Sage Journals: Discover world-class research

Abstract

The recent decade witnessed both steep ascending and descending trends in global oil prices. As a main energy resource, oil has been playing a substantial role in contemporary world’s economics. Hence, analyzing and understanding short- to long-term dynamics of oil prices is still one of challenging issues in the global economic- and energy-related debate topics. Although the literature is full of significant analytical studies that focus on particular issues related to the real oil prices, less can be found that integrates various factors affecting its dynamics. An integrated model, in which focal variables are included in main building blocks are illustrated with their interrelationships, helps policy makers to better understand the system. In this article, using a systems approach, a conceptual framework is developed to demonstrate various (economic and financial, technological, political, demographic, and industrial) factors that impact on the dynamics of the futures and spot prices with their interrelations. It is shown that unilateral (and univariate) analyzes is not sufficient in oil market dynamics analysis and systems approach should be applied. To do so, a subsystems diagram is developed based on literature review and analysis of oil market statistics. To validate the framework, windowed correlation analysis, Granger causality test, and regression analysis are utilized. Accordingly, a causal loop diagram is developed to describe interrelationships between main variables making the dynamics of oil prices. The framework provides practitioners with a foundation to conduct different related analyses. A quantified system dynamics model can be built based on this framework to simulate the market behavior and predict probable future trends and fluctuations.

Keywords

Oil market oil price systems approach subsystems diagram causal loop diagram regression analysis granger causality test

Introduction

Oil price is still the most important factor affecting world’s economic growth, industrial development, inflation and interest rate, food price, poverty, and other socio-economic variables, besides its dynamics is the main index demonstrating oil market circumstances (Álvarez et al., 2011; Jiménez-Rodríguez, 2008; Naranpanawa and Bandara, 2012; Narayan et al., 2014).

World’s oil market witnessed considerable and continuous oil price increments but a deep decrease during the end of 2014; oil price analysis is one of the most challenging energy related issues. There exist multiple factors influencing oil price dynamics, which makes oil market analysis so complicated whereas this complexity is increasing continuously (Huntington et al., 2012). Among those factors, the following ones are intensifying complexity of oil price analysis: changes in supply and demand side strategies, level of oil proved reserves, demand of oil substitutes, policies for oil strategic reserving, efficiency of transportation and industrial appliances, rate of economic growth in emerging and developed economies, financial market circumstances, and international political challenges. Consequently, numerous researchers tried to understand and investigate consequences of different factors on oil price dynamics as the main index of global energy market.

One research area focused on supply and demand side strategies in the market. A theoretical framework is developed based on the intersection of supply and demand curves and it is used to explain the reason of high level and oscillatory behavior of oil price (Stevens, 1995). Another research used the theory of exhaustible resources to estimate the equilibrium point of oil price and showed that oil price changes between 1974 and 1980 occurred as a response to supply-side changes (Roumasset et al., 1983). As well, it is argued that two co-integration relations are effective in oil price dynamics; the first one is Organization of the Petroleum Exporting Countries (OPEC) behavior based on its market power and the second one is the coverage rate of Organisation for Economic Co-operation and Development (OECD)- expected future demand using inventory behaviors (Chevillon and Rifflart, 2009). OPEC’s behavior analysis is done through an event analytical method based on available information about the future of global oil price in the market. Results are representative of the fact that there is no significant differences between the information and knowledge level of OPEC and Non-OPEC countries on oil price expectations (Lin and Tamvakis, 2010). One more research, using econometrics methods, investigated impact of OPEC announcements on expectation and volatility of daily oil price changes (Schmidbauer and Rösch, 2012). Moreover, the lifecycle of oil basins in supply side countries and significant suppliers’ production peak should be considered (Hosseini et al., 2014). Furthermore, the main reason of recent deep oil decreases could be found in supply side of the market wherein oil supply has not reacted to demand levels. On one side, main exporting countries tried to drive out rival producers (both new conventional and unconventional small but promising producers), and on the other hand, minor OPEC members and Non-OPEC producers tried to maintain and even increase, if possible, their market share through an oil export competition.

Another research area focused on relationships between oil prices and refineries’ performance (capacity production, demand, etc.). As an instance, it is showed that refinery sector had a significant role on oil price increases between 2004 and 2006 (Kaufmann et al., 2008).

The next category of literature is about effects of financial markets. The relationship between spot and futures prices is investigated for a monthly time period of 1990 to 1991 in United Kingdom and United States (Foster, 1996). In another research, the market’s data between 1990 and 2008 is applied to compare the capability of futures and spot prices in oil price discovery (Caporale et al., 2014). Moreover, it is tried to measure the effect of futures market on price discovery in spot market for crude oil benchmarks and the influence of changes in financial markets on oil futures market is investigated (Silvério and Szklo, 2012).

In the field of oil strategic storage (speculation), it is showed that there is a significant relationship between daily oil price oscillations and the level of oil strategic reserves (Cifarelli and Paladino, 2010).

Some researches focused on the response of oil substitutes demand to oil price changes. As an instance, using an econometrics approach, the relationship among alternative energy stock prices with technology stock prices, oil prices, and interest rates is analyzed and indicated that a shock in technology stock prices will result in a higher shock in alternative energy stock prices that have shocks on oil price (Henriques and Sadorsky, 2008).

In other researches in which some different factors have been investigated together, a research, using Vector Error Correction Model (VECM) and Directed Acyclic Graph (DAG), discussed the role of different factors on oil price dynamics including, near-term supply and demand, oil strategic reserves, exchange rate, commodity market, stock market, oil market, and US dollar index (Ji, 2012). Another research used Path Analysis and BVAR-TVP (Time varying parameter) model and concluded that followings with specific time-lags are influencing on oil price dynamics: changes in supply, level of oil reserves, oil consumption, and US dollar index (Chai et al., 2011). Moreover, the relationship between tanker freight rates and oil prices, using correlation coefficients and other tools, is investigated and showed that tanker freight rates and oil prices have high levels of correlation in medium to long term (Sun et al., 2014).

Although some of the aforementioned researches tried to investigate the dynamics of oil price in the global market more systematically (using vector regression), they almost ignored the structure and framework through which different factors are interrelating and used methods which are just based on the systems data; therefore, they do not have sufficient efficacy to model future trend of global oil price. However, systems approach focuses on the structure which makes the dynamics in a phenomenon and after showing this structure in a format of causal loop diagram, conducts quantitative analysis and validation consequently (Senge, 1997; Sterman, 2000). Furthermore, there exist different factors in form of a loop, which have considerable effect on oil price changes; these loops could not be modeled by unilateral relationships, instead, oil price dynamics should be considered in feedback structure over time, which considers various multilateral cause and effect relationships together.

In this regard, some researches applied systems approach to analyze oil market. Based on an expert panel, a conceptual framework developed to investigate OPEC’s power and stability, oil price dynamics, and investment opportunity of Non-OPEC countries. In this model, effects of market situation (based on practitioner’s decisions in the market) on producers’ decisions are studied however focus of the framework is on supply side of the market and other sectors were not studied completely (Morecroft and van der Heijden, 1992). Another research, critiquing behavioral simulation modeling approaches in related literature, which almost had modeled the supply side based on target capacity utilization, used a bottom-up approach to model demand endogenously although the model is too concise and suffers from several simplifications (i.e. an exogenous oil production rate) (Greenman, 1994). Regarding oscillatory behavior of oil price, a dynamic hypothesis (based on infrastructure development policies of oil importing and exporting countries) is developed and validated to investigate causes of the oscillations through a system dynamics modeling approach (Mashayekhi, 1998, 2001). Another research discussed effects of energy sector policies based on global economy’s trend, oil demand, oil substitutes, new oil discoveries, oil supply, and oil price (Samii and Teekasap, 2010).

In this research, it is not tried to carry out a complete review of the literature, instead, it is tried to determine the research gap through a selective but systematically review of the literature (Table 1).

Table 1.

Structure of literature review.

Row	Research area	Author(s)	Main methodology
1	Supply and demand side strategies in the market	Stevens (1995)	Economic analysis (intersection of supply and demand curves)
2		Roumasset et al. (1983)	The theory of exhaustible resources
3		Chevillon and Rifflart (2009)	Co-integration relations
4		Lin and Tamvakis (2010)	Event analytical method
5		Schmidbauer and Rösch (2012)	Econometrics methods
6	Relationships between oil prices and refineries’ performance	Kaufmann et al. (2008)	Co-integrating relationship
7	Effects of financial markets	Foster (1996)	Co-integration
8		Caporale et al. (2014)	Co-integration
9		Silvério and Szklo (2012)	Co-integration and Kalman filter technique
10	Oil strategic storage	Cifarelli and Paladino (2010)	Multivariate GARCH-M
11	Oil substitutes demand	Henriques and Sadorsky (2008)	Four variable vector auto-regression model
12	With different focus area	Ji (2012)	System analysis and VECM and the DAG method
13		Chai et al. (2011)	BVAR-TVP model
14		Sun et al. (2014)	Multi scale correlation, Ensemble Empirical Mode Decomposition
15		Morecroft and van der Heijden (1992)	System dynamics
16		Greenman (1994)	System dynamics
17, 18		Mashayekhi (1998, 2001)	System dynamics
19		Samii and Teekasap (2010)	System dynamics

VECM: vector error correction model; DAG: directed acyclic graph.

In all, the researches with econometrics and statistical approaches almost consider limited relationships, which do not present an integrated view of the system’s structure in which oil price dynamics is being formed. Furthermore, researches with systems approach have not developed a comprehensive framework consisting all influencing subsystems in oil price dynamics analysis; meanwhile, oil market dynamics analysis needs such a comprehensive framework (Fattouh, 2007) upon which further researches could be conducted more systematically. In this research, after an illustration of affecting factors on oil price dynamics through statistical tools, a subsystem diagram is developed and finally the detailed causal structure of interrelationships among main variables are explained in a causal loop diagram format.

The structure of the article is as follows: The next section describes research methodology and different techniques and tools used in this study. “Building blocks of oil price dynamics” section explains the dynamic hypothesis in building blocks of oil market and uses different tools (i.e. drawing trajectory of variables and their differences, Granger causality test, windowed correlation analysis, regression analysis, etc.) to validate the developed hypothesis. Consequently, a subsystem diagram is developed in “The generic conceptual framework of oil price dynamics (a subsystems diagram)” section, which clarifies the macro relationships shaping oil price dynamics. Finally, the last section explains detailed causal loops which interrelate different identified variables and makes structure of the system in which oil price dynamics occurs.

Methodology¹

Figure 1 shows research process in this article. At the first step, as reviewed in the previous section, related literature should be surveyed briefly in order that affecting factors on oil market dynamics be determined. In the next step, it is necessary to analyze market data² (about identified variables) to shape a dynamic hypothesis by which oil price dynamics could be explained. Herein, following supportive analysis should be performed:

Drawing and scrutinizing trajectory of main variables (and their difference values) versus oil price (and its difference values) in a multi-variable analysis format (Shone, 2002)

Calculating and drawing windowed correlation (Boker et al., 2002) among variables with mentioned relationship

Figure 1.

Research steps and tools.

At this step, some hypothesis about oil price dynamics could be formulated although to validate it, more statistical analysis based on the gathered data is required; therefore, following activities and analysis should be done:

Testing stationarity of data and making data series stationary³ (unit root and co-integration test (Ross, 2005))

Evaluating statistical causality between variable with different time lags (granger causality test (Tsay, 2005))

Evaluating explanatory property of identified causal relationships through regression analysis (Ross, 2005)

It should be mentioned that unconventional oil data (due to lack of adequate data) and political issues (being too qualitative) could not be involved in the above analysis.

The next two steps are to develop an integrated comprehensive framework using systems approach tools: subsystems diagram and causal loop diagram (Sterman, 2000).

Building blocks of oil price dynamics

In a general viewpoint, affecting factors on oil market dynamics could be categorized as follows:

Fundamental factors: These factors have long-term impact on market dynamics, which determines overall oil price trend, including oil supply and demand volume, volume and price of futures contracts, producer’s spare capacity, level of oil strategic reserves, rate of oil substitution (with coal, nuclear, renewables, unconventional oil, etc.), technological progress in oil discovery and recovery, etc. (Derakhshan, 2004).

Technical factors: These factors have short-term impact on market dynamics, which includes capacity of oil and oil-product storage in demand side (refineries and governmentally owned facilities), capacity of oil containers and tankers, other facility’s limitations, etc. (Lorié, 2013).

Political factors: These have short- to medium-term impact on market dynamics, including sanctions, wars, terrorist attacks to industry’s infrastructure, etc. (British Petrolum (BP), 2013a).

In fundamental factors, the demand side, regarding supply, oil price, and its economic situation, implements different short- to long-term reciprocal strategies especially oil substitution, energy intensity controls, and even temporary downturn in economic activities (Aldunate and Casassus, 2012; Aleklett, 2007; Cooper, 2005; International Energy Agency (IEA), 2008). On the other hand, it should be highlighted that some demand-side countries are Non-OPEC oil producers, determine their supply with regard to their demand and oil market’s situation. Meanwhile, oil spare capacity (in OPEC and Non-OPEC countries), oil strategic reserves in demand-side countries, and refineries storage level of oil are of high importance since they could damp or intensify market’s short-term oscillations (Andrews and Pirog, 2012; Bhanushali and Neema, 2007; Energy Security Partners LLC., 2013; Leiby et al., 2000; Morse, 2009; Singleton, 2013).

Another important factor is development rate of unconventional oil discovery and exploitation, which have considerable relative resources in the world in comparison with conventional oil resources (Kuuskraa et al., 2013). Besides, the main share of world’s unconventional oil resources is in demand-side countries especially in United States and China; in this regard, United Sates has implemented thorough development programs, which gradually affect its oil import rate as one of the largest oil consumers in the world (Maugeri, 2012).

It seems that if average oil price remains at a specific threshold, with the growth in futures market contracts and economically development in unconventional oil exploitation, OPEC power in the market will be weaken or even lost. On the other side, it should be noted that intentional decreases in oil price to put limitations on unconventional oil development is in contrast with utility of OPEC’s oil dependent economies (Brammer and Reuter, 2009; Morriss and Meiners, 2012). Furthermore, passing time, technological progress in unconventional oil and learning phenomenon would result in an economic competitive unconventional oil production which will control price manipulation power of OPEC (Brammer and Reuter, 2009).

In the following, a summarized and explanatory investigation of oil market dynamics is explained in four sections, namely, oil producers’ strategies, oil consumers’ strategies, unconventional oil development, financial market dynamics, and political issues.

Oil producers’ strategies

As a main supplier with 81 percent of world’s conventional oil reserves and 43 percent of world’s crude oil production, OPEC still plays a major role in the oil market (OPEC, 2012). OPEC's long-term strategy is to supply oil for market in a sustainable manner so that oil prices remain fair and stable to secure a regular supply for consumers and enhance the role of oil in meeting future energy demand, seeking to secure and enhance the collective interests of member countries (OPEC, 2010). Besides, whenever OPEC reduces oil supply to increase oil price, ceteris paribus, Non-OPEC will increase their supply to setback OPEC’s strategy (Aguiar-Conraria and Wen, 2012; Gately, 2011; Kohl, 2002).

OPEC’s spare capacity is also important since while spare capacity is tighten and economic condition is improving, oil price would ascend with more acceleration (Energy Security Partners LLC., 2013).

To conclude, it seems that the following items affect oil production rate of OPEC and Non-OPEC:

Oil production rate of OPEC and Non-OPEC (Y)

OPEC’s spare capacity (SC)

World’s Gross Domestic Production (GDP)

World’s proved reserves of oil (PR)

Oil real price (P)

OECD’s strategic oil reserves (SR)

Following sections “Unit root test on data,” “Granger causality tests for hypothetical causal relationships,” and “Regression analysis of oil production” tried to investigate above proposition statistically to conclude up with a reliable statement about effective factors on oil producers’ strategies in oil market.

Unit root test⁴ on data

Before conducting statistical analysis, unit root test is done to test stationarity of data and avoid erroneous relationship in Granger causality test and regression analysis. Accordingly, stationarity is gained by replacing the first-order difference of data; Table 2 shows the test results for the original data (level values) and the first-order differences (change rates).

Table 2.

Unit root test results for first-order difference of data.

Variable	T-test for level	T-test for first difference
Y^OPEC	−2.9266	−4.6464
Y^Non-OPEC	−1.1665	−4.6441
GDP	−3.1159	−3.8134
Price	−2.2315	−5.1776
PR^OPEC	0.2998	−4.0572
PR^OECD	−2.1128	−4.1463
SR	−1.2849	−19.595
SC	−2.9374	−5.2820

OPEC: Organization of the Petroleum Exporting Countries; GDP: Gross Domestic Production; PR: proved reserves of oil; SR: strategic oil reserves; SC: spare capacity.

Therefore, the first-order difference of data is used for Granger causality test and regression analysis consequently.⁵

Granger causality tests for hypothetical causal relationships.⁶

Based on the identified hypothetical causal relationships, Table 6 ⁷ shows the statistical causality test for the variables mentioned above and determines significant causal relationships.

Results imply that there exists a mutual relationship among OPEC, Non-OPEC oil production rates, and other variables (especially among independent variables affecting OPEC and Non-OPEC production rate), which are disregarded in Table 6 and should be considered in the final integrated framework.

Regression analysis of oil production

To perform a regression analysis, following variables are selected based on Table 6 although results were not significant based on T-stat and consequently another variable set is regarded, so that the resulted regression is more reliable as shown in Table 3. This implies that unilateral (and univariate) approaches for oil market dynamics analysis are not sufficient.

OPEC oil production:

${PR}_{t - 1}^{OECD}$

${PR}_{t - 2}^{OECD}$

${PR}_{t - 2}^{OPEC}$

${PR}_{t - 3}^{OPEC}$

$SC t - 1$

Non-OPEC oil production:

${PR}_{t - 1}^{OECD}$

P( $t - 1$ )

$GDP t - 2$

${PR}_{t - 3}^{OPEC}$

Table 3 shows that aforementioned propositions are valid; Figure 2 shows results of regression analysis of OPEC and Non-OPEC oil production rate and their first-order difference. In depicting oil production rates, following formula is applied.

Y_{t}^{OPEC} = \sum_{t 0}^{t} (α 1 {Δ SC}_{t}^{OPEC} + α 2 {Δ PR}_{t}^{OECD} + α 3 Δ SR t + α 4 Δ GDP t) + Y_{t 0}^{OPEC}

Y_{t}^{Non - OPEC} = \sum_{t 0}^{t} (α 1 {Δ Y}_{t - 1}^{OPEC} + α 2 Δ GDP t + α 3 Δ P t) + Y_{t 0}^{Non - OPEC}

Table 3.

Parameter values and related statistics for oil producers’ regression analysis.

OPEC	${Δ Y}_{t}^{OPEC} = α 1 {Δ SC}_{t}^{OPEC} + α 2 {Δ PR}_{t}^{OECD} + α 3 Δ SR t + α 4 Δ GDP t$
Non-OPEC	${Δ Y}_{t}^{Non - OPEC} = α 1 {Δ Y}_{t - 1}^{OPEC} + α 2 Δ GDP t + α 3 Δ P t$
Parameters value	$α 1$	$α 2$	$α 3$	$α 4$	R
OPEC	−0.86	−12.86	2.70	0.40	0.897
t-stat	−6.206	−2.770	2.571	3.025	0.897
Non-OPEC	−0.15	0.5607	−51.3054	**	0.808
t-stat	−2.1231	5.4594	−3.8506	**	0.808

OPEC: Organization of the Petroleum Exporting Countries; GDP: Gross Domestic Production; PR: proved reserves of oil; SR: strategic oil reserves; SC: spare capacity.

Figure 2.

OPEC and Non-OPEC oil production and its first-order difference—historical data vs. regression results.

Co-integration test

To test reliable explanatory property of resulted regressions, co-integration test is used. Since residuals should not have significant trend or intercept, statistical hypothesis testing is implemented for θ in

Δ Y t = θ Δ Y t - 1 + u t

. As shown in Table 4, both regressions have explanatory property and consequently they are valid.

Table 4.

T-stats for co-integration test of oil production regressions.

Regressions	T-test value
Reg. for OPEC production	−4.3471
Reg. for Non-OPEC Production	−2.8389

OPEC: Organization of the Petroleum Exporting Countries.

To sum up, the resulted remarks from Granger test and regression analysis reveals that the structure of relationships, which is realized with regards to historical data of the system, is not unique. To conduct numerical analysis, a conceptual integrated framework is necessary in which, not only cause and effect relationships between independent variables are seen but also bilateral relationships between dependent and independent variables could be studied in an integrated feedback structure. This could be achieved by systems approach in which the problem is investigated in a system of effective variables.

In the following, sections “Oil consumers’ strategies”, “Oil market players’ strategies to raise oil contracts in financial markets”, “Oil market players’ strategies to develop exploitation of unconventional oil”, and “Effect of political issues in the market” present similar analysis in summarized form for the other mentioned areas.

Oil consumers’ strategies

Oil consumption trends significantly depend on current and future outlook of world’s economic situation (U.S. Energy Information Administration (EIA), 2014); i.e. after 2008 economic crisis, oil consumption increased consequently all over the world (European Commission, 2009).

Among demand-side players’ medium- to long-term strategies, facing oil price increases, energy intensity control is one of the most effective ones in which technological improvement and market regulatory policies occur (BP, 2013b; Liddle, 2012; Matheny, 2010). Energy intensity control happens gradually due to considerable requisite time for technological progress and change (Fisher-Vanden et al., 2004). According to selected countries in EIA reports, Russia has reacted to oil price increases severely and has decreased 60% of its energy intensity from 1992 to 2011 (Karghiev, 2006); Moreover, China has reduced the index by 45% in the period of time (Long-term Incentive Strategies for Energy Efficiency, 2008) which affected world oil demand considerably (U.S. Energy Information Administration (EIA), 2013).

Another strategy is to substitute oil by other energy carriers (i.e. coal, natural gas, renewables, etc.) whenever oil price and its trend is not desirable regarding consumer’s economic programs and supply condition (Benes et al., 2015; Clark, 2013; European Commission, 2010; Kumhof and Muir, 2012).

The next action in demand side is to store oil and its derivatives strategically or in refineries (as the main crude oil consumers) so that they can adjust their activity level in refineries to minimize losses due to oil price undesirable oscillations.

To conclude, it seems that the following items influence oil demand rate in OECD and Non-OECD countries:

Gross Domestic Production (GDP)

Energy intensity (EI)

Oil substitutes consumption (OE)

Petroleum stocks in refineries (PS)

Oil real price (P)

Petroleum products stocks in refineries (PPS)

Refinery capacity utilization (CU)

As follows, sections “Granger causality tests for hypothetical causal relationships” and “Regression analysis of oil consumption” tried to investigate above proposition statistically to conclude up with a reliable statement about effective factors on oil consumers’ strategies in oil market.

Granger causality tests for hypothetical causal relationships

Similar to section “Unit root test on data,” Unit root test on data are performed and first-order difference of data is selected with valid stationarity. Table 6 shows significant causal relationships among the variables mentioned above.

Regression analysis of oil consumption

Similar to section “Regression analysis of oil production,” different variable set (which is not the same with what granger test concludes) is considered so that the resulted regression is reliable, which is shown in Table 5.

Table 5.

Parameter values and related statistics for oil consumers’ regression analysis.

OECD		${Δ D}_{t}^{OECD} = α 1 {Δ GDP}_{t}^{OECD} + α 2 {Δ EI}_{t}^{OECD} + α 3 {Δ OE}_{t}^{OECD} + α 4 Δ P t$
Non-OECD		${Δ D}_{t}^{Non - OECD} = α 1 {Δ GDP}_{t}^{Non - OECD} + α 2 {Δ OE}_{t}^{Non - OECD} + α 3 Δ P t$
Parameters		$α 1$	$α 2$	$α 3$	$α 4$	R
OECD	Value	2.99	8.34	−317.08	−20.79	0.923
OECD	t-stat	6.30	5.60	−3.20	−1.42	0.923
Non-OECD	Value	2.01	57.81	−51.92	**	0.933
Non-OECD	t-stat	2.9	1.19	−3.38	**	0.933

Table 6.

Granger causality test results for main variable in oil production, demand, and financial market.

Row	Lags	Cause var.	Affected var.	Null hypotheses	P-value
Results for main variable in OECD demand sector
1	1	ΔD^OECD	ΔP_t	A = 0	0.2176	−
1		ΔD^OECD	ΔP_t	B = 0	0.001	✓
2		ΔD^OECD	ΔGDP^OECD	A = 0	0.0096	✓
2		ΔD^OECD	ΔGDP^OECD	B = 0	0.0129	✓
3		ΔD^OECD	ΔOE_t	A = 0	0.0342	✓
3		ΔD^OECD	ΔOE_t	B = 0	0.0895	−
4	2	ΔD^OECD	ΔGDP^OECD	A = 0	0.0333	✓
4		ΔD^OECD	ΔGDP^OECD	B = 0	0.0055	✓
5		ΔD^OECD	ΔOE_t	A = 0	0.0182	✓
5		ΔD^OECD	ΔOE_t	B = 0	0.0892	✓
6		ΔD^OECD	ΔP_t	A = 0	0.0351	✓
6		ΔD^OECD	ΔP_t	B = 0	0.0018	✓
7	3	ΔD^OECD	ΔGDP^OECD	A = 0	0.0358	✓
7		ΔD^OECD	ΔGDP^OECD	B = 0	0.0297	✓
8		ΔD^OECD	ΔOE_t	A = 0	0.0068	✓
8		ΔD^OECD	ΔOE_t	B = 0	0.333	−
9		ΔD^OECD	ΔP_t	A = 0	0.0623	−
9		ΔD^OECD	ΔP_t	B = 0	0.0211	✓
Results for main variable in Non-OECD demand sector
10	1	ΔD^Non^-OECD	ΔGDP^Non^-OECD	A = 0	0.0042	✓
10		ΔD^Non^-OECD	ΔGDP^Non^-OECD	B = 0	0.6819	−
11		ΔD^Non^-OECD	ΔOE_t	A = 0	0.0121	✓
11		ΔD^Non^-OECD	ΔOE_t	B = 0	0.0808	−
12		ΔD^Non^-OECD	ΔP_t	A = 0	0.0105	✓
12		ΔD^Non^-OECD	ΔP_t	B = 0	0.7056	−
13	2	ΔD^Non^-OECD	ΔGDP^Non^-OECD	A = 0	0.0191	✓
13		ΔD^Non^-OECD	ΔGDP^Non^-OECD	B = 0	0.3696	−
14		ΔD^Non^-OECD	ΔOE_t	A = 0	0.0339	✓
14		ΔD^Non^-OECD	ΔOE_t	B = 0	0.129	−
15		ΔD^Non^-OECD	ΔP_t	A = 0	0.0148	✓
15		ΔD^Non^-OECD	ΔP_t	B = 0	0.8572	−
16	1	ΔY^OPEC	ΔPR^OECD	A = 0	0.0927	−
16		ΔY^OPEC	ΔPR^OECD	B = 0	0.0314	✓
17		ΔY^OPEC	ΔSC_t	A = 0	0.0642	−
17		ΔY^OPEC	ΔSC_t	B = 0	0.4024	−
18	2	ΔY^OPEC	ΔPR^OECD	A = 0	0.0686	−
18		ΔY^OPEC	ΔPR^OECD	B = 0	0.0316	✓
19		ΔY^OPEC	ΔPR^OPEC	A = 0	0.0055	✓
19		ΔY^OPEC	ΔPR^OPEC	B = 0	0.7465	−
Results for main variable in Non-OPEC production sector
20	3	ΔY^OPEC	ΔPR^OPEC	A = 0	0.0294	✓
20	3	ΔY^OPEC	ΔPR^OPEC	B = 0	0.7998	−
21	1	ΔY^Non^-OPEC	ΔPR^OECD	A = 0	0.2814	−
21		ΔY^Non^-OPEC	ΔPR^OECD	B = 0	0.0035	✓
22		ΔY^Non^-OPEC	ΔP_t	A = 0	0.3924	−
22		ΔY^Non^-OPEC	ΔP_t	B = 0	0.0219	✓
23	2	ΔY^Non^-OPEC	ΔGDP_t	A = 0	0.8496	−
23	2	ΔY^Non^-OPEC	ΔGDP_t	B = 0	0.0232	✓
24	3	ΔY^Non^-OPEC	ΔPR^OPEC	A = 0	0.0433	✓
24	3	ΔY^Non^-OPEC	ΔPR^OPEC	B = 0	0.0317	✓
Results for main variable in Financial Market
25	1	ΔFC	ΔFP	A = 0	0.0042	✓
25		ΔFC	ΔFP	B = 0	0.8422	−
26		ΔFC	ΔSP	A = 0	0.0101	✓
26		ΔFC	ΔSP	B = 0	0.8124	−
27	2	ΔFC	ΔFP	A = 0	0.0376	✓
27		ΔFC	ΔFP	B = 0	0.373	−
28		ΔFC	ΔSP	A = 0	0.0265	✓
28		ΔFC	ΔSP	B = 0	0.6761	−

OPEC: Organization of the Petroleum Exporting Countries; SP: spot price; GDP: Gross Domestic Production; PR: proved reserves of oil.

Table 5 satisfactorily justifies the aforementioned propositions and Figure 3 illustrates results of the regression analysis of OECD and Non-OECD oil demand and its first-order difference. It is necessary to mention that co-integration test is also used here and showed that both regressions have explanatory property and therefore they are valid.

Figure 3.

OECD and Non-OECD oil demand and its first-order difference—historical data vs. regression results (1000 barrels per day).

Oil market players’ strategies to raise oil contracts in financial markets

In recent years, oil contracts in financial markets (options, futures, and other derivatives) have grown noticeably; while maximum oil consumption in 2012 was about 100 million barrels per day, maximum of oil futures contracts reached to 1.5 million contract (and each contract equals to 1000 barrels per day) (U.S. Energy Information Administration (EIA), 2013). On one side, it should be noted that lack of oil in the market (consumption-production) is one of the main stimulus of oil contracts growth in the financial markets. On the other side, since OPEC does not have a major contribution in financial markets, price controlling power would be by Non-OPEC players of the market (i.e. in United States and Europe) in the future (Fattouh, 2011).

Moreover, since oil spot price oscillations is one another important motivation for market players to contribute additionally in financial markets to damp undesired changes, futures price has a more smooth trend in contrast to spot price (U.S. Energy Information Administration (EIA), 2013), which indicates ambition of market players to hedge contract risks in OTC that itself is a stimulus for the growth in financial markets (Hull, 2009).

Despite the fact that the major share of oil demand is met in the OTC market and therefore spot prices should be the major input for future decisions, futures prices play the main role (Bernard, 2013; Kuuskraa et al., 1985).

In case that more and more volume of oil demand is met in futures markets, oscillations of spot price would be partially controlled which makes a reliable and safe market for demand side; this will gradually result in a growth in futures contracts, which is determined by a more predictable future price.

In conclusion, it seems that the following are influencing oil futures contracts:

Changes in real oil spot price (ΔSP)

Difference between oil demand and supply (Y-D)

Oil futures price (FP)

Sections “Granger causality tests for hypothetical causal relationships” and “Regression analysis of oil futures contracts” investigate above proposition statistically to come up with a reliable statement about effective factors on oil futures contracts.

Granger causality tests for hypothetical causal relationships

Similar to section “Granger causality tests for hypothetical causal relationships,” Unit root test on data is performed and first-order difference of data is selected with valid stationarity. Table 6 shows significant causal relationships between the variables mentioned above.

Regression analysis of oil futures contracts

Similar to section “Regression analysis of oil consumption,” different variable set (which is not the same with what granger test concludes) is considered so that the resulted regression is reliable, which is shown in Table 7.

Table 7.

Parameter values and related statistics for oil futures contracts regression analysis.

World	$Δ FC t = α 1 Δ (y t - D t) + α 2 Δ FP t$
Parameters	$α 1$	$α 2$	R
Value	0.104	6.212	0.837
T-stat	4.348	3.410	0.837

Figure 4 shows the results of regression analysis of oil futures contracts and its first-order difference. It should be mentioned that co-integration test is also applied here and showed that both regressions have explanatory property and therefore they are valid.

Figure 4.

Oil futures contracts and its first-order difference—historical data vs. regression results.

Oil market players’ strategies to develop exploitation of unconventional oil

Recent estimation of EIA in 2013 about world’s unconventional oil technically recoverable resources is 345 billion barrels which approximately equals 10 percent of total oil resources (conventional and unconventional) of world (Kuuskraa et al., 2013). The important point is that United States and China, as largest oil consumers, have main unconventional resources among others. In this regard, United States and Canada are developing exploitation of unconventional oil more rapidly and effectively with a promising trend (Bernard, 2013; Kuuskraa et al., 1985). It is argued that unconventional oil production for United States is economic with a WTI oil price between 50 and 65 dollar (Maugeri, 2012); this increases US power in controlling oil price in a defined interval. Unconventional oil production in United States, especially in North Dakota and Texas, has a growing trend so that 847,000 barrels per day of oil production increase of United States in 2012 was due to unconventional oil production, i.e. 29 percent of US oil production belonged to unconventional oil. In this year, the production rate of Canada reached to 291,489 barrels per day (Kuuskraa et al., 2013).

Hence, it could be concluded that demand side would consider developing unconventional oil as a long-term strategy, so that they could limit OPEC’s pricing power and provide a desired economic situation.

On the other side, technological progress in unconventional oil first makes unconventional oil production more economic, although it can enhance recovery factor of conventional oil, from its current average value of 35%, as well (Maugeri, 2012); thus, this bilateral relationship should be considered too.

Effect of political issues in the market

There is no doubt that political issues have substantial influence on oil market (especially in short to medium term). Global oil market has encountered structural changes in last decades. In 1960s, oil industry was controlled by large multinational companies. In 1970s, OPEC played a major role in the market. Arab-Israeli War in 1973, Iranian revolution in 1997, and Iran-Iraq war in 1981, imposed many oscillations to oil price in the market. In 1990s, oil market was almost stable politically except when Iraq invaded Kuwait, which imposed a short run increase in prices. Invasion of Iraq in 2000s and recently political changes in Arab countries put unplanned oscillations in the market (BP, 2013a; Derakhshan, 2005). Therefore, political issues (revolutions, wars, etc.) should be considered in the framework by which oil price dynamics would be studied.

The generic conceptual framework of oil price dynamics (a subsystems diagram)

As mentioned above, different factors affecting oil price changes, which make oil price analysis complicated; on the other hand, statistical analysis in previous sections showed that unilateral (and univariate) approaches for oil market dynamics analysis are not suitable. Therefore, it is suggested to apply a systems approach. Here, at least a generic conceptual framework is necessary and useful. This section provides a subsystems diagram of oil price dynamics building blocks, which has concluded from review of literature and reports issued by international energy related organizations and results of “Building blocks of oil price dynamics” section (Figure 5).

Figure 5.

Building blocks of oil price dynamics—subsystems diagram.

Oil production subsystem is considered in two sections, i.e. conventional and unconventional in which following variables are determined: production rate, production capacity, capacity expansion economic profitability, production and capacity expansion unit costs, proved reserves, discovery rate, etc. In production technology subsystem recovery factor for conventional and unconventional oil and their interrelationships are determined.

Oil contracts, mainly, in OTC and spot market, futures and options market, are considered in oil contracts subsystem in which following variables are determined: supply rate and demand rate of each contract type (from oil and non-oil participants), price of oil in each contract type, purchase and buy strategies of oil and non-oil participants, etc.

In demand subsystem, oil and its substitutes demand are determined; moreover, following variables are determined: world’s economic activities, energy intensity, refinery capacity, refinery oil stocks, etc.

The detailed conceptual framework of oil price dynamics (a causal loop diagram)

Based on the developed subsystem diagram of oil price dynamics in “The generic conceptual framework of oil price dynamics (a subsystems diagram)” section, in this section, a more detailed structure of mechanisms, by which oil price changes over time, is described through a causal loop diagram.

Figure 6 shows main influencing factors on oil spot price dynamics, oil futures price (for 1- to 12-month delivery), and expectations of oil spot price.

Figure 6.

Main influencing factors on oil spot price (1- to 12-month delivery is shown by “…”).

Supply and demand imbalance, oil spot price in previous periods of time, expectations of oil spot price in one and next two forthcoming months, and market environment stability are influencing the oil spot price (Hamilton, 2008). Factors forming oil futures price for 1- to 12-month delivery, include, supply and demand for futures contracts, trend of oil spot price, futures price in previous time period, and factors which determine desirability of futures contracts (opportunity costs: oil inventory and storage costs, oil transportation costs, and interest rate).

Also, spot price trend, futures price for delivery in the time period of expectation, and futures expectation weight (relative share of oil supplied by futures market) are of high importance in forming the expectations of oil spot price.

Figure 7 depicts two reinforcing and balancing loops which affect conventional and unconventional oil production and capacity expansion rates. In the reinforcing loop, an increase in production rate would lead to an increase in revenues and consequently an increase in investments in capacity expansion regarding normal field development delay time; therefore, in case that demand is growing, production rate would increase. In the balancing loop, an increase in production rate, regarding limited oil resources, all else being equal, decreases reserve to production ratio (R/P). The decrease mentioned above accordingly results in a decrease in long term expected profitability of investment in oil industry which finally constraints capacity and production enhancement programs.

Figure 7.

Reinforcing and balancing loops of oil production and capacity expansion.

Figure 8 shows two reinforcing loops in which capacity expansion unit costs reduction (because of learning curve effect during the time) and production technology (which here is indexed by recovery factor) development are explained.

Figure 8.

Reinforcing loops for learning effect in capacity expansion and production technology development.

Figure 9 illustrates the structure which results in confronting OPEC and Non-OPEC strategies. OPEC and privately owned Non-OPEC producers adopt their strategy almost in a same manner since their ultimate goal is to gain the maximum economic added value from oil market; i.e. an increase in oil price, ceteris paribus, would result in more oil production to gain more revenue with new high prices. This, all else being equal, brings about oil surplus and overproduction in the market and ultimately closes the loop by reducing the oil price. The fact that should be considered, is oil strategic storage by Non-OPEC countries so that they can damp undesired fluctuations of oil price. Hence, an increase in spot price, if it considerably is far from targeted oil price which is came from long term spot price trend, causes an increase in oil supply from strategic reserves by Non-OPEC countries to control their market. In contrast, a decrease in spot price brings on more strategic oil storage in demand side of the oil market.

Figure 9.

Effective factors on OPEC and Non-OPEC strategies.

Figure 10 shows loops related to production and capacity dynamics of refineries as the main consumers of crude oil. Here, the main point is that refineries production and capacity expansion rates are based on short term and long term expected profitability which is calculated by means of expectations of oil spot prices. Figure 11 illustrates the structure by which fluctuations of futures prices for 1- to 12-month delivery affect storage level of petroleum and its products. Here, the key variable is the difference between futures price for 1- and 12-month delivery. An increase in this value, which means market players’ estimation of prices increase in the future, ceteris paribus, results in an increase in short term demand of oil which brings on oil lack in the market. This, all else being equal, motivates players to buy futures oil for 1- to 6-month delivery which causes an increase in price of these contracts. The increase closes the loop by decreasing the difference between futures price for 1- and 12-month delivery.

Figure 10.

Reinforcing and balancing loops for production rate and capacity expansion of refineries.

Figure 11.

Balancing loop for effect of futures prices dynamics on the storage of oil and its products in refineries.

Figure 12 shows that how increase or decrease in oil demand affects volume of oil contracts in financial markets. An increase in oil demand, all else being equal, would result in oil lack in the market, which motivates oil demand side players to buy more futures contracts which finally increases futures prices. This causes an increase in spot price and its future estimation, which motivates demand side to reduce their oil demand by promoting energy intensity control programs which brings on lower demand in unit economic activity after a delay time. In contrast, a decrease in oil demand, ceteris paribus, results in oil surplus in the market which finally stimulates demand by lower prices.

Figure 12.

Balancing loops for relationships between demand and price with the volume of futures contracts (1- to 12-month delivery is shown by “…”).

Figure 13 indicates to the structure through which speculators in financial market affect oil price. A raise in oil futures prices brings on a raise in estimations of spot price in the future, which increases motivation of speculators to buy more futures contracts; on one hand, this itself results in an increase in futures price, and on the other hand, by decreasing motivation of speculators for selling contracts, brings on futures price increment. Therefore, speculation can intensify oil increases or decreases direction which is called price bubble.

Figure 13.

Effect of speculation in financial markets on oil price (1- to 12-month delivery is shown by “…”).

Figure 14 shows factors determining oil demand in residential, transportation, commercial, and industrial sectors. Here, the main point is relative price of petroleum products vs. substitutes. Finally, based on petroleum products demand and refineries technology to produce a specific mix of different petroleum products from crude oil, oil demand is calculated.

Figure 14.

Loops for interrelationship between oil demand and oil price.

The main inputs for petroleum products demand in residential sector are welfare (here is considered by GDP) and population; the demand in transportation sector could be generally calculated by GDP, Population, and IT infrastructure; the main input of the demand in commercial and industrial sectors is GDP.

Conclusion

A concise review of literature about oil market dynamics showed that there is lack of an integrated framework for the system by which not only building block of influencing factors on oil price dynamics is explained but also causal loops and feedback structure of relationships among main variables in the market is illustrated. In this article, factors influencing on oil price dynamics are discovered through literature review and statistical analysis of oil markets data. In the next step, based on a multivariate analysis of oil price dynamics, hypothesizes are shaped and validated through a Granger causality test and regression analysis. In this analysis, unit root and co-integration test are utilized to validate the stationarity of the data and explanatory property of regression analysis results. It is shown that local analysis of the entire system with complex interrelations cannot explain the phenomena of price fluctuations well and an integrated model is missing.

In the next step, a generic conceptual framework of oil price dynamics is developed, so that the subsystems which make market dynamics are identified and their interrelationships are determined. Accordingly, a detailed conceptual framework of oil price dynamics is developed where feedback structures leading to fluctuations in the oil market are illustrated.

The developed two-stage framework can be used to conduct further quantitative research considering feedback structures in the market. Moreover, it facilitates qualitative analysis about new situations in the market, i.e. current downturn in international oil price. However, other factors such as export policies of both OPEC and Non-OPEC countries should be explicitly involved in the model.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

1% level	−4.4983
5% level	−3.6584
10% level	−3.2690

Notes

References

Aguiar-Conraria

Wen

(2012) OPEC’s oil exporting strategy and macroeconomic (in) stability. Energy Economics 34(1): 132–136.

Aldunate

Casassus

(2012) Consumption and hedging in oil importing developing countries. European Financial Management 18(5): 896–928.

Aleklett

(2007) Peak Oil and the Evolving Strategies of Oil Importing and Exporting Countries-Facing the hard truth about an import decline for the OECD countries, Uppsala, Sweden: Joint Transport Research Centre, Uppsala University, pp. 63.

Álvarez

Hurtado

Sánchez

(2011) The impact of oil price changes on Spanish and euro area consumer price inflation. Economic Modelling 28(1): 422–431.

Andrews

Pirog

(2012) The Strategic Petroleum Reserve: Authorization, Operation, and Drawdown Policy. CRS Report R42460. Washington, DC: Library of Congress, Congressional Research Service, p.19.

Benes

Chauvet

Kamenik

(2015) The future of oil : Geology versus technology. International Journal of Forecasting 31(1): 207–221.

Bernard P (2013) Oil shale Rush in North Dakota. Le Monde.fr newspaper. Available at: http://www.lemonde.fr/ameriques/visuel/2013/05/12/ruee-vers-le-gaz-de-schiste-dans-le-dakota-du-nord_3175426_3222.html (accessed 27 January 2016).

Bhanushali T and Neema S (2007) Crude Oil: Prices on a firm footing. India Infoline Ltd. Available at: http://www.indiainfoline.com/article/general-na/crude-oil-prices-on-a-firm-footing-113111400519_1.html (accessed 27 January 2016).

Boker

Rotondo

(2002) Windowed cross-correlation and peak picking for the analysis of variability in the association between behavioral time series. Psychological Methods 7(3): 338–355.

10.

Brammer

Reuter

(2009) The Viability of Non-Conventional Oil Development, USA: Innovest Strategic Value Advisor.

11.

British Petrolum (BP) (2013a) BP Statistical Review of World Energy. London, UK: BP p.l.c.

12.

British Petrolum (BP) (2013b) Energy Outlook 2030. London, UK: BP p.l.c.

13.

Caporale

Ciferri

Girardi

(2014) Time-varying spot and futures oil prices dynamics. Scottish Journal of Political Economy 61(1): 78–97.

14.

Chai

Guo

Meng

(2011) Exploring the core factors and its dynamic effects on oil price: An application on path analysis and BVAR-TVP model. Energy Policy 39(12): 8022–8036.

15.

Chevillon

Rifflart

(2009) Physical market determinants of the price of crude oil and the market premium. Energy Economics 31(4): 537–549.

16.

Cifarelli

Paladino

(2010) Oil price dynamics and speculation—A multivariate financial approach. Energy Economics 32(2): 363–372.

17.

Clark

(2013) Energy Market Review: A Story of U.S. Supply and EM Demand, USA: Roubini Global Economic LLC, pp. 213.

18.

Cooper

(2005) A World Agreement on Oil?, South Africa: Centre for International Governance Innovation (CIGI).

19.

Derakhshan M (2005) Derivatives and Risk Management in Oil Markets. Tehran, Iran: Institute for International Energy Studies (IIES).

20.

Energy Security Partners LLC (2013) Crude Oil and Petroleum Product Price Outlook, UK: WTRG Economic.

21.

European Commission (2009) Addressing the Impact of the Financial Crisis on EU-Russia Energy Cooperation. HTSPE Limited.

22.

European Commission (2010) Impact of the Use of Biofuels on Oil Refining and Fuels Specifications. Wood Mackenzie & Ricardo.

23.

Fattouh

(2007) OPEC Pricing Power: The Need for a New Perspective, Oxford, England: Oxford Institute for Energy Studies.

24.

Fattouh

(2011) An Anatomy of the Crude Oil Pricing System, Oxford, England: Oxford Institute for Energy Studies, pp. 187.

25.

Fisher-Vanden

Jefferson

Liu

(2004) What is driving China’s decline in energy intensity? Resource and Energy Economics 26(1): 77–97.

26.

Foster

(1996) Price discovery in oil markets: a time varying analysis of the 1990-1991 Gulf conflict. Energy Economics 18(3): 231–246.

27.

Gately

(2011) OPEC at 50: Looking Back and Looking Ahead, Oklahoma, USA: National Energy Policy Institute, pp. 238.

28.

Greenman JV (1994) The price of oil: A system dynamic approach. In: Monaghan C and Wolstenholme E (eds) 12th International System Dynamics Conference. Sterling, Scotland: System Dynamics Society.

29.

Hamilton JD (2008) Understanding crude oil prices. University of California Energy Institute. UC Berkeley. Available at: http://escholarship.org/uc/item/3fg2r29s (accessed 27 January 2016).

30.

Henriques

Sadorsky

(2008) Oil prices and the stock prices of alternative energy companies. Energy Economics 30(3): 998–1010.

31.

Hosseini

Ganjavi

Kiani

(2014) Examination of Iran’s crude oil production peak and evaluating the consequences: a system dynamics approach. Energy, Exploration & Exploitation 32(4): 673–690.

32.

Hull

(2009) Options, Futures and Other Derivatives, 7th ed. New Jersey, USA: Pearson Education International.

33.

Huntington H, Al-Fattah SM, Huang Z, et al. (2012) Oil Price Drivers and Movements: The Challenge for Future Research. Available at SSRN 2257675.

34.

International Energy Agency (IEA) (2008) Medium-Term Oil Market Report, Paris, France.

35.

(2012) System analysis approach for the identification of factors driving crude oil prices. Computers & Industrial Engineering 63(3): 615–625.

36.

Jiménez-Rodríguez

(2008) The impact of oil price shocks: Evidence from the industries of six OECD countries. Energy Economics 30(6): 3095–3108.

37.

Karghiev

(2006) Energy and Sustainable Development in the Russian Federation, Moscow, Russia: Helio International, pp. 26–27.

38.

Kaufmann

Dees

Gasteuil

(2008) Oil prices: The role of refinery utilization, futures markets and non-linearities. Energy Economics 30(5): 2609–2622.

39.

Kohl

(2002) OPEC behavior 1998-2001. The Quarterly Review of Economics and Finance 42(2): 209–233.

40.

Kumhof

Muir

(2012) Oil and the World Economy: Some Possible Futures. . IMF Working Paper No. 12/256. Available at: SSRN 2183018.

41.

Kuuskraa VA, Sedwick K and Yost AB (1985) Technically recoverable Devonian shale gas in Ohio, West Virginia, and Kentucky. In: SPE Eastern Regional Meeting, Society of Petroleum Engineers, West Virginia, USA, p.15.

42.

Kuuskraa VA, Stevens SH and Moodhe KD (2013) Technically Recoverable Shale Oil and Shale Gas Resources: An Assessment of 137 Shale Formations in 41 Countries Outside the United States. USA: Energy Information Administration (EIA).

43.

Leiby PN, Bowman D and UT-Battelle LLC (2000) The value of expanding the US strategic petroleum reserve. Report NO: ORNL/TM-2000/179, Oak Ridge National Laboratory, Tennessee, USA.

44.

Liddle

(2012) Breaks and trends in OECD countries’ energy–GDP ratios. Energy Policy 45: 502–509.

45.

Lin

Tamvakis

(2010) OPEC announcements and their effects on crude oil prices. Energy Policy 38(2): 1010–1016.

46.

Long-term Incentive Strategies for Energy Efficiency (2008) Interim Report of the Task Force on Economic Instruments for Energy Efficiency and the Environment in China, Annual General Meeting of CCICED, China.

47.

Lorié

(2013) Oil Market Outlook, The Netherlands: Atradius Credit Insurance N.V.

48.

Mashayekhi

(1998) Public finance, oil revenue expenditure and economic performance: a comparative study of four countries. System Dynamics Review 14(2): 189–219.

49.

Mashayekhi AN (2001) Dynamics of oil price in the world market. In: 19th International System Dynamics Conference, Atlanta, GA.

50.

Matheny

(2010) Reducing the Impact of Price Shocks in Energy-intensive Economies, USA: John F. Kennedy School of Government.

51.

Maugeri

(2012) Oil: The Next Revolution, The Unprecedented Upsurge of Oil Production Capacity and What It Means for the World, Harvard Kennedy School, USA: Belfer Center for Science and International Affairs.

52.

Morecroft

JDW

van der Heijden

KAJM

(1992) Modelling the oil producers—Capturing oil industry knowledge in a behavioural simulation model. European Journal of Operational Research 59(1): 102–122.

53.

Morriss AP and Meiners R.E. (2012) Competition in global oil markets: a meta-analysis and review. In: Securing America’s Future Energy Commissioned Research, USA.

54.

Morse

(2009) What We Have Learned About Oil Markets, UK: LCM Research, British Institute of Energy Economics.

55.

Naranpanawa

Bandara

(2012) Poverty and growth impacts of high oil prices: Evidence from Sri Lanka. Energy Policy 45: 102–111.

56.

Narayan

Sharma

Poon

(2014) Do oil prices predict economic growth? New global evidence. Energy Economics 41: 137–146.

57.

Organization of the Petroleum Exporting Countries (OPEC) (2010) OPEC Long-Term Strategy. Vienna, Austria: Organization of the Petroleum Exporting Countries.

58.

Organization of the Petroleum Exporting Countries (OPEC) (2012) Annual Statistical Bulletin. Vienna, Austria: Organization of the Petroleum Exporting Countries.

59.

Ross

(2005) Introductory Statistics. Elsevier Academic Press p. 230.

60.

Roumasset

Isaak

Fesharaki

(1983) Oil prices without OPEC: a walk on the supply-side. Energy Economics 5(3): 164–170.

61.

Samii

Teekasap

(2010) Energy policy and oil prices : system dynamics approach to modeling oil market. Journal of Global Commerce Research 2(3): 1–7.

62.

Schmidbauer

Rösch

(2012) OPEC news announcements: Effects on oil price expectation and volatility. Energy Economics 34(5): 1656–1663.

63.

Senge

(1997) The fifth discipline. Measuring Business Excellence 1(3): 46–51.

64.

Shone

(2002) Economic Dynamics, 2nd ed. New York, USA: Cambridge University Press, pp. 322.

65.

Silvério

Szklo

(2012) The effect of the financial sector on the evolution of oil prices: Analysis of the contribution of the futures market to the price discovery process in the WTI spot market. Energy Economics 34(6): 1799–1808.

66.

Singleton

(2013) Investor flows and the 2008 boom/bust in oil prices. Management Science 60(2): 300–318.

67.

Sterman

(2000) Business dynamics: Systems thinking and modeling for a complex world, Boston, USA: Irwin/McGraw-Hill, pp. 168.

68.

Stevens

(1995) The determination of oil prices 1945-95. Energy Policy 23(10): 861–870.

69.

Sun

Tang

Yang

(2014) Identifying the dynamic relationship between tanker freight rates and oil prices: In the perspective of multiscale relevance. Economic Modelling 42: 287–295.

70.

The World Bank (2013) World Development Indicators. Available at: http://data.worldbank.org.

71.

Tsay

(2005) Analysis of Financial Time Series, 3rd ed. New Jersey, USA: John Wiley & Sons, pp. 216.

72.

U.S. Energy Information Administration (EIA) (2013). Available at: http://www.eia.gov/petroleum/data.cfm.

73.

U.S. Energy Information Administration (EIA) (2014) What drives crude oil prices? Available at: www.eia.gov/finance/markets (accessed 27 January 2016).

A conceptual framework for the oil market dynamics: A systems approach

Abstract

Keywords

Introduction

Methodology 1

Building blocks of oil price dynamics

Oil producers’ strategies

Unit root test 4 on data

Granger causality tests for hypothetical causal relationships. 6

Regression analysis of oil production

Co-integration test

Oil consumers’ strategies

Granger causality tests for hypothetical causal relationships

Regression analysis of oil consumption

Oil market players’ strategies to raise oil contracts in financial markets

Granger causality tests for hypothetical causal relationships

Regression analysis of oil futures contracts

Oil market players’ strategies to develop exploitation of unconventional oil

Effect of political issues in the market

The generic conceptual framework of oil price dynamics (a subsystems diagram)

The detailed conceptual framework of oil price dynamics (a causal loop diagram)

Conclusion

Footnotes

Declaration of conflicting interests

Funding

Notes

References

Methodology¹

Unit root test⁴ on data

Granger causality tests for hypothetical causal relationships.⁶