The Global Impact of Oil Revenue Dependency: Analysis of Key Indicators from Leading Energy-Producing Countries

Abstract

This study investigates how energy production, which plays a significant role in the economies of countries dependent on oil revenues, affects global energy price dynamics. Drawing particular attention to the Rentier State Theory, the study analyzes the long- and short-term interactions among five key indicators (oil price, public expenditure, exchange rate, corruption control, and carbon emissions) using data from 16 countries between 2001 and 2016, a period of high volatility in global energy markets. The Panel Vector Error Correction Model (PVECM) was used for this study. The analysis results indicate that oil prices are significantly affected in the long term by macroeconomic indicators, environmental factors, and, in particular, GDP growth and carbon emissions, but their short-term effects are more limited. Furthermore, the findings also reveal that corruption control, economic, and environmental factors affect energy market stability. Policymakers are encouraged to develop solutions that consider longer-term dynamics rather than short-term plans and measures. This study provides new insights into how local structural conditions, particularly in the Rentier States, significantly influence and shape the volatility of global oil price movements.

1. Introduction

This research examines the perspective of suppliers in oil-exporting countries and focuses on potential domestic factors that could influence global oil prices. Specifically, it argues that representatives of these countries engage in bargaining processes at the national level for local motivations. The study questions the possibility that the high dependence on oil revenues could lead to internal instability, such as civil conflicts, in these countries, which could have significant impacts on global energy prices. In this context, the research is theoretically based on the Rentier State Theory. In Rentier State structures, a large portion of the state’s income is derived from external sources. Revenues from natural resource exports, in particular, influence the development of other sectors, affecting countries’ economic development both positively and negatively. This situation profoundly impacts both economic and political dynamics.

The study analyzes five key indicators (oil price, public expenditure, exchange rate, corruption score, and carbon emissions) for 16 major energy producers, examining how these indicators relate to domestic instabilities and rentier structures.

As the recent pandemic has highlighted, trade is vital to the global economy and profoundly impacts various aspects of life [1]. Energy is a critical commodity in international trade, and its limited geographical resources make it both a blessing and a challenge. While energy exports generate significant revenue, they also create dependency problems for resource-rich countries [2]. This dependency reduces the importance given to other sectors, limits production diversity in energy-producing economies, and can lead to the neglect of other sectors. This situation poses a problem for countries during any crisis.

The period chosen for the study was specifically chosen due to the high volatility in energy markets. This period has one of the highest volatility levels and is quite unique in its scale, frequency, and geopolitical origins. I would like to emphasize that extraordinary fluctuations in oil prices, geopolitical crises, and supply/demand shocks were experienced during this period. The key events during this period can be summarized as follows:

• 2003–2008: China’s growth → Record increase in oil prices.
• 2008–2010: Global Financial Crisis → Price collapse.
• 2014–2016: Oil oversupply and OPEC conflicts → Second major crash. Furthermore, I would like to remind you that these events created an environment where domestic economic imbalances in countries dependent on rentier income could influence global prices within the framework of the Rentier State Theory [3].

Brent crude oil prices are shaped by global supply and demand, futures market dynamics, and the roles of market participants such as hedgers and speculators [4]. Hedgers aim to manage risk in future transactions, while speculators influence prices based on their predictions. Organizations such as OPEC and OPEC+ control supply to stabilize prices [5]. Furthermore, energy-exporting countries often use their revenues for public investments and security measures, but this can also lead to corruption and economic instability.

• The study hypothesis is as follows:
• H₀ = Global energy prices are affected by the domestic economic, institutional, national security, technological progress, and influence issues of the top 20 oil-exporting countries.
• H₁ = Global energy prices are not significantly related to the domestic affairs of the top 20 oil-exporting countries.
• This study analyzes 16 years of annual data from the top 20 oil-exporting countries to examine how domestic factors affect global oil prices. It uses an econometric model similar to that used by the authors [6] in their study on the determinants of entrepreneurship in the Middle East and North Africa. The study categorizes explanatory variables as economic conditions, institutional quality, national security, technological progress, and welfare.
• Key variables include military spending (national security), public spending (welfare), CO₂ emissions (technological progress), exchange rates (economic context), and control of corruption (institutional quality). The study’s innovative approach involves examining these local factors as predictors of global oil prices, making it the first study to apply this model to analyze oil prices using these predictors. The quantitative significance of these variables is assessed using various forecasting techniques, as detailed in the Methodology section.

General Overview

Academic interest has explored the impact of oil prices on various economic and institutional indicators, including economic performance [7], financial stability [8], and technological progress [9]. However, less attention has been paid to how local factors within energy-producing countries affect global energy prices. This challenging economic development in such countries often leads to inefficient energy use and causes environmental pollution [10]. Figure 1 shows the dependence of oil-producing countries on energy export revenues as a percentage of their GDP in 2015. This dependence is associated with the phenomenon of Dutch Disease, which refers to the negative economic impact that countries with large reserves of natural resources often experience on their economies. The term originates from the economic problems experienced by the Netherlands after the discovery of natural gas in the North Sea in 1959.

The resulting currency appreciation has reduced the competitiveness of exports and increased imports, leading to economic imbalances [12]. To mitigate these problems, policymakers have increased revenues by allocating more financing to growth models focused on small- and medium-sized enterprises to maintain economic competitiveness and protect the economy from energy price fluctuations. Figure 1 shows that more developed countries, such as the United States and the United Kingdom, are less dependent on energy revenues and less sensitive to price shocks despite being large energy producers. In contrast, it is important to note that less developed countries, heavily dependent on energy exports, are more vulnerable to global economic turmoil. In less developed countries, energy revenues are primarily allocated to public spending, security, and imports. However, as mentioned earlier, a significant portion of these revenues also goes to corruption. The figure in Appendix B also shows the oil revenue dependence of both the oil-producing countries in our study, based on their 2015 GDP. The disadvantages of dependence on energy export revenues are both assessed through the Rentier State Theory and explained in numerous studies in relation to the Dutch Disease phenomenon. This economic term dates back to 1959, when large natural gas fields were discovered in the North Sea. This prediction created a paradox that increased the value of local currencies in some countries.

Table 1. Fuel exports (% of merchandise exports) [13].

Country	µ	2001	2003	2004	2005	2006	2007	2009	2010	2012	2013	2014	2015	2016
Angola	96.91						92.23	97.92	97.91	98.37	98.22	97.72	94.02	92.70
UAE	54.12	91.78	48.27	54.06	57.73	62.25	63.72	43.52	49.99	47.23	55.85	50.76	37.65	31.56
Brazil	7.57	3.59	5.19	4.57	5.99	7.68	8.32	9.00	9.90	11.02	7.66	9.35	7.36	6.45
Canada	21.99	15.30	16.12	17.76	21.40	21.14	22.11	24.42	25.32	25.62	26.36	28.89	20.60	17.63
China	1.93	3.16	2.54	2.44	2.31	1.83	1.71	1.70	1.69	1.51	1.53	1.47	1.23	1.28
Algeria	97.73	97.62	98.04	98.16	98.40	98.05	98.38	98.34	98.31	98.40	98.34	97.23	95.84	95.31
UK	10.04	8.03	8.18	8.74	9.35	9.52	10.28	11.01	12.37	13.87	11.42	10.90	7.02	6.26
Iran	73.88	85.21	79.48	78.96	83.08	83.28			70.80		71.72	67.80	57.79	67.59
Kuwait	94.18	93.23	93.44	94.60		96.48	96.30	93.21	92.75		94.22	95.22	92.50	92.74
Mexico	12.17	7.97	11.23	12.39	14.89	15.46	15.67	13.28	13.77	14.06	12.81	10.41	5.89	4.82
Nigeria	92.05	99.66	97.90			98.24	93.67	90.36	87.13	84.04	87.62	90.85	87.87	96.48
Norway	64.00	61.75	61.17	63.62	67.68	67.80	64.30	63.66	63.70	69.82	67.62	64.88	57.69	52.97
Oman	81.98	80.56	76.82	91.09	91.83	91.39	89.06	74.97	77.76	83.54	82.61	81.67	76.16	76.05
Russia	61.41	51.81	54.51	54.70	61.78	62.88	61.46	63.01	65.65	70.29	70.56	69.53	62.84	48.29
Saudi A	87.73	89.47	89.68	89.82	91.04	90.98	90.17	87.80	87.62	88.63	87.57	85.17	78.43	77.55
USA	6.01	1.88	2.11	2.57	3.31	3.74	3.96	5.80	7.18	10.06	10.76	10.99	8.00	7.58

presents the dataset of fuel exporting countries. More than half of the exports in more than 60% of the examined countries are based on fuel exports; moreover, this ratio is over 80% for one-third of the countries. In addition, the table highlights the vital importance of oil revenues for oil producing countries, while revealing their dependence on energy revenues. As a result, political leaders of energy supplying countries formulate policies considering their national priorities and needs regarding factors affecting supply constraints and geopolitical risks. This study is the first to examine the impact of local indicators of energy suppliers on global oil prices. In addition, the study will contribute to the literature by providing an alternative perspective to understand the determinants of global oil prices and may provide additional tools for price forecasting for both hedging and speculators.

2. Theoretical Framework and Literature Review

This study is based on the Rentier State Theory (RST). RST was put forward by Mahdavy [3] in 1970 as a view characterized by the financial dependence of a state on its source revenues (in this case oil).

2.1. Theoretical Framework

The Rentier State Theory posits that a state derives its revenues not from domestic production activities but from rent revenues generated from external sources. In this context, external dependence could pose serious risks and problems for both oil-exporting and importing economies in the future. Furthermore, it is believed that this country will create serious bottlenecks for countries purchasing its oil from abroad and negatively impact production costs as a result of oil price fluctuations. Similar views are also evident in the studies of [14,15], who analyze the impact of Rentier on the global energy transition and the resilience of fossil fuel economies.

The Basic Characteristics and Results of Rentier States can be summarized as follows:

2.1.1. Political Fragility and Autocracy

Rentier states are described as having an “autocratic structure,” which can make them “more vulnerable to internal and international conflicts.” This can also lead to adverse economic conditions. For example, if China’s oil imports from Iran were to cease due to war or conflict in the region, multinational companies producing in China would experience significant fluctuations in their production and costs. This would also cause fluctuations in global markets. Consequently, countries that rely on oil for their income could experience international supply disruptions and global crises as a result of events such as wars, mismanagement, or general political instability.

Some studies contradict this view. In particular, comparative studies on Gulf monarchies have shown that renewable energy projects generally strengthen rather than weaken rentier state structures [16,17].

2.1.2. Deterioration of Institutions

This dependence on a specific source of revenue strengthens the link between resource output and “deteriorating institutional quality” and “civil unrest.” This perspective is also anticipated and supported in broader research on renewable energy transitions, which emphasizes the role of governance quality and institutional frameworks [18,19]. Consequently, institutional deterioration inevitably leads to corruption, operational inefficiencies, poor revenue management, and the deterrence of foreign direct investment (FDI), which in turn reduces the inflow of capital and technology.

2.1.3. Short-Term, Rent-Seeking Behavior

Decision-making processes are heavily influenced by short-term rent-seeking. Similarly, this problem has been studied in cases such as Nigeria and Oman, where rentier financial structures discourage disruptive renewable energy investments [20,21]. This can lead to “unexpected and sudden production decisions, deliberate supply restrictions, and inefficiencies” that contribute to global price fluctuations. The governments of Iran, Iraq, Venezuela, and Nigeria are also examples.

Conditions such as “rent psychology” or “petromania” can arise, in which government officials develop cognitive impairments intertwined with oil wealth.

RST can also have serious impacts on global oil supply and prices. Internal problems in rentier states, in particular, have been observed to directly affect global oil supply and prices in various ways. Some studies also link this to emerging research on “green rentierism,” where states attempt to adapt rentier structures to capture renewable rents [15,19].

Future Supply Expectations and speculation also significantly influence political instability and market expectations in oil-producing countries. The market may anticipate a future supply decrease, which could lead to higher prices “even if actual supply remains unchanged.”

The OPEC/OPEC+ Decisions on production quotas set by these groups are not based solely on global supply and demand. They are heavily influenced by “member country internal difficulties and intra-group adjustment dynamics”, such as the “fiscal requirements” of member states such as Russia and Saudi Arabia.

2.2. Literature Review

In this literature review, we have included the views that can be effective in our study by considering the studies that question the various quantitative methods used to analyze oil prices, focusing on West Texas Intermediate (WTI) futures, Brent futures, and OPEC basket prices. The studies reviewed have included different views to explore the economic, political, and environmental dynamics affected by oil price volatility. Another study [22] analyzed historical data to understand the effects of global supply and demand on oil prices. These studies provide a basic approach to examine long-term oil price movements, especially in relation to changes in global supply, production levels, and consumption patterns.

In another study [23], various factors affecting oil prices were categorized as supply, demand, financial, commodity market, speculative, and political factors. I would also like to make an interesting point: Geopolitical risks, such as terrorist attacks, have been identified as important variables in determining oil prices.

Similarly, some comparative studies on rentier states have noted that renewable energy is often used to maintain rentier privileges, soft power, and state legitimacy [16,21]. It has been observed that while oil rents can certainly support investment, they also carry some risks of negatively impacting the economy without strong institutional reforms and diversification of production [14].

The “Dutch Disease” effect is also directly related to these discussions. Research on the geopolitics of renewable energy envisions how rentier states might restructure themselves as “green rentier” states [4,19].

In the current study, these risks are expressed as increases in military spending. Wars and military activities, particularly in the Middle East, can be considered geopolitical events that generally affect oil prices [23].

One study [24] conducted a comprehensive analysis of eight oil-rich MENA countries (Algeria, Iran, Iraq, Kuwait, Oman, Qatar, Saudi Arabia, and the UAE) between 1996 and 2020 to investigate the relationship between oil revenues, financial development, governance, and renewable energy production. The study stated that oil revenues are directed to renewable energy production, and that it has positively affected the Gulf Cooperation Council countries, where oil revenues are strategically allocated to renewable energy projects. The study emphasizes the importance of effective management in ensuring that oil revenues are used efficiently, while it is less effective in countries where poor management is more common. One research [25] in 2024 analyzed the relationship between oil prices and exchange rates in oil-exporting countries (Canada, Mexico, Russia, Brazil, and Norway) and major oil-importing regions (UK, South Korea, Japan, China, the EU, and India) from 1998 to 2023. In their study, they used a VAR model with time-varying parameters and noted that oil price fluctuations affect exchange rates, especially in oil-exporting countries, while oil price increases increase national wealth, leading to overvaluation of the currency. However, the study noted that there were no significant long-term relationships between oil prices and exchange rates, and the analyzed data presented mixed results.

• One study [26], which is conducted in 2024, focuses on the effects of oil prices on GDP and its components in both oil-exporting and oil-importing countries. The results of the study showed that increasing oil prices tend to stimulate economic growth in oil-exporting countries by increasing government revenues, which in turn facilitates public investment and economic expansion. The findings showed that oil prices have some negative effects on economic growth. In particular, it was noted that excessive dependence on oil exports could reduce growth in the long run due to volatility in oil prices and poor management.
• Another research [27] applied network analysis and quantile regression to examine the impact of oil dependence on political stability in 155 countries from 1995 to 2019. They supported the “resource curse” hypothesis and found that oil-dependent economies, especially those with weak institutions, were more prone to political instability. Their study highlighted the negative effects of oil price volatility on political stability. Similarly, one researcher found that increasing oil prices negatively affected political stability in oil-exporting countries due to economic shocks and civil conflicts. This study also found that although oil price increases can increase government revenues, they can also worsen governance problems such as corruption and mismanagement. One another study [28] examined the economic structure of Bahrain from 1960 to 2010 and focused on the relationship between oil revenues, government expenditures, and economic growth. They concluded that oil revenues are the main driver of economic growth and public expenditures. This is also reflected in the current study, which shows a long-term positive relationship between government spending and oil prices. As can be seen, when the literature is examined, it is seen that there are very few studies similar to our study. The results of our study show that government spending can sometimes contribute to inefficiencies and may not always translate into sustainable economic growth. In a similar study [29,30], it is also found that geopolitical risks, especially wars and conflicts, are effective in affecting military spending and oil price fluctuations. Studies show that there is a positive correlation between oil wealth and military expenditures, especially in oil-exporting regions such as the Middle East and North Africa (1). Another researcher [31] pointed out that there is an inverse relationship between oil price volatility and military expenditures in Gulf Cooperation Council countries [7]. The author stated that oil revenues significantly affect Iran’s military expenditures. These findings confirm the complex interaction between oil prices and military expenditures; geopolitical tensions generally increase oil price volatility. Another study [32,33] investigated the effects of oil price fluctuations on carbon emissions in oil-exporting countries. These studies noted that there was an increase in carbon emissions until 2007, followed by a decrease after the Global Financial Crisis. Ultimately, oil price fluctuations, combined with investments in renewable energy, indicate that the environmental footprint of oil-exporting countries is changing.
• The ‘Dutch Disease’, first proposed in connection with the Groningen gas discovery, describes how a resource boom can lead to currency appreciation and reduce the competitiveness of other exports [34]. While many studies focus on the US dollar exchange rate, this research is based on [2,35] adopting the swap terms of trade index to assess how currency fluctuations affect the decision-making processes of oil exporters. Corruption remains a significant problem in resource-rich countries, where poor governance and institutional weaknesses can undermine economic growth [36,37]. This study offers a unique perspective by examining how local governance issues, including corruption, affect global oil prices and resource management practices. The literature review reveals a complex and multifaceted relationship between oil prices, economic growth, political stability, renewable energy development, and governance. While oil wealth can stimulate economic growth and public spending, volatility, management influence, and misuse of planned resources, especially in oil-exporting countries outside the Gulf Cooperation Council, affect economic success. Effective governance, strategic investments in renewable energy, and diversification efforts are vital to ensure sustainable development of oil-dependent economies. Our research also highlights the need for robust institutional frameworks to effectively manage oil revenues and mitigate the negative impacts of oil price volatility on political stability and economic growth.

3. Data and Methodology

This study uses the Panel Vector Error Correction Model (PVECM) to analyze both long-run and short-run relationships affecting oil prices across countries. PVECM improves the traditional Vector Error Correction Model (VECM) by integrating panel data, enabling it to capture the dynamics within and between countries.

The data examined in our study covers the 16-year period between 2000 and 2016. This period, characterized by one of the highest volatile periods, is quite unique in its scale, frequency, and geopolitical origins. This period also allows for the analysis of three major price cycles: the 2004–2008 boom, the 2008–2010 bust and recovery, and the 2014–2016 second bust.

Below is a list of key oil price moments during the thesis data period.

• 2000–2003: This period saw oil prices rise steadily from USD20 to USD30 due to the recovery of global demand and limited spare capacity.
• 2003–2008: This period, marked by strong demand from China, speculative investments in commodities, tight supply, and geographical tensions (e.g., the Iraq War), saw oil prices reach an all-time high of USD147 per barrel. This period is known as the “Super Peak.”
• 2008–2009: This period is marked by a price decline due to the sharp decline in demand caused by the Global Financial Crisis (GFC).
• 2009–2011: This is the recovery period following the Global Financial Crisis. The recovery in global energy demand brought prices back to USD100 per barrel.
• 2014–2016: This period was affected by the US shale oil boom, OPEC’s supply glut, and the slowdown in demand, particularly from China. Oil prices fell by 25% during this period.

Between 2000 and 2016, crude oil prices reached a low of USD26 and a high of USD147. The largest percentage decline was recorded during the 2008 Global Financial Crisis, which saw three boom–bust cycles. In summary, the data period for this thesis witnessed the financialization of oil markets, three rapid demand booms and busts, the US shale boom revolution, and the significant geopolitical events the thesis aims to analyze.

Methodology Overview

Indicator Analysis: The study evaluates five key indicators, namely security, public policies, technology, finance, and governance quality. These indicators are selected based on their potential impact on global oil prices and reflect various aspects of local conditions.

PVECM Application: PVECM uses panel data that combines cross-sectional data across countries with time series data, enabling the model Equation (1) to capture both interactions within and between countries. The econometric model investigated in this study is shown below:

$$\mathit{d}\mathit{l}\mathit{n}\mathit{O}\mathit{i}\mathit{l}={\mathit{\beta }}_{\mathbf{0}}+{\mathit{\beta }}_{\mathbf{1}}\mathit{d}\mathit{X}\mathit{C}+{\mathit{\beta }}_{\mathbf{2}}\mathit{d}\mathit{l}\mathit{n}\mathit{G}\mathit{X}+{\mathit{\beta }}_{\mathbf{3}}\mathit{d}\mathit{l}\mathit{n}\mathit{G}\mathit{D}\mathit{P}+{\mathit{\beta }}_{\mathbf{4}}\mathit{d}\mathit{l}\mathit{n}\mathit{C}\mathit{O}\mathbf{2}+{\mathit{\beta }}_{\mathbf{5}}\mathit{d}\mathit{C}\mathit{C}+έ$$

(1)

where

dlnOil is the natural logarithm of the Brent Petro Crude oil prices globally;

dlnGX natural logarithm of the government expenditures;

dlnGDP natural logarithm of the gross domestic product;

dlnCO2 is the natural logarithm of carbon dioxide emissions;

dXC is the exchange rate;

dCC is the corruption score.

It analyses how sudden changes in local factors affect oil prices by evaluating long-term balances and short-term deviations. The model also uses error correction to adjust for short-term imbalances and helps determine which indicators have the most significant impact on oil price dynamics. The data investigated in the study covers 16 years from 2001 to 2016. The period was specifically chosen because of its high volatility in energy markets. The list is arranged from the least to the most according to the country’s LPG and oil production (E-views).

Countries Investigated

• United Kingdom, Oman, Angola, Algeria, Nigeria, Mexico, Norway, Kuwait, Brazil, United Arab Emirates, Iran, China, Canada, Saudi Arabia, Russia, United States; Countries Excluded: Iraq, Kazakhstan, Qatar, Libya.
• In this study, Iraq, Kazakhstan, Qatar, and Libya were excluded from the country sample due to a lack of reliable data. Specifically, while the period we studied for Iraq and Libya coincided with periods of war, it was not possible to obtain consistent data for Kazakhstan and Qatar, and these countries were excluded from our study.

Variables employed in the research are demonstrated at

Table 2. List of variables.

Variable	Description	Source	Unit
lnOIL	Natural logarithm of Brent Crude Oil Prices.	SIPRI	Current USD
lnGX	Natural logarithm of the general government final consumption expenditures.	World Bank Data	Current USD
lnGDP	Natural logarithm of gross domestic product.	World Bank data	Current USD
lnCO2	Natural logarithm of carbon dioxide emissions of those stemming from the burning of fossil fuels and the manufacture of cement. They include carbon dioxide produced during consumption of solid, liquid, and gas fuels.	CAIT	Kg per PPP USD of GDP
XC	Exchange rate of the local currency regarding barter terms of trade index, which is calculated as the percentage ratio of the export unit value indexes, measured relatively to the base year 2000.	World Bank Data	100 Base Year 2000
CC	Control of corruption that reflects perceptions of the extent to which public power is exercised for private gain, including both petty and grand forms of corruption, as well as “capture” of the state by elites and private interest.	WGI	−2.5 (weak) to +2.5 (strong) governance performance

.

4. Empirical Estimation

Table 3. Descriptive group statistics.

	DLNOIL	DXC	DLNGX	DLNGDP	DLNCO2	DCC
Mean	0.038596	1.076843	0.085141	0.076770	−0.023227	−0.005167
Median	0.105913	1.068100	0.082515	0.097830	−0.026903	−0.005000
Maximum	0.355079	77.39100	1.874950	0.536820	0.395744	1.010000
Minimum	−0.637438	−104.9340	−0.728380	−0.559600	−0.269901	−0.350000
Std. Dev.	0.278769	24.59223	0.192448	0.155927	0.075259	0.136728
Skewness	−1.044286	−1.302253	3.003147	−0.851938	0.720687	1.554956
Kurtosis	3.306819	7.098326	34.96020	4.682826	7.556270	15.21296
Jarque–Bera	44.56273	235.7973	10575.30	57.35097	228.3716	1588.279
Probability	0.000000	0.000000	0.000000	0.000000	0.000000	0.000000
Sum	9.262960	258.4423	20.43388	18.42476	−5.574575	−1.240000
Sum Sq. Dev.	18.57316	144541.9	8.851696	5.810831	1.353681	4.467993
Observations	240	240	240	240	240	240

Source: Author’s estimation.

was calculated using time series data covering 240 observations (the dataset is annual) using the E-views 13 program. It includes descriptive statistics such as mean, median, maximum, minimum, standard deviation, skewness, and kurtosis. The Jarque–Bera normality test was applied to determine whether the series exhibited a normal distribution. These statistics include the central tendency, distribution, shape, and normality of the data. All calculations were performed using first-derivative transformed variables and logarithmic transformations.

Table 3 presents descriptive statistics for the parameters used in the analysis and includes their basic characteristics throughout the sample period. The table presents statistics for six variables: DLNOIL (change in oil prices), DXC (exchange rate), DLNGX (change in natural gas exports), DLNGDP (change in GDP), DLNCO2 (change in CO₂ emissions), and DCC (capital–labor ratio).

Descriptive analysis values help identify patterns, possible outliers, and deviations from normality, which are important in selecting appropriate econometric models for further analysis.

Table 3 presents descriptive statistics for six variables: DLNOIL, DXC, DLNGX, DLNGDP, DLNCO2, and DCC. Most variables except DLNCO2 and DCC have small positive means. The other two variables have negative means. DXC shows high variability and extreme outliers, while DCC has low variability. Skewness and kurtosis indicate non-normal distributions; DLNGX and DCC have heavy tails and extreme values. The Jarque–Bera test confirms non-normality for all variables. DXC has the highest variance, while DCC and DLNCO2 are negatively skewed with moderate variability. Overall, the dataset shows significant outliers and non-normal distributions.

4.1. Unit Root Tests Results

Results from the unit root tests from

Table 4. Unit root test result at the level trend for lnOil, CC, XC, lnGDP, lnCO2, and ln GX for the period between 2001 and 2016.

Variable	Unit Root Tests
At Level Trend and Intercept
CC	Null: Unit root (assumes common unit root process)
	Test	Test Statistic	Probability value
	Levin, Lin & Chu t *	−1.77969	0.0376
	Null: Unit root (assumes individual unit root process)
	Im, Pesaran and Shin W-stat	1.08484	0.8610
	ADF-Fisher Chi-square	26.5715	0.7378
	PP-Fisher Chi-square	57.7539	0.0035
XC	Null: Unit root (assumes common unit root process)
	Levin, Lin & Chu t *	6.56148	1.0000
	Null: Unit root (assumes individual unit root process
	Im, Pesaran and Shin W-stat	5.64590	1.0000
	ADF-Fisher Chi-square	2.97051	1.0000
	PP-Fisher Chi-square	6.73614	1.0000
lnCO2	Null: Unit root (assumes common unit root process)
	Levin, Lin & Chu t *	−2.02437	0.0215
	Null: Unit root (assumes individual unit root process
	Im, Pesaran and Shin W-stat	1.65796	0.9513
	ADF-Fisher Chi-square	26.9212	0.7215
	PP-Fisher Chi-square	14.7480	0.9961
lnGDP	Null: Unit root (assumes common unit root process)
	Levin, Lin & Chu t *	3.38887	0.9996
	Null: Unit root (assumes individual unit root process
	Im, Pesaran and Shin W-stat	5.91842	1.0000
	ADF-Fisher Chi-square	8.03248	1.0000
	PP-Fisher Chi-square	2.38539	1.0000
lnGX	Null: Unit root (assumes common unit root process)
	Levin, Lin & Chu t *	2.27095	0.9884
	Null: Unit root (assumes individual unit root process)
	Im, Pesaran and Shin W-stat	4.65135	1.0000
	ADF-Fisher Chi-square	9.81231	0.9999
	PP-Fisher Chi-square	7.02463	1.0000
lnOil	Null: Unit root (assumes common unit root process)
	Levin, Lin & Chu t *	8.33001	1.0000
	Null: Unit root (assumes individual unit root process)
	Im, Pesaran and Shin W-stat	8.13931	1.0000
	ADF-Fisher Chi-square	0.35135	1.0000
	PP-Fisher Chi-square	0.03192	1.0000

Source: estimated by author. Note: * indicate that variables are statistically significant at the 10% levels.

and

Table 5. Unit root test result at first difference for lnOil, CC, XC, lnGDP, lnCO2, and ln GX for the period between 2001 and 2016.

Variable	Unit Root Tests
First differentiation Trend and Intercept
dCC	Null: Unit root (assumes common unit root process)
	Test	Test Statistic	Probability value
	Levin, Lin & Chu t *	3.15011	0.0092
	Null: Unit root (assumes individual unit root process)
	Im, Pesaran and Shin W-stat	−1.76658	0.0386
	ADF-Fisher Chi-square	45.5085	0.0574
	PP-Fisher Chi-square	171.409	0.0000
dXC	Null: Unit root (assumes common unit root process)
	Levin, Lin & Chu t *	−5.29316	0.0000
	Null: Unit root (assumes individual unit root process
	Im, Pesaran and Shin W-stat	−2.57873	0.0050
	ADF-Fisher Chi-square	52.9798	0.0113
	PP-Fisher Chi-square	162.551	0.0000
lnCO2	Null: Unit root (assumes common unit root process)
	Levin, Lin & Chu t *	−5.19560	0.0000
	Null: Unit root (assumes individual unit root process
	Im, Pesaran and Shin W-stat	−3.64945	0.0001
	ADF-Fisher Chi-square	69.3400	0.0001
	PP-Fisher Chi-square	148.671	0.0000
lnGDP	Null: Unit root (assumes common unit root process)
	Levin, Lin & Chu t *	−5.03727	0.0000
	Null: Unit root (assumes individual unit root process
	Im, Pesaran and Shin W-stat	−2.31011	0.0104
	ADF-Fisher Chi-square	49.9843	0.0224
	PP-Fisher Chi-square	141.227	0.0000
dlnGX	Null: Unit root (assumes common unit root process)
	Levin, Lin & Chu t *	2.27095	0.9884
	Null: Unit root (assumes individual unit root process)
	Im, Pesaran and Shin W-stat	4.65135	1.0000
	ADF-Fisher Chi-square	9.81231	0.9999
	PP-Fisher Chi-square	7.02463	1.0000
dlnOil	Null: Unit root (assumes common unit root process)
	Levin, Lin & Chu t *	−6.78486	0.0000
	Null: Unit root (assumes individual unit root process)
	Im, Pesaran and Shin W-stat	−2.61354	0.0045
	ADF-Fisher Chi-square	51.4743	0.0160
	PP-Fisher Chi-square	155.651	0.0000

Source: estimated by author. Note: * indicate that variables are statistically significant at the 10% levels.

reveal that oil prices, exchange rates, government expenditures, and control of corruption demonstrate stationarity at first difference. Therefore, long and short run relationships between the variables are investigated, employing panel VECM technique. The table below shows the unit root tests at the level trend and intercept for and in the first difference.

In the first differentiation, all variables become stationary, as observed in Table 5. The probability values for all considered parameters for the unit root tests, namely ADF, PP, P/S, were almost at zero-level exemption of the governmental expenditures. We can reject the null hypothesis and we accept the alternative hypothesis where the parameters become stationary.

4.2. Cointegration Tests

Table 6. Johansen cointegration test.

Null Hypothesis	Fisher Stat.* Trace Statistic	Probability	Fisher Stat.* Max-EigenStatistic	Probability
r ≤ 0 *	0.000	1.0000	0.000	1.0000
r ≤ 1 *	0.000	1.0000	0.000	1.0000
r ≤ 2 *	196.6	0.0000	700.9	0.0000
r ≤ 3 *	473.8	0.0000	423.4	0.0000
r ≤ 4 *	563.4	0.0000	620.6	0.0000
r ≤ 5 *	4214.	0.0000	4214.	0.0000

Source: Authors’ calculations. * Probabilities are computed using asymptotic Chi-square distribution.

presents the results of a Johansen Cointegration Test, which evaluates the number of cointegration relationships (r) among time series.

The null hypothesis at each level (r ≤ 0, r ≤ 1, etc.) tests whether there are at most this many cointegration relationships. Fisher and Trace Statistics, along with their associated p-values, help determine whether the null hypothesis can be rejected. The p-values for r ≤ 0 and r ≤ 1 are 1.0000, indicating no cointegration. However, for r ≥ 2, both statistics consistently reject the null hypothesis, suggesting at least five cointegration relationships with p-values of 0.0000. Therefore, the results indicate r = 5 or 6 cointegration relationships.

4.3. Lag Length Results

Table 7. Lag length.

Lag	LogL	LR	FPE	AIC	SC	HQ
0	−78.48387	NA	1.05 × 10−7	0.960044	1.068129 *	1.003883
1	−16.45786	119.1181	7.83 × 10−8 *	0.664294 *	1.420887	0.971164 *
2	2.607659	35.31454	9.50 × 10−8	0.856731	2.261832	1.426633
3	44.17649	74.16258	8.95 × 10−8	0.793449	2.847058	1.626383
4	78.67576	59.19761 *	9.17 × 10−8	0.810503	3.512620	1.906468

Source: Author’s estimation, * indicates lag order selected by the criterion, LR: sequentially modified LR test statistic (each test at 5% level), FPE: final prediction error, AIC: Akaike information criterion, SC: Schwarz information criterion. Source: Authors’ estimation.

presents the VAR (Vector Autoregression) Lag Order Selection Criteria for a model with six endogenous variables and one exogenous variable covering the period 2001–2016. The criteria used to determine the optimal lag length include LogL, LR (Likelihood Ratio), FPE (Post-Forecast Error), AIC (Akaike Information Criterion), SC (Schwarz Criterion), and HQ (Hannan–Quinn Criterion). The results show that Lag 1 is the optimal choice by most of the criteria (LR, FPE, AIC, and HQ) providing the best balance of model fit and complexity. Lag 4 shows a strong LR statistic but is penalized by the higher values of AIC, SC, and HQ, making it less preferred than Lag 1. Therefore, Lag 1 is the optimal lag length for the model.

Lag length was determined based on the Posterior Error of Prediction (FPE), Akaike Information Criterion (AIC), and Hannan–Quinn (HQ) criteria, with the latter primarily suggesting lag 1. While the LR test suggested Lag 4, longer lag structures were observed to carry a risk of overfitting given the limited timeframe (T = 240). Therefore, Lag 1 was chosen for parsimony and model stability. However, it should be noted that robustness checks at Lags 2 and 4 produced qualitatively similar long-term results, confirming the consistency of the main findings.

$$\begin{array}{cc}\hfill D\left(DLNOIL\right)& =C\left(1\right)\ast \left(DLNOIL\right(-1)+0.0105712655577\ast DXC(-1)+2.76205718208\ast DLNGX(-1)\hfill \\ & -5.83149660039\ast DLNGDP(-1)-1.56781197867\ast DLNCO2(-1)+5.41920132527\hfill \\ & \ast DCC(-1)+0.169257605158)+C(2)\ast D(DLNOIL(-1))+C(3)\ast D(DLNOIL(-2))\hfill \\ & +C\left(4\right)\ast D\left(DXC\right(-1\left)\right)+C\left(5\right)\ast D\left(DXC\right(-2\left)\right)+C\left(6\right)\ast D\left(DLNGX\right(-1\left)\right)+C\left(7\right)\hfill \\ & \ast D\left(DLNGX\right(-2\left)\right)+C\left(8\right)\ast D\left(DLNGDP\right(-1\left)\right)+C\left(9\right)\ast D\left(DLNGDP\right(-2\left)\right)+C\left(10\right)\hfill \\ & \ast D\left(DLNCO2\right(-1\left)\right)+C\left(11\right)\ast D\left(DLNCO2\right(-2\left)\right)+C\left(12\right)\ast D\left(DCC\right(-1\left)\right)+C\left(13\right)\hfill \\ & \ast D\left(DCC\right(-2\left)\right)+C\left(14\right)\hfill \end{array}$$

(2)

$$\begin{array}{cc}\hfill D\left(DXC\right)& =C\left(15\right)\ast \left(DLNOIL\right(-1)+0.0105712655577\ast DXC(-1)+2.76205718208\ast DLNGX(-1)\hfill \\ & -5.83149660039\ast DLNGDP(-1)-1.56781197867\ast DLNCO2(-1)+5.41920132527\hfill \\ & \ast DCC(-1)+0.169257605158)+C(16)\ast D(DLNOIL(-1))+C(17)\hfill \\ & \ast D\left(DLNOIL\right(-2\left)\right)+C\left(18\right)\ast D\left(DXC\right(-1\left)\right)+C\left(19\right)\ast D\left(DXC\right(-2\left)\right)+C\left(20\right)\hfill \\ & \ast D\left(DLNGX\right(-1\left)\right)+C\left(21\right)\ast D\left(DLNGX\right(-2\left)\right)+C\left(22\right)\ast D\left(DLNGDP\right(-1\left)\right)+C\left(23\right)\hfill \\ & \ast D\left(DLNGDP\right(-2\left)\right)+C\left(24\right)\ast D\left(DLNCO2\right(-1\left)\right)+C\left(25\right)\ast D\left(DLNCO2\right(-2\left)\right)+C\left(26\right)\hfill \\ & \ast D\left(DCC\right(-1\left)\right)+C\left(27\right)\ast D\left(DCC\right(-2\left)\right)+C\left(28\right)\hfill \end{array}$$

(3)

$$\begin{array}{cc}\hfill D\left(DLNGX\right)& =C\left(29\right)\ast \left(DLNOIL\right(-1)+0.0105712655577\ast DXC(-1)+2.76205718208\ast DLNGX(-1)\hfill \\ & -5.83149660039\ast DLNGDP(-1)-1.56781197867\ast DLNCO2(-1)+5.41920132527\hfill \\ & \ast DCC(-1)+0.169257605158)+C(30)\ast D(DLNOIL(-1))+C(31)\hfill \\ & \ast D\left(DLNOIL\right(-2\left)\right)+C\left(32\right)\ast D\left(DXC\right(-1\left)\right)+C\left(33\right)\ast D\left(DXC\right(-2\left)\right)+C\left(34\right)\hfill \\ & \ast D\left(DLNGX\right(-1\left)\right)+C\left(35\right)\ast D\left(DLNGX\right(-2\left)\right)+C\left(36\right)\ast D\left(DLNGDP\right(-1\left)\right)+C\left(37\right)\hfill \\ & \ast D\left(DLNGDP\right(-2\left)\right)+C\left(38\right)\ast D\left(DLNCO2\right(-1\left)\right)+C\left(39\right)\ast D\left(DLNCO2\right(-2\left)\right)+C\left(40\right)\hfill \\ & \ast D\left(DCC\right(-1\left)\right)+C\left(41\right)\ast D\left(DCC\right(-2\left)\right)+C\left(42\right)\hfill \end{array}$$

(4)

$$\begin{array}{cc}\hfill D\left(DLNGDP\right)& =C\left(43\right)\ast \left(DLNOIL\right(-1)+0.0105712655577\ast DXC(-1)+2.76205718208\hfill \\ & \ast DLNGX(-1)-5.83149660039\ast DLNGDP(-1)-1.56781197867\ast DLNCO2(-1)\hfill \\ & +5.41920132527\ast DCC(-1)+0.169257605158)+C(44)\ast D(DLNOIL(-1))+C(45)\hfill \\ & \ast D\left(DLNOIL\right(-2\left)\right)+C\left(46\right)\ast D\left(DXC\right(-1\left)\right)+C\left(47\right)\ast D\left(DXC\right(-2\left)\right)+C\left(48\right)\hfill \\ & \ast D\left(DLNGX\right(-1\left)\right)+C\left(49\right)\ast D\left(DLNGX\right(-2\left)\right)+C\left(50\right)\ast D\left(DLNGDP\right(-1\left)\right)+C\left(51\right)\hfill \\ & \ast D\left(DLNGDP\right(-2\left)\right)+C\left(52\right)\ast D\left(DLNCO2\right(-1\left)\right)+C\left(53\right)\ast D\left(DLNCO2\right(-2\left)\right)+C\left(54\right)\hfill \\ & \ast D\left(DCC\right(-1\left)\right)+C\left(55\right)\ast D\left(DCC\right(-2\left)\right)+C\left(56\right)\hfill \end{array}$$

(5)

$$\begin{array}{cc}\hfill D\left(DLNCO2\right)& =C\left(57\right)\ast \left(DLNOIL\right(-1)+0.0105712655577\ast DXC(-1)+2.76205718208\hfill \\ & \ast DLNGX(-1)-5.83149660039\ast DLNGDP(-1)-1.56781197867\ast DLNCO2(-1)\hfill \\ & +5.41920132527\ast DCC(-1)+0.169257605158)+C(58)\ast D(DLNOIL(-1))+C(59)\hfill \\ & \ast D\left(DLNOIL\right(-2\left)\right)+C\left(60\right)\ast D\left(DXC\right(-1\left)\right)+C\left(61\right)\ast D\left(DXC\right(-2\left)\right)+C\left(62\right)\hfill \\ & \ast D\left(DLNGX\right(-1\left)\right)+C\left(63\right)\ast D\left(DLNGX\right(-2\left)\right)+C\left(64\right)\ast D\left(DLNGDP\right(-1\left)\right)+C\left(65\right)\hfill \\ & \ast D\left(DLNGDP\right(-2\left)\right)+C\left(66\right)\ast D\left(DLNCO2\right(-1\left)\right)+C\left(67\right)\ast D\left(DLNCO2\right(-2\left)\right)+C\left(68\right)\hfill \\ & \ast D\left(DCC\right(-1\left)\right)+C\left(69\right)\ast D\left(DCC\right(-2\left)\right)+C\left(70\right)\hfill \end{array}$$

(6)

$$\begin{array}{cc}\hfill D\left(DCC\right)& =C\left(71\right)\ast \left(DLNOIL\right(-1)+0.0105712655577\ast DXC(-1)+2.76205718208\ast DLNGX(-1)\hfill \\ & -5.83149660039\ast DLNGDP(-1)-1.56781197867\ast DLNCO2(-1)+5.41920132527\hfill \\ & \ast DCC(-1)+0.169257605158)+C(72)\ast D(DLNOIL(-1))+C(73)\hfill \\ & \ast D\left(DLNOIL\right(-2\left)\right)+C\left(74\right)\ast D\left(DXC\right(-1\left)\right)+C\left(75\right)\ast D\left(DXC\right(-2\left)\right)+C\left(76\right)\hfill \\ & \ast D\left(DLNGX\right(-1\left)\right)+C\left(77\right)\ast D\left(DLNGX\right(-2\left)\right)+C\left(78\right)\ast D\left(DLNGDP\right(-1\left)\right)+C\left(79\right)\hfill \\ & \ast D\left(DLNGDP\right(-2\left)\right)+C\left(80\right)\ast D\left(DLNCO2\right(-1\left)\right)+C\left(81\right)\ast D\left(DLNCO2\right(-2\left)\right)+C\left(82\right)\hfill \\ & \ast D\left(DCC\right(-1\left)\right)+C\left(83\right)\ast D\left(DCC\right(-2\left)\right)+C\left(84\right)\hfill \end{array}$$

(7)

4.4. Long-Run Effects for Equations RGDP and Ordinary Least Squares (OLS) for Other Dependent Variables

Appendix A estimates the long-run effects from the equations given above for the variables considered. The Ordinary Least Squares (OLS) results show that the long-run adjustment speed for Equation (2) is −2 percent to bring the entire system to equilibrium. Even though the probability value is greater than 0.05, there is long-run causality from the independent variables. This means that the independent variables have an effect on oil.

4.5. Wald Test Results, Short-Run Effects

The results of Equation (2) are presented in

Table 8. Wald test results for DlnOil.

Dependent Variable	Independent Variable	Chi-Square	Probability
DlnOil	DXC	2.979743	0.2254
	DlnGX	0.471914	0.7898
	DlnGDP	2.353991	0.3082
	DlnCO2	0.824290	0.6622
	DCC	1.864390	0.3937

Source: Author’s calculations.

, which shows the results of the Wald test conducted on the dependent variable Dlnoil. This test is applied to determine the effect of combined lags on the dependent variable oil. If the probability value is both greater than 0.05 and zero (i.e., the probability values for all coefficients should be greater than 0.05), we can reject the null hypothesis. A probability value greater than zero means that these coefficients cannot affect the common dependent variable. Table 7 reveals that the independent variables considered in Equation (2) do not have a significant effect on the dependent variable Fat.

When we look at Equation (3), where DXC becomes a dependent variable with short run causality, and compare it with the independent variables Dlnoil, DlnGX, DlnGDP, DlnCO2, and DCC with short run causality, it is shown from the estimated results that there is no short run causality running from independent variables to DXC (see

Table 9. Wald test results for DXC.

Dependent Variable	Independent Variable	Chi-Square	Probability
DXC	DlnOil	0.302745	0.8595
	DlnGX	1.644317	0.4395
	DlnGDP	0.622052	0.7327
	DlnCO2	0.663451	0.0734
	DCC	1.740318	0.4189

Source: Author’s calculations.

).

Equation (4) shows the short run results for the dependent variable DlnGX. Except for the CO₂, all other independent variables Dlnoil, DXC, DlnGDP, and DCC have no short run causality running from independent variables to DlnGX (see

Table 10. Wald test results for DGX.

Dependent Variable	Independent Variable	Chi-Square	Probability
DlnGX	DlnOil	2.945348	0.2293
	DXC	2.350645	0.3087
	DlnGDP	3.043239	0.2184
	DlnCO2	6.584637	0.0372
	DCC	2.240943	0.3261

Source: Author’s calculations.

).

Equation (5) shows the short run results for the dependent variable DlnGDP. Only the DXC has some influence on DlnGDP; other independent variables Dlnoil, DlnGX, DlnCO2, and DCC have no short run causality running from independent variables to DlnGDP (see

Table 11. Wald test results for DGDP.

Dependent Variable	Independent Variable	Chi-Square	Probability
DlnGDP	DlnOil	3.876749	0.1439
	DXC	6.341761	0.0420
	DlnGX	0.381252	0.8264
	DlnCO2	2.179596	0.3363
	DCC	4.132809	0.1266

Source: Author’s calculations.

).

Equation (6) shows the short run results for the dependent variable DlnCO2. The independent variables Dlnoil, DXC, DlnGX, DlnGDP, and DCC have no short run causality running from independent variables to DlnCO2 (see

Table 12. Wald test results for DO2.

Dependent Variable	Independent Variable	Chi-Square	Probability
DlnCO2	DlnOil	1.672541	0.4333
	DXC	2.334875	0.3112
	DlnGX	2.259353	0.3231
	DlnGDP	3.398898	0.1828
	DCC	2.119528	0.3465

Source: Author’s calculations.

).

Equation (7) shows the short run results for the dependent variable DCC. Except for DlnOil, all other independent variables have some influence on DCC (see

Table 13. Wald test results for DCC.

Dependent Variable	Independent Variable	Chi-Square	Probability
DCC	DlnOil	3.976421	0.1369
	DXC	4.828307	0.0894
	DlnGX	24.94431	0.0001
	Dln GDP	22.23291	0.0000
	DlnCO2	7.301307	0.0260

Source: Author’s calculations.

).

4.6. Results and Discussion

The results obtained from the ARDL model estimation provided important insights into the long-term relationships between the key variables that may be related to energy and macroeconomic factors as well as the dependent variable (potentially related to oil).

The analysis revealed the following important findings:

In the long run, in the second equation, the independent variables’ gross national product, C(8) and C(9), have a statistically significant negative relationship with the dependent variable, while the exchange rate, C4 and C5, and corruption control, C4 and C5, also have a negative relationship, but these variables are not statistically significant. As is known, a negative coefficient in ARDL models indicates an increase in the independent variable, with all else being equal.

In the fourth equation, only carbon emissions affect government spending. A previous study [25] found that increases in oil prices lead to currency appreciation in oil-exporting countries, but their study did not find significant long-term relationships between oil prices and exchange rates. This is consistent with the existing study, which also found that exchange rates have weak long-term effects on the dependent variables.

In the fifth equation, oil prices and corruption scores affect gross national product. Another study [26] shows that increasing oil prices stimulate economic growth in oil-exporting countries by increasing government revenues, which in turn stimulates public investment. In contrast, the current study finds that oil prices have a negative impact of 13% on economic growth, indicating that over-reliance on oil exports can harm long-term growth due to volatility and mismanagement.

In the sixth equation, only gross national product affects carbon emissions. In the seventh equation, all independent variables appear to have an impact on the corruption score. Increases in these variables (GDP, corruption score, oil prices, and CO₂ emissions, which are probably energy-related factors) are thought to cause long-term decreases in the dependent variable. These results are an important warning for policymakers who aim to influence energy markets or related sectors.

The probability values of many coefficients are statistically insignificant, indicating that certain variables may not contribute significantly to the model. In particular, the effects of exchange rates and public expenditures can be questioned in future studies or evaluated by removing them from the model.

From an economic perspective, it has been observed that some variables have significant negative effects on corruption scores, GDP, and emissions. Negative coefficients indicate emission reductions or energy efficiency improvements. Long-term policies are needed in energy markets, especially in sectors such as corruption scores, consumption, and emissions interactions.

Short-term effects: Wald test results show that independent variables generally have limited effects on dependent variables in the short term. Equation (2) (Dlnoil as a dependent variable): The Wald test shows that independent variables do not have a significant effect on oil prices (Dlnoil) in the short term, because the p-values of all coefficients are greater than 0.05, indicating that there is no significant effect on the dependent variable. In contrast, another research [23] identified geopolitical risks such as terrorist attacks as important factors affecting oil prices. However, these effects are shown in this study, especially in the short term; government and military expenditures have no effect on oil prices.

Equation (3) (DXC as dependent variable): There is no short-term causality from the independent variables (Dlnoil, DlnGX, DlnGDP, DlnCO2, DCC) to DXC, indicating that these factors do not have an immediate effect on DXC.

Equation (4) (DlnGX as dependent variable): Except for CO₂, none of the independent variables (Dlnoil, DXC, DlnGDP, DCC) show short-term causality to DlnGX, indicating limited short-term effect from these factors.

Equation (5) (DlnGDP as dependent variable): Only DXC shows short-term effect on DlnGDP. Other variables (Dlnoil, DlnGX, DlnCO2, DCC) do not have significant short-term effects on DlnGDP.

Equation (6) (DlnCO2 as dependent variable): None of the independent variables (Dlnoil, DXC, DlnGX, DlnGDP, DCC) show short-term causality to DlnCO2, indicating no immediate effects from the factors.

Equation (7) (DCC as dependent variable): Except for DlnOil, all other independent variables showed that they have a short-term effect on DCC [23]. A study on oil-rich MENA countries, especially those with strong governance, emphasized that directing oil rents to renewable energy production has positive effects on sustainable production and economy. Similarly, our study draws attention to the importance of governance in effectively managing oil revenues and reducing corruption. This is also consistent with a previous study [24], which shows that the quality of governance is important in the effective use of oil rents.

The results show that the short-term effects between the independent and dependent variables are generally weak. In particular, no significant short-term effects are observed for most independent variables, such as oil prices (Dlnoil), exchange rates (DXC), government expenditure (DlnGX), and CO² emissions (DlnCO2), indicating that changes in economic policies are unlikely to lead to sudden changes in oil prices and government speculation.

Our research draws attention to the fact that an economic policy based solely on oil revenues will not be sufficient for a sustainable economic structure, and that volatility in oil prices will have negative effects on political stability and economic growth. Therefore, in order to reduce the negative effects on economic growth, environmentally friendly production techniques and production that reduce carbon emissions should be supported. Again, a proper and sustainable economic policy and solid institutional frameworks are also needed.

In this context, I would like to state that economic growth should be achieved with long-term economic plans according to the results of our study. According to the analysis findings, although some variables such as gross domestic product and oil prices have negative relationships with dependent variables in the long term, the weak effects of exchange rates and government spending indicate that long-term policies are needed to address issues such as corruption and economic growth.

5. Conclusions

This study examines the effects of domestic indicators of rentier states (GDP, government spending, exchange rates, carbon emissions, and corruption control) between 2001 and 2016, particularly during the period of greatest global oil price volatility, using a Panel VECM.

This study examines the political, economic, and environmental indicators of local governments, defined in this study as GDP, government spending, exchange rates, carbon emissions, and corruption control. Including these variables in the analysis of global oil price dynamics challenges existing approaches in energy economics and Rentier State Theory.

As the results suggest, GDP and CO₂ emissions significantly affect oil prices and government spending in the long run. All variables, however, are observed to strongly influence corruption control. Exchange rates and public spending, on the other hand, were found to have weaker and generally insignificant effects—often statistically insignificant. Short-term causality also has limited effects. This suggests that global oil prices may prompt urgent policy changes in oil-exporting countries. In the long term, structural and institutional conditions were observed to be influential.

An important assessment of this research is its emphasis on governance: corruption control emerged as the most affected variable in the study. It emphasizes that fragility in institutional structures is the fundamental problem of rentier economies. In particular, it demonstrates that dependence on oil revenues undermines the development of other sectors and that macroeconomic volatility in oil prices creates governance challenges.

Policy implications: Sustainable economic growth cannot be based solely on oil; diversification in production and the development of other sectors will allow the economy to achieve a more robust and enduring structure. A sustainable economic model can be established, particularly by increasing investments in environmentally friendly production. Volatility in oil prices and the temporary revenues generated by high prices during certain periods are not sufficient for a stable economic structure; if similarly weak governance persists, long-term stability will not be achieved.

This study examines economic indicators by integrating them with the dynamics of energy markets, particularly by considering political indicators, and demonstrates that oil prices are not only determined by external shocks but also influenced by the internal structures of exporting countries. From this perspective, this research contributes to the existing literature. Authors’ contribution: Akalpler largely wrote and analyzed this paper. Aker collected the data review literature and contributed partly to the writing of this paper.

5.1. Short-Term Policy Recommendations

In the short term, no significant impact has been observed on oil prices, exchange rates, or public spending. Therefore, necessary adjustments should be made to maintain macroeconomic stability against sudden shocks.

It is recommended that oil revenues be directed to areas that will increase production diversity rather than covering the short-term budget deficit.

In the short term, measures should be increased to effectively combat corruption, while transparency in revenues and taxes should be increased.

Due to exchange rate fluctuations, central banks should implement regulations to increase their reserves to counter speculative pressures.

Socioeconomic measures should be implemented to counter volatility in energy prices, and energy expenses should be subsidized.

5.2. Long-Term Policy Recommendations

The results of the study indicate that oil prices negatively affect economic growth in the long term. To mitigate these negative economic challenges, diverse sectors in production should be supported, product diversification should be implemented, and excessive oil dependence should be reduced.

In the long term, carbon emissions should be reduced, and environmentally friendly energies and technologies should be encouraged. Renewable energy investments should be increased to alleviate the energy crisis in the economy.

Governmental oversight mechanisms and measures should be increased to prevent and control corruption. Improving institutional quality also plays a decisive role in both the increase in clean energy resources and the price of oil in production. From this perspective, establishing strong and meritocratic governance will support the proper enforcement of oversight and laws.

Oil revenues should be directed to human capital investments (education, healthcare, R&D) in the long term, creating the necessary environment for sustainable growth.

Fiscal discipline must be brought under control, and budget plans for a balanced and sustainable economy should be developed instead of budget expenditures based on oil revenues. Decision-makers should develop energy and economic policies that are both resilient to short-term revenue fluctuations and based on long-term sustainability.