOURNAL OF ENERGY AND DEVELOPMENT
Belgium: We note from the Granger test a unidirectional causality relationship
at the 5% threshold from EGR to PDE. Variance decomposition shows that 88%
of the forecast error variance in EGR is due to its own shocks, and 12% to those of
PDE. Regarding debt, variance decomposition shows that 68% of the forecast error
variance in its growth is due to its own shocks, and 32% to those of EGR. How-
ever, the Granger test does not indicate causality in this direction at the 5% thresh-
old, but at the 10% threshold, the response effect of debt is significant, implying
bidirectional causality. Impulse response functions show that PDE reacts immedi-
ately and negatively to a shock on EGR, but the effect becomes positive after
period 2 before tapering off. In the case of a shock on PDE, EGR does not react
immediately but shows a maximum negative effect after 2 years. At the 10%
threshold, we note a bidirectional causality relationship between EGR and ESU.
The response of the latter to a shock on EGR is immediate and positive, but the
effect becomes negative in period 2 before canceling out. Conversely, EGR does
not react immediately to a shock on ESU, but a positive effect is noted after 2 years
(see Appendix 1).
Finland: We do not observe a causality relationship between ESU and the other
variables. The results from the Granger causality tests indicate EGR causes PDE.
The variance decomposition table confirms this result, showing that 64% of the
forecast error variance in PDE is due to shocks in EGR, with only 36% attributable
to its own shocks (see Appendix 2). Therefore, a shock on EGR leads to a negative
response in PDE, with a maximum effect seen at 2 years before tapering off after
periods of 4-5 years.
Germany: The results from the bivariate Granger causality tests indicate that
there is no causality relationship between PDE and ESU. Only EGR, as per
Granger causality, influences PDE. This finding suggests a unidirectional causal
relationship between these two variables, primarily driven by EGR. The variance
decomposition analysis shows that 73% of the forecast error variance in PDE is
due to its own shocks, while 27% is attributable to EGR (see Appendix 3). Exam-
ining the impulse response functions of the stationary VAR, we observe an imme-
diate negative reaction of PDE following a shock to EGR. The maximum effect is
seen at period 2, which is 2 years after the shock in PDE, and diminishes after peri-
ods of 4-5 years.
Netherlands: We observe a unilateral causality relationship from PDE to EGR
at the 5% significance level. According to the variance decomposition table, 63%
of the forecast error variance in EGR is due to its own shocks, while 37% is attrib-
uted to shocks in PDE, which has an instantaneous negative impulse response (see
Appendix 4). The maximum effect of the shock is reached at period 2 and gradu-
ally diminishes until periods 5-6. Furthermore, we notice another causality relation-
ship from EGR to ESU at the 10% significance level. The response of ESU is
immediate and positive, lasting until periods 5-6.
303ENERGY, DEBT, AND GROWTH
Spain: The results of the Granger causality tests indicate three causal relation-
ships at the 5% significance level. First, PDE causes EGR. Additionally, there is a
bidirectional relationship between ESU and EGR. According to the variance
decomposition tables, 50% of the forecast error variance in EGR is due to its own
shocks, and 50% is attributed to shocks in ESU (see Appendix 5). The response is
not instantaneous but starts positively at period 2, peaks at period 3, and then
diminishes. Regarding the feedback effect (from EGR to ESU), we observe that
86% of the forecast error variance in ESU is due to its own shocks, while 14% is
due to shocks in EGR. A positive response in ESU begins in period 1 (not instanta-
neous), continues until period 2, and then turns negative from period 3 onward.
Similarly, the variance decomposition shows that 58% of the forecast error vari-
ance in EGR is due to its own shocks, and 42% is due to shocks in PDE. Analyz-
ing the impulse response functions, a shock in PDE to EGR does not result in an
immediate reaction but rather a response starting from period 2 with a positive and
sustained effect.
Ireland: Based on the Granger causality test results, it is evident that ESU is
influenced by PDE. Despite the p-value being above 5%, EGR also causes ESU.
The variance decompositions (see Appendix 6) reveal that 76% of the forecast
error variance in ESU is due to its own shocks, and 24% is due to shocks in PDE.
Additionally, the variance of the forecast error in ESU is attributed to 79% of its
own shocks and 21% to EGR. Examining the impulse response functions, a shock
in PDE leads to a negative response in ESU) over 2-4 years before slowly dissipat-
ing. This same shock also triggers an instantaneous negative response in EGR over
a period of 3 years. Finally, a shock in EGR positively and instantaneously affects
ESU up to period 4. Hence, it can be inferred that the effect of EGR on ESU
results from the significant impact of PDE on EGR (cross-effect). The faster PDE
accelerates, the weaker EGR and ESU becomes.
France: Based on the tests, at the 5% significance level, PDE causes EGR. The
variance decomposition table indicates that 70% of the forecast error variance
in EGR is due to its own shocks, and 30% is due to shocks in PDE (see
Appendix 7). Impulse response functions reveal that EGR responds immediately
but negatively to a shock in PDE, with effects tapering off over a period of
4-5 years. The bidirectional nature of the relationship, meaning causality from
EGR to PDE, only appears at the 10% significance level. However, the response is
not instantaneous, as we observe a negative and maximal reaction 2 years later.
Also at the 10% significance level, we note a unilateral causal relationship from
ESU to PDE.
Italy: For Italy, Granger causality tests reveal a unidirectional relationship from
the increase in PDE to ESU. According to the variance decomposition table, 72%
of the forecast error variance of ESU is attributable to its own shocks, while
28% is due to shocks from PDE (See Appendix 8). Impulse response functions
304 THE JOURNAL OF ENERGY AND DEVELOPMENT
indicate that ESU does not react immediately, but its response peaks negatively
after two years.
Greece: The Granger causality tests indicate that EGR affects both PDE and
ESU. According to the variance decomposition table, 60% of the forecast error
variance of ESU is due to its own shocks, while 40% is attributable to shocks from
EGR (See Appendix 9). There is an immediate positive response of ESU to this
shock, reaching its peak effect at 2 years before attenuating around period 5. The
variance decomposition table indicates that 90% of the forecast error variance of
PDE is due to its own shocks, and 10% to EGR, with a diffusion that attenuates
over 5 years. There is an immediate positive response at period 1, but it turns nega-
tive by period 2.
Our country-specificfindings highlight a new dimension for categorizing coun-
tries, considering both debt-growth classification and energy supply classification.
Regarding the relationship between debt and growth, countries can be divided
into two groups (See Appendix 10). Firstly, group 1 –lowly indebted: PDE does
not exceed 100% of EGR for northern countries: in 2023, Germany was at
63.60%, Ireland at 43.70%, Finland at 75.80%, and the Netherlands at 46.50%.
Among northern European countries, only Belgium has a debt exceeding 100% at
105.2%. This group can be divided into two parts. On one side, Germany and Fin-
land are characterized by a unidirectional causality relationship from EGR to PDE
(a positive shock in EGR leads to a negative response in PDE). On the other side,
Ireland and the Netherlands are characterized by a unidirectional causality relation-
ship from PDE to EGR.
Secondly, group 2 –highly indebted: In 2023, PDE exceeded 100% of EGR for
Southern European countries as follows: Spain was at 107.70%, France at 110.60%,
Greece at 161.90%, and Italy at 137.30%. This group can be divided into 2 sub-
groups. On one hand, France and Spain mainly exhibit a 5% causal relationship
from debt to economic growth. On the other hand, Belgium and Greece are charac-
terized by a causal relationship from economic growth to debt (debt reacts nega-
tively to a shock in growth). Italy alone among Southern countries shows no causal
relationship between these two variables (debt and economic growth).
In terms of energy supply, countries can be grouped accordingly (a different
distribution from the literature). One group (A) consists of countries with an energy
dependency rate exceeding 50% and at least one causal relationship (at 5%), either
to or from energy supply: Spain, Italy, Greece, and Ireland. These countries are
part of the GIIPS, the most affected countries after the global financial crisis. The
other group (B) consists of energetically independent countries with no significant
causal relationship (at 5%): Germany, Finland, the Netherlands, and France.
Belgium is an exception as it shows no significant causal relationship but has a
bidirectional relationship at 10% and is subsequently dependent on energy imports.
Therefore, a country with an energy dependency rate below 50% shows no
305ENERGY, DEBT, AND GROWTH
causality to or from ESU, with equal proportions of production and importation.
Conversely, a country with a high dependency rate shows at least one significant
causal relationship between ESU and one of the other variables (PDE or EGR),
with imports comprising a very high percentage compared to production.
Overall, countries in the North have a PDE to EGR ratio below 100% and a
low energy dependence rate, and vice versa. Resolving the debt issues in troubled
countries is necessary to encourage them to produce and import more energy to
achieve growth objectives. These findings are derived from Granger causality tests
(See Appendix 10).
5. Conclusion
This study examines the relationships between energy supply, public debt, and
economic growth in nine OECD European countries. Using VAR models, as well
as Granger and Toda-Yamamoto causality tests, we find that economic dynamics
differ significantly between northern and southern European nations.
In northern countries like Germany and Finland, economic growth influences
debt levels, indicating growth’s role in debt management. Conversely, in Ireland
and the Netherlands, debt levels impact economic growth. In southern Europe,
high debt negatively affects growth in countries like Spain and France, while
Greece and Belgium show the reverse causality.
Energy dependency is critical, with countries like Spain, Italy, Greece, and Ire-
land showing significant causal links between energy supply and economic vari-
ables due to high import reliance. In contrast, energy-independent nations like
Germany, Finland, and the Netherlands exhibit more stable interactions.
These findings highlight the importance of tailored policies. For highly indebted
and energy-dependent countries, enhancing energy security and managing debt
sustainably are crucial for economic stability. For less indebted and energy-
independent countries, balancing energy production and imports while leveraging
growth for debt management remains essential. This study provides insights for
policymakers aiming to promote sustainable economic development and energy
security across OECD European nations.
306 THE JOURNAL OF ENERGY AND DEVELOPMENT
APPENDIX 1: BELGIUM RESULTS
APPENDIX FIGURE 1: BELGIUM —IMPULSE RESPONSE FUNCTIONS
APPENDIX TABLE 1: BELGIUM —GRANGER CAUSALITY
(ACCORDING TO THE TODA-YAMAMOTO APPROACH)
Null hypothesis Statistic Prob.
PDE does not Granger Cause ESU 0.615 0.437
ESU does not Granger Cause PDE 0.186 0.667
EGR does not Granger Cause ESU 3.115 0.084
ESU does not Granger Cause EGR 2.942 0.093
EGR does not Granger Cause PDE 7.7 0.022
PDE does not Granger Cause EGR 4.9 0.087
APPENDIX TABLE 2: BELGIUM —VARIANCE DECOMPOSITION
Variance decomposition of EGR Period EGR PDE
Cholesky Ordering : PDE EGR 10 87.735 12.264
Variance decomposition of PDE Period EGR PDE
Cholesky Ordering : EGR PDE 10 32.023 67.977
307ENERGY, DEBT, AND GROWTH
APPENDIX 2: FINLAND RESULTS
APPENDIX TABLE 3: FINLAND —GRANGER CAUSALITY
Null hypothesis Statistic Prob.
PDE does not Granger Cause ESU 0.209 0.651
ESU does not Granger Cause PDE 0.141 0.709
EGR does not Granger Cause ESU 0.215 0.645
ESU does not Granger Cause EGR 1.853 0.181
EGR does not Granger Cause PDE 25.325 1.E-05
PDE does not Granger Cause EGR 1.367 0.249
APPENDIX TABLE 4: FINLAND —VARIANCE DECOMPOSITION
Variance decomposition of PDE Period EGR PDE
Cholesky Ordering : EGR PDE 10 64.081 35.919
APPENDIX FIGURE 2: FINLAND —IMPULSE RESPONSE FUNCTIONS
308 THE JOURNAL OF ENERGY AND DEVELOPMENT
APPENDIX 3: GERMANY RESULTS
APPENDIX TABLE 5: GERMANY —GRANGER CAUSALITY
Null hypothesis Statistic Prob.
PDE does not Granger Cause ESU 1.833 0.181
ESU does not Granger Cause PDE 2.077 0.155
EGR does not Granger Cause ESU 1.020 0.317
ESU does not Granger Cause EGR 0.157 0.693
EGR does not Granger Cause PDE 5.451 0.023
PDE does not Granger Cause EGR 0.035 0.852
APPENDIX FIGURE 3: GERMANY — IMPULSE RESPONSE FUNCTIONS
Response of PDE to EGR
APPENDIX TABLE 6: GERMANY —VARIANCE DECOMPOSITION
Variance decomposition of PDE Period EGR PDE
Cholesky Ordering : EGR PDE 10 27.703 72.297
309ENERGY, DEBT, AND GROWTH
APPENDIX 4: THE NETHERLANDS RESULTS
APPENDIX FIGURE 4: THE NETHERLANDS —IMPULSE RESPONSE FUNCTIONS
APPENDIX TABLE 7: THE NETHERLANDS —GRANGER CAUSALITY
Null hypothesis Statistic Prob.
PDE does not Granger Cause ESU 0.601 0.552
ESU does not Granger Cause PDE 1.227 0.301
EGR does not Granger Cause ESU 2.537 0.089
ESU does not Granger Cause EGR 1.101 0.340
EGR does not Granger Cause PDE 1.174 0.317
PDE does not Granger Cause EGR 9.178 0.0004
APPENDIX TABLE 8: THE NETHERLANDS —VARIANCE DECOMPOSITION
Variance decomposition of EGR Period EGR PDE
Cholesky Ordering : PDE EGR 10 36.981 63.019
310 THE JOURNAL OF ENERGY AND DEVELOPMENT
APPENDIX 5: SPAIN RESULTS
APPENDIX FIGURE 5: SPAIN —IMPULSE RESPONSE FUNCTIONS
APPENDIX TABLE 9: SPAIN —GRANGER CAUSALITY
(ACCORDING TO THE TODA-YAMAMOTO APPROACH)
Null hypothesis Statistic Prob.
PDE does not Granger Cause ESU 0.002 0.965
ESU does not Granger Cause PDE 1.574 0.218
EGR does not Granger Cause ESU 6.8 0.03
ESU does not Granger Cause EGR 5.7 0.04
EGR does not Granger Cause PDE 1.8 0.41
PDE does not Granger Cause EGR 5.4 0.04
APPENDIX TABLE 10: SPAIN —VARIANCE DECOMPOSITION
Variance decomposition of ESU Period ESU EGR
Cholesky Ordering : EGR ESU 10 85.757 14.243
Variance decomposition of EGR Period ESU EGR
Cholesky Ordering : ESU EGR 10 51.307 48.693
Variance decomposition of EGR Period PDE EGR
Cholesky Ordering : PDE EGR 10 41.542 58.458
311ENERGY, DEBT, AND GROWTH
APPENDIX 6: IRELAND RESULTS
APPENDIX FIGURE 6: IRELAND —IMPULSE RESPONSE FUNCTIONS
APPENDIX TABLE 11: IRELAND —GRANGER CAUSALITY
Null hypothesis Statistic Prob.
PDE does not Granger Cause ESU 3.614 0.023
ESU does not Granger Cause PDE 0.715 0.549
EGR does not Granger Cause ESU 2.849 0.052
ESU does not Granger Cause EGR 0.010 0.998
EGR does not Granger Cause PDE 1.624 0.203
PDE does not Granger Cause EGR 3.067 0.041
APPENDIX TABLE 12: IRELAND —VARIANCE DECOMPOSITION
Variance decomposition of ESU Period ESU EGR
Cholesky Ordering : PDE ESU 10 23.616 76.384
Variance decomposition of ESU Period ESU EGR
Cholesky Ordering : EGR ESU 10 21.480 78.520
Variance decomposition of EGR Period PDE EGR
Cholesky Ordering : PDE EGR 10 41.311 58.689
312 THE JOURNAL OF ENERGY AND DEVELOPMENT
APPENDIX 7: FRANCE RESULTS
APPENDIX FIGURE 7: FRANCE —IMPULSE RESPONSE FUNCTIONS
APPENDIX TABLE 13: FRANCE —GRANGER CAUSALITY
Null hypothesis Statistic Prob.
PDE does not Granger Cause ESU 0.079 0.780
ESU does not Granger Cause PDE 3.560 0.065
EGR does not Granger Cause ESU 0.146 0.704
ESU does not Granger Cause EGR 0.897 0.348
EGR does not Granger Cause PDE 3.423 0.071
PDE does not Granger Cause EGR 4.175 0.047
APPENDIX TABLE 14: FRANCE —VARIANCE DECOMPOSITION
Variance decomposition of EGR Period PDE EGR
Cholesky Ordering : PDE EGR 10 29.835 70.165
313ENERGY, DEBT, AND GROWTH
APPENDIX 8: ITALY RESULTS
APPENDIX FIGURE 8: ITALY —IMPULSE RESPONSE FUNCTIONS
APPENDIX TABLE 15: ITALY —GRANGER CAUSALITY
Null hypothesis Statistic Prob.
PDE does not Granger Cause ESU 6.457 0.014
ESU does not Granger Cause PDE 0.834 0.365
EGR does not Granger Cause ESU 0.695 0.407
ESU does not Granger Cause EGR 0.733 0.395
EGR does not Granger Cause PDE 0.643 0.426
PDE does not Granger Cause EGR 0.615 0.436
APPENDIX TABLE 16: ITALY —VARIANCE DECOMPOSITION
Variance decomposition of ESU Period PDE ESU
Cholesky Ordering : PDE ESU 10 28.337 71.662
314 THE JOURNAL OF ENERGY AND DEVELOPMENT
APPENDIX 9: GREECE RESULTS
APPENDIX FIGURE 9: GREECE —IMPULSE RESPONSE FUNCTIONS
APPENDIX TABLE 17: GREECE —GRANGER CAUSALITY
Null hypothesis Statistic Prob.
PDE does not Granger Cause ESU 0.038 0.844
ESU does not Granger Cause PDE 0.291 0.591
EGR does not Granger Cause ESU 5.712 0.025
ESU does not Granger Cause EGR 1.420 0.238
EGR does not Granger Cause PDE 14.884 0.0003
PDE does not Granger Cause EGR 0.0003 0.985
APPENDIX TABLE 18: GREECE —VARIANCE DECOMPOSITION
Variance decomposition of ESU Period EGR ESU
Cholesky Ordering : EGR ESU 10 40.150 59.850
Variance decomposition of PDE Period EGR PDE
Cholesky Ordering : EGR PDE 10 9.944 90.056
315ENERGY, DEBT, AND GROWTH
APPENDIX 10: GENERAL RESULTS
APPENDIX FIGURE 10: GROUPING OF COUNTRIES (DEBT AND GROWTH)
APPENDIX FIGURE 11: GROUPING OF COUNTRIES (ENERGY SUPPLY)
APPENDIX TABLE 19: GRANGER CAUSALITY RESULTS
a
Country
PDE➔
ESU
ESU➔
PDE
EGR➔
ESU
ESU➔
EGR
EGR➔
PDE
PDE➔
EGR
Belgium * * ** *
Finland ***
Germany **
Netherlands * **
Spain ** ** **
Ireland ** ** **
France * * **
Italy **
Greece ** ***
a
*p,0.1; **p,0.05; ***p,0.01.
316 THE JOURNAL OF ENERGY AND DEVELOPMENT
REFERENCES
Abaidoo, R. “Economic Growth and Energy Consumption in an Emerging Economy: Aug-
mented Granger Causality Approach.”Research in Business and Economics Journal,vol.4,2011,
pp. 1.
Adam, C. S., and D. L. Bevan. “Fiscal Deficits and Growth in Developing Countries.”Journal
of Public Economics, vol. 89, no. 4, 2005, pp. 571–597. DOI: 10.1016/j.jpubeco.2004.02.006.
Adebola, S. S. “Electricity Consumption and Economic Growth: Trivariate Investigation in
Botswana with Capital Formation.”International Journal of Energy Economics and Policy,vol.1,
no. 2, 2011, pp. 32–46.
Aizenman, J., K. Kletzer, and B. Pinto. Economic Growth with Constraints on Tax Revenues
and Public Debt: Implications for Fiscal Policy and Cross-Country Differences. National Bureau of
Economic Research, 2007. DOI: 10.3386/w12750.
Akinlo, A. E. “Energy Consumption and Economic Growth: Evidence from 11 Sub-Sahara Afri-
can Countries.”Energy Economics,vol.30,no.5,2008,pp.2391–2400. DOI: 10.1016/j.eneco.
2008.01.008.
Altinay, G., and E. Karagol. “Electricity Consumption and Economic Growth: Evidence from
Turkey.”Energy Economics, vol. 27, no. 6, 2005, pp. 849–856. DOI: 10.1016/J.ENECO.2005.07.
002.
Andriosopoulos, K., and S. Silvestre. “French Energy Policy: A Gradual Transition.”Energy
Policy, vol. 106, 2017, pp. 376–381. DOI: 10.1016/j.enpol.2017.04.015.
Antonin, C., M. Guerini, M. Napoletano, and F. Vona. Italie: Sortir du Double Pi
ege de l’Endet-
tement
Elev
e et de la Faible Croissance. Sciences Po, 2019.
Apergis, N., and J. E. Payne. “Renewable Energy Consumption and Economic Growth: Evidence
from a Panel of OECD Countries.”Energy Policy, vol. 38, no. 1, 2010, pp. 656–660. DOI: 10.1016/
j.enpol.2009.09.002.
Asafu-Adjaye, J. “The Relationship between Energy Consumption, Energy Prices and Economic
Growth: Time Series Evidence from Asian Developing Countries.”Energy Economics, vol. 22, no.
6, 2000, pp. 615–625. DOI: 10.1016/S0140-9883(00)00050-5.
Baudry, M., S. Beck, K. Chaput, Y. Coltier, F.-X. Dussud, A. Foussard, et al. Bilan
Energ
etique
de la France pour 2017,2019.
Behname, M., et al. “La Consommation d’
Energie Renouvelable et la Croissance
Economique
dans l’Europe de l’Ouest.”Romanian Journal of Economics, vol. 2, 2012, pp. 160–171.
Belke, A., F. Dobnik, and C. Dreger. “Energy Consumption and Economic Growth: New
Insights into the Cointegration Relationship.”Energy Economics, vol. 33, no. 5, 2011, pp. 782–789.
DOI: 10.1016/j.eneco.2011.02.005.
Bell, A., R. Johnston, and K. Jones. “Stylised Fact or Situated Messiness? The Diverse Effects
of Increasing Debt on National Economic Growth.”Journal of Economic Geography, vol. 15, no. 2,
2015, pp. 449–472. DOI: 10.1093/jeg/lbu005.
Benassy-Quere, A. “La Zone Euro en Qu^
ete de Coordination.”Revue d’
Economie Financi
ere,
vol. 1, 2017, pp. 209–224.
Carminel, T. “L’Impossible D
ecouplage entre
Energie et Croissance.”
Economie de l’Apr
es
Croissance: Politiques de l’Anthropoc
ene II, Presses de Sciences Po, 2015, pp. 97–115.
Checherita-Westphal, C. D., and P. Rother. “The Impact of High and Growing Government Debt
on Economic Growth: An Empirical Investigation for the Euro Area.”European Economic Review,
vol. 56, no. 7, 2012, pp. 1392–1405. DOI: 10.1016/j.euroecorev.2012.06.007.
Chen, S.-T., H.-I. Kuo, and C.-C. Chen. “The Relationship Between GDP and Electricity Con-
sumption in 10 Asian Countries.”Energy Policy, vol. 35, no. 4, 2007, pp. 2611–2621. DOI: 10.
1016/j.enpol.2006.10.001.
317ENERGY, DEBT, AND GROWTH
Cheng, B. S. “An Investigation of Cointegration and Causality Between Energy Consumption
and Economic Growth.”The Journal of Energy and Development, vol. 21, no. 1, 1995, pp. 73–84.
Chontanawat, J., L. C. Hunt, and R. Pierse. “Does Energy Consumption Cause Economic
Growth? Evidence from a Systematic Study of Over 100 Countries.”Journal of Policy Modeling,
vol. 30, no. 2, 2008, pp. 209–220. DOI: 10.1016/j.jpolmod.2006.10.003
Cochrane, J. H. “Understanding Policy in the Great Recession: Some Unpleasant Fiscal
Arithmetic.”European Economic Review, vol. 55, no. 1, 2011, pp. 2–30. DOI: 10.1016/j.
euroecorev.2010.11.002.
Codogno, L., C. Favero, and A. Missale. “Yield Spreads on EMU Government Bonds.”Eco-
nomic Policy, vol. 18, no. 37, 2003, pp. 503–532. DOI: 10.1111/1468-0327.00114_1.
Denison, E. Trends in American Economic Growth. Brookings Institution, 1985.
Deshaies, M. “
Energie et Environnement en Allemagne.”Revue G
eographique de l’Est, vol. 47,
no. 1, 2007.
Destek, M. A. “Renewable Energy Consumption and Economic Growth in Newly Industrialized
Countries: Evidence from Asymmetric Causality Test.”Renewable Energy, vol. 95, 2016, pp. 478–
484. DOI: 10.1016/j.renene.2016.04.049.
Diamond, P. A. “National Debt in a Neoclassical Growth Model.”The American Economic
Review,vol.55,no.5,1965,pp.1126–1150.
Dickey, D. A., and W. A. Fuller. “Distribution of the Estimators for Autoregressive Time Series
with a Unit Root.”Journal of the American Statistical Association, vol. 74, no. 366a, 1979, pp.
427–431. DOI: 10.1080/01621459.1979.10482531.
Dickey, D. A., and W. A. Fuller. “Likelihood Ratio Statistics for Autoregressive Time Series
with a Unit Root.”Econometrica: Journal of the Econometric Society, 1981, pp. 1057–1072. DOI:
10.2307/1912517.
Donatos, G. S., and G. J. Mergos. “Energy Demand in Greece: The Impact of the Two Energy
Crises.”Energy Economics, vol. 11, no. 2, 1989, pp. 147–152. DOI: 10.1016/0140-9883(89)90008-X.
Dos Santos Gaspar, J., A. C. Marques, and J. A. Fuinhas. “The Traditional Energy-Growth
Nexus: A Comparison Between Sustainable Development and Economic Growth Approaches.”Eco-
logical Indicators, vol. 75, 2017, pp. 286–296. DOI: 10.1016/J.ECOLIND.2016.12.048.
Egert, B., et al. R
eduire la Dette Publique Franc¸aise: Les Options de l’Assainissement
Budg
etaire. OECD Publishing, 2011.
Eichengreen, B., and D. Leblang. “Capital Account Liberalization and Growth: Was Mr.
Mahathir Right?”International Journal of Finance & Economics, vol. 8, no. 3, 2003, pp. 205–224.
DOI: 10.3386/w9427.
Elmendorf, D. W., and N. G. Mankiw. “Government Debt.”Handbook of Macroeconomics,
vol. 1, 1999, pp. 1615–1669. DOI: 10.3386/w6470.
Energiebilanzen, A. Evaluation Tables of the Energy Balance for Germany.2018.
Erol, U., and E. S. Yu. “On the Causal Relationship Between Energy and Income for Industrial-
ized Countries.”The Journal of Energy and Development, vol. 11, no. 2, 1987, pp. 113–122.
Ertur, C. “Une Strat
egie de Test de la Racine Unitaire.”1991.
Ertur, C. “M
ethodologies de Test de la Racine Unitaire.”1998.
Feldstein, M. “How to Achieve Stronger US Growth.”Journal of Policy Modeling,vol.36,
no.4,2014,pp.649–653. DOI: 10.1016/j.jpolmod.2014.03.007.
Ferreira, M. C. “Public Debt and Economic Growth: A Granger Causality Panel Data
Approach.”2009.
Fischer, S. “Les D
efis de l’Int
egration Europ
eenne pour la Politique
Energ
etique Allemande.”
Regards sur l’
Economie Allemande: Bulletin
Economique du CIRAC, no. 90, 2009, pp. 31–38. DOI:
10.4000/rea.3590.
318 THE JOURNAL OF ENERGY AND DEVELOPMENT
Gelo, T. “Causality Between Economic Growth and Energy Consumption in Croatia.”Zbornik
Radova Ekonomskog Fakulteta u Rijeci:
Casopis za Ekonomsku Teoriju i Praksu, vol. 27, no. 2,
2009, pp. 327–348.
Giraud, G., and Z., Kahraman. “How Dependent is Growth from Primary Energy? The Depen-
dency Ratio of Energy in 33 Countries (1970-2011).”2014.
G
omez-Puig, M., and S. Sosvilla-Rivero. “The Causal Relationship Between Debt and Growth
in EMU Countries.”Journal of Policy Modeling,vol.37,no.6,2015,pp.974–989. DOI: 10.1016/j.
jpolmod.2015.09.004.
Granger, C. W. “Investigating Causal Relations by Econometric Models and Cross-Spectral
Methods.”Econometrica: Journal of the Econometric Society, vol. 37, no. 3, 1969, pp. 424–438.
DOI: 10.1017/CBO9780511753978.002.
Granger, C. W. “Some Recent Developments in a Concept of Causality.”Journal of Economet-
rics, vol. 39, no. 1-2, 1988, pp. 199–211. DOI: 10.1016/0304-4076(88)90045-0.
Gusbin, D., and B. Hoornaert. Perspectives
Energ
etiques pour la Belgique
al’Horizon 2030.
Bureau F
ed
eral du Plan, 2004.
Hamilton, J. D. Time Series Analysis. Vol. 2, Princeton University Press, 1994.
Hansen, H. “The Impact of External Aid and External Debt on Growth and Investment.”Debt
Relief for Poor Countries, Springer, 2004, pp. 134–157. DOI: 10.1057/9780230522329_7.
Hatemi, A., M. Irandoust, et al. “Energy Consumption and Economic Growth in Sweden: A Lev-
eraged Bootstrap Approach, 1965–2000.”International Journal of Applied Econometrics and Quan-
titative Studies, vol. 2, no. 4, 2005, pp. 87–98. DOI: 10.1016/j.enpol.2008.09.077.
Hatzigeorgiou, E., H. Polatidis, and D. Haralambopoulos. “CO
2
Emissions, GDP and Energy
Intensity: A Multivariate Cointegration and Causality Analysis for Greece, 1977-2007.”Applied
Energy, vol. 88, no. 4, 2011, pp. 1377–1385. DOI: 10.1016/j.apenergy.2010.10.008.
H€
olsgens, R. “Resource Dependence and Energy Risks in the Netherlands Since the Mid-
Nineteenth Century.”Energy Policy, vol. 125, 2019, pp. 45–54. DOI: 10.1016/j.enpol.2018.10.020.
Hondroyiannis, G. “Estimating Residential Demand for Electricity in Greece.”Energy Econom-
ics, vol. 26, no. 3, 2004, pp. 319–334. DOI: 10.1016/j.eneco.2004.04.001.
Hondroyiannis, G., S. Lolos, and E. Papapetrou. “Energy Consumption and Economic Growth:
Assessing the Evidence from Greece.”Energy Economics, vol. 24, no. 4, 2002, pp. 319–336. DOI:
10.1016/S0140-9883(02)00006-3.
Jaramillo, L., and M. C. Cottarelli. Walking Hand in Hand: Fiscal Policy and Growth in
Advanced Economies. No. 12-137, International Monetary Fund, 2012.
Jobert, T., and F. Karanfil. “Sectoral Energy Consumption by Source and Economic Growth in
Turkey.”Energy Policy, vol. 35, no. 11, 2007, pp. 5447–5456. DOI: 10.1016/j.enpol.2007.05.008
Jumbe, C. B. “Cointegration and Causality Between Electricity Consumption and GDP: Empiri-
cal Evidence from Malawi.”Energy Economics, vol. 26, no. 1, 2004, pp. 61–68. DOI: 10.1016/
S0140-9883%2803%2900058-6.
Kraft, J., and A. Kraft. “On the Relationship Between Energy and GNP.”The Journal of Energy
and Development,1978,pp.401–403.
Krugman, P. R. Market-Based Debt-Reduction Schemes. National Bureau of Economic Research,
1988. DOI: 10.3386/w2587.
Kumar, M., and J. Woo. Public Debt and Growth. IMF Working Papers, 2010, pp. 1–47.
Lee, C.-C., C.-P. Chang, and P.-F. Chen. “Energy-Income Causality in OECD Countries Revis-
ited: The Key Role of Capital Stock.”Energy Economics, vol. 30, no. 5, 2008, pp. 2359–2373.
DOI: 10.1016/j.eneco.2008.01.005.
Linkohr, R. “Une Strat
egie Diff
erente de Celle de la France: La Politique
Energ
etique
Allemande.”Annales des Mines-R
ealit
es Industrielles, 2013, pp. 61–68.
319ENERGY, DEBT, AND GROWTH
Marbaix, P., and J.-P. van Ypersele de Strihou. Impacts des Changements Climatiques en Bel-
gique. 2004.
Masih, A. M., and R. Masih. “Energy Consumption, Real Income, and Temporal Causality:
Results from a Multi-Country Study Based on Cointegration and Error-Correction Modeling Tech-
niques.”Energy Economics, vol. 18, no. 3, 1996, pp. 165–183. DOI:10.1016/0140-9883(96)00009-6.
Mencinger, J., A. Aristovnik, and M. Verbic. “The Impact of Growing Public Debt on Economic
Growth in the European Union.”Amfiteatru Economic Journal, vol. 16, no. 35, 2014, pp. 403–414.
Mitze, T., and F. Matz. “Public Debt and Growth in German Federal States: What Can Europe
Learn?”JournalofPolicyModeling, vol. 37, no. 2, 2015, pp. 208–228. DOI: 10.1016/j.jpolmod.
2015.02.003.
Modigliani, F. “Long-Run Implications of Alternative Fiscal Policies and the Burden of the
National Debt.”The Economic Journal, vol. 71, no. 284, 1961, pp. 730–755. DOI: 10.2307/2228247.
Moss, T. J., and H. S. Chiang. The Other Costs of High Debt in Poor Countries: Growth, Policy
Dynamics, and Institutions. 2003.
Mozumder, P., and A. Marathe. “Causality Relationship Between Electricity Consumption and
GDP in Bangladesh.”Energy Policy, vol. 35, no. 1, 2007, pp. 395–402. DOI: 10.1016/J.ENPOL.
2005.11.033.
Murphy, G. “L'Irlande du «Tigre Celtique»: Un Cas Particulier d'«
Etat Comp
etitif».”2016.
Mutascu, M. “A Bootstrap Panel Granger Causality Analysis of Energy Consumption and Eco-
nomic Growth in the G7 Countries.”Renewable and Sustainable Energy Reviews, vol. 63, 2016, pp.
166–171. DOI: 10.1016/j.rser.2016.05.055.
Nguyen, T. Q., M. B. J. Clements, and M. R. Bhattacharya. External Debt, Public Investment,
and Growth in Low-Income Countries. No. 3-249, International Monetary Fund, 2003.
Nyambuu, U., and L. Bernard. “A Quantitative Approach to Assessing Sovereign Default Risk
in Resource-Rich Emerging Economies.”International Journal of Finance & Economics,vol.20,
no.3,2015,pp.220–241. DOI: 10.1002/ijfe.1512.
OECD. “OECD Economic Outlook No 115 (Edition 2024/1).”OECD Economic Outlook: Statis-
tics and Projections, 2024, DOI: 10.1787/65f9bbcb-en.
Ozturk, I., and A. Acaravci. “The Causal Relationship Between Energy Consumption and GDP
in Albania, Bulgaria, Hungary, and Romania: Evidence from ARDL Bound Testing Approach.”
Applied Energy, vol. 87, no. 6, 2010, pp. 1938–1943. DOI: 10.1016/j.apenergy.2009.10.010.
Ozturk, I., A. Aslan, and H. Kalyoncu. “Energy Consumption and Economic Growth Relation-
ship: Evidence from Panel Data for Low and Middle Income Countries.”Energy Policy, vol. 38,
no.8,2010,pp.4422–4428. DOI: 10.1016/j.enpol.2010.03.071.
Panizza, U., and A. F. Presbitero. “Public Debt and Economic Growth: Is There a Causal
Effect?”Journal of Macroeconomics, vol. 41, 2014, pp. 21–41. DOI: 10.1016/j.jmacro.2014.03.009.
Pattillo, C. A., H. Poirson, and L. A. Ricci. External Debt and Growth. No. 2002-2069, Interna-
tional Monetary Fund, 2002.
Paul, S., and R. N. Bhattacharya. “Causality Between Energy Consumption and Economic
Growth in India: A Note on Conflicting Results.”Energy Economics, vol. 26, no. 6, 2004, pp. 977–
983. DOI: 10.1016/j.eneco.2004.07.002.
Poirson, M. H., M. L. A. Ricci, and M. C. A. Pattillo. What Are the Channels Through Which
External Debt Affects Growth? No. 4-15, International Monetary Fund, 2004.
Puente-Ajov
ın, M., and M. Sanso-Navarro. “Granger Causality Between Debt and Growth: Evi-
dence from OECD Countries.”International Review of Economics & Finance, vol. 35, 2015, pp.
66–77. DOI: 10.1016/j.iref.2014.09.007.
Reinhart, C. M., and K. S. Rogoff. “Growth in a Time of Debt.”American Economic Review,
vol. 100, no. 2, 2010, pp. 573–578. DOI: 10.1257/aer.100.2.573.
Reinhart, C., and K. Rogoff. Debt and Growth Revisited.2010.
320 THE JOURNAL OF ENERGY AND DEVELOPMENT
Roubini, N., and J. D. Sachs. “Political and Economic Determinants of Budget Deficits in the
Industrial Democracies.”European Economic Review, vol. 33, no. 5, 1989, pp. 903–933. DOI: 10.
1016/0014-2921(89)90002-0.
Ruostetsaari, I. “Governance and Political Consumerism in Finnish Energy Policy-Making.”
Energy Policy, vol. 37, no. 1, 2009, pp. 102–110. DOI: 10.1016/j.enpol.2008.08.001.
Sacko, I. Analyse des Liens entre Croissance
Economique et Consommation d’
Energie au Mali.
CERFOD-FSJE, Universit
e du Mali, 2004.
Saintherant, N., C. Lerouge, and A. Welcker. “Energy in Ireland: Context, Strategy and
Research.”
Energie en Irlande: Contexte, Strat
egie et Recherche,2008.
Saint-Paul, G. “Fiscal Policy in an Endogenous Growth Model.”The Quarterly Journal of Eco-
nomics, vol. 107, no. 4, 1992, pp. 1243–1259. DOI: 10.2307/2118387.
Schclarek, A., et al. Debt and Economic Growth in Developing and Industrial Countries.Lund
University Department of Economics Working Paper, 2005.
Schifano, F., and D. Moriconi. “L’
Equation
Energ
etique
a Trois Inconnues.”Confluences
M
editerran
ee, no. 1, 2009, pp. 151–160.
Shiu, A., and P.-L. Lam. “Electricity Consumption and Economic Growth in China.”Energy
Policy, vol. 32, no. 1, 2004, pp. 47–54. DOI: 10.1016/S0301-4215(02)00250-1.
Sievers, L., B. Breitschopf, M. Pfaff, and A. Schaffer. “Macroeconomic Impact of the German
Energy Transition and Its Distribution by Sectors and Regions.”Ecological Economics, vol. 160,
2019, pp. 191–204. DOI: 10.1016/j.ecolecon.2019.02.017.
Smyth, D. J., and Y. Hsing. “In Search of an Optimal Debt Ratio for Economic Growth.”Contem-
porary Economic Policy, vol. 13, no. 4, 1995, pp. 51–59. DOI: 10.1111/J.1465-7287.1995.TB00731.X.
Solow, R. M. “Resources and Economic Growth.”The American Economist,vol.61,no.1,
2016, pp. 52–60.
Spilioti, S., and G. Vamvoukas. “The Impact of Government Debt on Economic Growth: An
Empirical Investigation of the Greek Market.”The Journal of Economic Asymmetries,vol.12,
no. 1, 2015, pp. 34–40. DOI: 10.1016/j.jeca.2014.10.001.
Squalli, J. “Electricity Consumption and Economic Growth: Bounds and Causality Analyses of
OPEC Members.”Energy Economics,vol.29,no.6,2007,pp.1192–1205. DOI: 10.1016/j.eneco.
2006.10.001.
Storm, S., and C. Naastepad. “Crisis and Recovery in the German Economy: The Real Lessons.”
Structural Change and Economic Dynamics, vol. 32, 2015, pp. 11–24. DOI: 10.1016/j.strueco.2015.
01.001.
Tanzi, V., and N. Chalk. “Impact of Large Public Debt on Growth in the EU: A Discussion of
Potential Channels.”European Economy, vol. 2, 2000, pp. 23–43.
Toda, H. Y., and T. Yamamoto. “Statistical Inference in Vector Autoregressions with Possibly
Integrated Processes.”Journal of Econometrics, vol. 66, no. 1-2, 1995, pp. 225–250. DOI: 10.1016/
0304-4076(94)01616-8.
Tsani, S. Z. “Energy Consumption and Economic Growth: A Causality Analysis for Greece.”
Energy Economics, vol. 32, no. 3, 2010, pp. 582–590. DOI: 10.1016/j.eneco.2009.09.007.
Vihri€
al€
a, V. The Finnish Economy. 2017.
Whelan, K. “Ireland’s Economic Crisis: The Good, the Bad and the Ugly.”Journal of Macro-
economics, vol. 39, 2014, pp. 424–440. DOI: 10.1016/j.jmacro.2013.08.008.
Woo, J., and M. S. Kumar. “Public Debt and Growth.”Economica, vol. 82, no. 328, 2015, pp.
705–739. DOI: 10.1111/ecca.12138.
Yu, E. S., and J.-Y. Choi. “The Causal Relationship Between Energy and GNP: An International
Comparison.”The Journal of Energy and Development, 1985, pp. 249–272.
321ENERGY, DEBT, AND GROWTH
