University of Ghana  http://ugspace.ug.edu.gh
UNIVERSITY OF GHANA 
COLLEGE OF HUMANITIES 
 
 
INTRA- AND INTER-GROUP PERFORMANCE OF OIL PRODUCING 
COUNTRIES: A META- AND GLOBAL FRONTIER ANALYSIS 
 
 
BY 
VICTOR SOSU GAKPEY 
(10444001) 
 
THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA IN PARTIAL 
FULFILMENT OF THE REQUIREMENT FOR THE AWARD OF MPHIL 
OPERATIONS MANAGEMENT DEGREE 
 
DEPARTMENT OF OPERATIONS AND MANAGEMENT INFORMATION 
SYSTEMS 
 
 
JULY 2016 
i 
 
University of Ghana  http://ugspace.ug.edu.gh
DECLARATION 
I do hereby declare that this work is the result of my own research and has not been presented 
by anyone for any academic award in this or any other university. All references used in the 
work have been fully acknowledged. 
I bear sole responsibility for any shortcomings. 
 
 
 
………………………………………….   ………………………… 
      VICTOR SOSU GAKPEY        DATE 
                  (Student) 
 
 
 
 
 
 
 
  
i 
 
University of Ghana  http://ugspace.ug.edu.gh
CERTIFICATION 
I hereby certify that this thesis was supervised in accordance with procedures laid down by 
the University of Ghana. 
 
 
………………………………………  ……………………………….. 
DR. KWAKU OHENE-ASARE         DATE 
           (SUPERVISOR) 
 
 
 
 
 
………………………………………….  ………………………………….. 
DR. ANTHONY AFFUL-DADZIE         DATE 
       (CO-SUPERVISOR) 
 
  
ii 
 
University of Ghana  http://ugspace.ug.edu.gh
DEDICATION 
 
To My Father and Mother  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
iii 
 
University of Ghana  http://ugspace.ug.edu.gh
ACKNOWLEDGEMENT 
I owe gratitude to my supervisors, Dr Kwaku Ohene-Asare and Dr Anthony Afful-Dadzie, 
both of the Department of Operations and Management Information Systems of the 
University of Ghana Business School, for their guidance and support. Further appreciation 
goes to all other faculty members of the Department especially Dr Francis Yaw Banuro and 
other staff of the department for their support. 
Finally, my sincerest thanks goes to my family, siblings, colleagues and friends for their 
diverse support and undying willingness to support me. 
 
 
 
 
 
 
 
 
 
 
 
 
iv 
 
University of Ghana  http://ugspace.ug.edu.gh
TABLE OF CONTENTS 
DECLARATION .................................................................................................................. i 
CERTIFICATION .............................................................................................................. ii 
DEDICATION ................................................................................................................... iii 
ACKNOWLEDGEMENT ................................................................................................. iv 
TABLE OF CONTENTS .................................................................................................... v 
LIST OF TABLES .......................................................................................................... viii 
LIST OF FIGURES ............................................................................................................ x 
ABSTRACT ........................................................................................................................ xi 
 
CHAPTER ONE ................................................................................................................. 1 
INTRODUCTION ............................................................................................................... 1 
1.0  Background of the Study ........................................................................................ 1 
1.1  Statement of Problem ............................................................................................. 3 
1.2  Contributions of the Study ..................................................................................... 6 
1.3  Research Objectives ............................................................................................... 7 
1.4  Research Questions ................................................................................................ 8 
1.5  Limitations of the Study ......................................................................................... 8 
1.6  Thesis Structure ...................................................................................................... 9 
 
CHAPTER TWO .............................................................................................................. 10 
2.0  INTRODUCTION .............................................................................................. 11 
2.1  Theoretical Review ............................................................................................... 11 
2.1.1 Intra-Group Performance Theories ............................................................... 11 
2.1.2 Inter-Group Performance Theories ............................................................... 17 
2.2 Empirical Review ...................................................................................................... 24 
2.2.1 Frontier Efficiency in the Oil Industry ............................................................... 24 
2.2.2 Group Performance and Efficiency .................................................................... 27 
v 
 
University of Ghana  http://ugspace.ug.edu.gh
2.3 Conceptual Framework ............................................................................................. 31 
2.4  Conclusion ............................................................................................................ 32 
 
CHAPTER THREE .......................................................................................................... 33 
3.0 INTRODUCTION .............................................................................................. 33 
3.1 International Oil and Gas Supply and Demand .................................................... 34 
3.1.1 Proven Oil and Gas Reserves ........................................................................ 34 
3.1.2 Oil and Gas Supply ....................................................................................... 38 
3.1.2 Oil and Gas Consumption ............................................................................. 42 
3.2 Intergovernmental organizations in the oil industry ............................................. 44 
3.2.1 Organization of the Petroleum Exporting Countries (OPEC) ....................... 45 
3.2.2 Organization of Arab Petroleum Exporting Countries (OAPEC) ................. 47 
3.2.3 The Former Soviet Union (FSU)................................................................... 48 
3.2.4 The International Energy Agency (IEA) ....................................................... 49 
3.3 Conclusion ............................................................................................................ 51 
 
CHAPTER FOUR ............................................................................................................. 52 
METHODOLOGY ............................................................................................................ 52 
4.0 Introduction .......................................................................................................... 52 
4.1 Research design .................................................................................................... 52 
4.2 Data for modelling ................................................................................................ 53 
4.3 Efficiency modelling and other considerations .................................................... 54 
4.3.1 Metafrontier analysis ..................................................................................... 55 
4.3.2 Global-frontier differences ............................................................................ 61 
4.3.3 Bootstrapping ................................................................................................ 65 
4.3.4 Testing returns to scale.................................................................................. 68 
4.3.5 Testing of differences in the distribution of efficiency scores ...................... 70 
4.4 Modelling inputs and outputs ............................................................................... 72 
4.4.1 Inputs ............................................................................................................. 73 
4.4.2 Outputs .......................................................................................................... 75 
vi 
 
University of Ghana  http://ugspace.ug.edu.gh
4.5 Other DEA consideration ..................................................................................... 75 
4.6 Conclusions .......................................................................................................... 76 
 
CHAPTER FIVE ............................................................................................................... 77 
ANALYSIS AND DISCUSSIONS ................................................................................... 77 
5.0 Introduction .......................................................................................................... 77 
5.1 Data Description ................................................................................................... 77 
5.2 Scale Elasticity in the International Oil Industry ................................................. 81 
5.3  Meta and Group Analysis of Inter- and Intra-IGO Performance ........................ 83 
5.4 Global Frontier Differences Analysis of Inter-IGO Performance ...................... 105 
5.5 Conclusions ........................................................................................................ 116 
 
CHAPTER SIX ............................................................................................................... 117 
SUMMARY, CONCLUSIONS AND RECOMMENDATIONS ................................ 117 
6.0 Introduction ........................................................................................................ 117 
6.1 Summary of study .............................................................................................. 117 
6.2  Conclusions of the study .................................................................................... 119 
6.3 Recommendations of the study .......................................................................... 122 
 
REFERENCES ................................................................................................................ 125 
APPENDIX A .................................................................................................................. 155 
APPENDIX B .................................................................................................................. 159 
APPENDIX C .................................................................................................................. 164 
APPENDIX D .................................................................................................................. 166 
APPENDIX E .................................................................................................................. 171 
 
 
vii 
 
University of Ghana  http://ugspace.ug.edu.gh
LIST OF TABLES 
Table                 Page 
Table 3.1: Crude Oil Proved Reserves (Billion Barrels) .................................................... 35 
Table 3.2: Proved Reserves of Natural Gas (Trillion Cubic Feet) ...................................... 36 
Table 3.3: Total Oil Supply (Thousand Barrels Per Day) ................................................... 38 
Table 3.4: Exports of Crude Oil (Thousand Barrels Per Day) ............................................ 39 
Table 3.5: Exports of Natural Gas (Billion Cubic Feet) ..................................................... 41 
Table 3.6: Total Petroleum Consumption (Thousand Barrels Per Day) ............................. 42 
Table 3.7: Dry Natural Gas Consumption (Billion Cubic Feet) ......................................... 44 
Table 4.1 : Membership distribution of IGOs ..................................................................... 54 
Table 4.2: Data of Hypothetical OPCs and IGOs ............................................................... 59 
Table 4.3 : Group and Meta Efficiencies ............................................................................ 61 
Table 4.4: Efficiencies for Global Frontier Difference Analysis ........................................ 64 
Table 4.5: Global Frontier Differences ............................................................................... 65 
Table 4.6: Model Variable Selection .................................................................................. 73 
Table 5.1: Descriptive Statistics of Pooled Data from 2000 to 2013 .................................. 78 
Table 5.2: Correlations between Inputs and Outputs .......................................................... 80 
Table 5.3: Scale Elasticity Tests (Simar & Wilson, 2002) using the Mean of Ratios ........ 82 
Table 5.4: Intra-Group Analysis of FSU Member States .................................................... 85 
Table 5.5: Intra-Group Analysis of IEA Member States .................................................... 89 
Table 5.6: Intra-Group Analysis of OAPEC Member States .............................................. 91 
Table 5.7: Intra-Group Analysis of OPEC Member States ................................................. 93 
Table 5.8: Inter-IGO Metafrontier Results.......................................................................... 97 
Table 5.9: Pairwise Comparisons of Inter-IGO Performance ........................................... 100 
Table 5.10: Technology Indices for the IGOs ................................................................... 106 
viii 
 
University of Ghana  http://ugspace.ug.edu.gh
Table 5.11: Global Frontier Differences against OPEC Frontier ...................................... 109 
Table 5.12: Global Frontier Differences against IEA Frontier ......................................... 110 
Table 5.13: Global Frontier Differences against FSU Frontier......................................... 110 
Table 5.14: Global Frontier Differences against OAPEC Frontier ................................... 111 
Table 5.15: Pairwise Comparisons of Technology Indices ............................................... 113 
 
 
 
 
 
 
 
 
 
 
 
 
 
ix 
 
University of Ghana  http://ugspace.ug.edu.gh
LIST OF FIGURES 
Figure                 Page 
Figure 2. 1: Outcomes of Inter and Intra Group Collaboration........................................... 20 
Figure 2.2: Distribution of Efficiency Studies by IGO ....................................................... 28 
Figure 2.3: The Cooper Conceptual Framework................................................................. 31 
Figure 3.1: Oil and Gas Reserve Trends ............................................................................. 37 
Figure 4.1: Graphical Illustration of Hypothetical Data ..................................................... 59 
Figure 5.1: Distribution of Metafrontier Scores of the IGOs .............................................. 98 
Figure 5.2: Distribution of Technological Gap Ratios ...................................................... 102 
Figure 5.3: Distribution of Technological Indices ............................................................ 108 
 
 
 
 
 
 
 
 
 
 
x 
 
University of Ghana  http://ugspace.ug.edu.gh
ABSTRACT 
Oil and gas is an important factor in the economic growth and development of most 
economies and an integral part of the global economy. Oil producing countries (OPCs) play 
a substantial part in the ownership, production and marketing of extracted hydrocarbons. 
This has made oil and gas production and supply decisions at the country level key to 
operations managers globally. To enhance their oil and gas management capabilities, some 
countries in the international oil industry have coalesced into intergovernmental 
organisations (IGOs) that are aimed at improving their bargaining powers through 
consensus and cooperation. However, although these IGOs play an important role in the 
industry, little is known on the production and supply efficiencies of member states. Based 
on this background, the study provides insights on the dynamics of intra and inter group 
performance of OPCs in order to inform operations managers in these countries on their 
performance and benchmarks in the industry. This was done by assessing performance of 
OPCs in a particular IGO and comparing their performance with other OPCs in other IGOs.  
An unbalanced panel of OPCs in four IGOs was used by mainly drawing data from the U.S 
Energy Information Administration and the World Bank Development Indicators. In all 53 
OPCs from the four IGOs in the international oil and gas industry for a 14-year period from 
2000 to 2013 were used. Relevant performance measurement techniques in management 
science and operations management were used. The study identified that the International 
Energy Agency’s (IEA) production frontier outperforms the production frontiers of the other 
three IGOs in the inter group performance. IEA, on average, were the best performers 
followed by the Organization of Arab Petroleum Exporting Countries (OAPEC), 
Organization of the Petroleum Exporting Countries (OPEC) and the Former Soviet Union 
(FSU) in that order. Finally, IEA and OAPEC production frontiers were seen to be 
consistently better than both the OPEC and FSU frontiers for all the 14 years of study. 
xi 
 
University of Ghana  http://ugspace.ug.edu.gh
CHAPTER ONE 
INTRODUCTION 
1.0  Background of the Study 
Oil and gas play important roles in the economic growth and development of most 
economies and have become an integral part of global economic life (Barros & Assaf, 2009; 
Cleveland, Costanza, Hall, & Kaufmann, 1997; Kashani, 2005b; Murphy & Hall, 2011b; 
Ramachandra, Loerincik, & Shruthi, 2006).  Nearly two billion dollars’ worth of petroleum 
are traded globally everyday (Tordo, Tracy, & Arfaa, 2011) and oil is the primary largest 
commodity of international trade in terms of volume and value traded globally (Desta, 2003; 
Ismail et al., 2013). The sector supports such industries as energy, transportation, 
industrialized agriculture, steel and plastic pipe production, health care and chemical 
industries, which generate income and employment (Francisco, de Almeida, & da Silva, 
2012; Ismail et al., 2013). This has made oil and gas production and supply decisions key 
to most economies globally.  
The importance of oil and gas has made oil producing countries (OPCs) play crucial roles 
in the supply of crude oil (Managi, Opaluch, Jin, & Grigalunas, 2006). OPCs through the 
production and export of oil and gas, hold substantial amount of the world petroleum 
reserves, production and marketing (Ike & Lee, 2014). Oil and gas affect relationships of 
countries and is a factor in determining foreign policy of producer and consumer countries 
(Hawdon, 2003; Stevens, 2008; Yergin, 2011). Of the 1,148 million barrels of stated oil 
reserves in the world, 77% are managed by states (Eller, Hartley, & Medlock, 2011). 
However, efficiency and productivity assessments in the oil and gas industry have ignored 
the cross-country assessment though OPCs play a critical role in the global oil and gas 
supply. 
1 
 
University of Ghana  http://ugspace.ug.edu.gh
Some countries in the international oil sector have coalesced into intergovernmental 
organisations (IGOs) in order to remain efficient and competitive in oil and gas production 
and supply. IGOs are international associations established by governments or their 
representatives that have been  sufficiently institutionalized by treaty to require regular 
meetings, rules governing decision making, a permanent staff and its headquarters (Shanks, 
Jacobson, & Kaplan, 1996). IGOs’ goal is to deliver high quality project outcomes and 
innovations to operational and economic issues at regional and global levels (Cissokho, 
Haughton, Makpayo, & Seck, 2013; Dorussen & Ward, 2008; Escobar & Le Chaffotec, 
2015; Holland, 1998). In meeting the world’s growing economic demands, there have been 
a number of IGOs with policy direction towards the oil and gas industry. The four IGOs at 
the forefront of forging global policy in the oil and gas industry are Organization of the 
Petroleum Exporting Countries (OPEC), Former Soviet Union (FSU) Organization of Arab 
Petroleum Exporting Countries (OAPEC) and International Energy Agency (IEA). Without 
IGOs, it may be difficult to monitor performance of oil and gas production and supply over 
time and set up international regulations and restrictions for countries (Biermann & Bauer, 
2004).  
Although many oil efficiency studies exist  (Al-Obaidan & Scully, 1995; Barros & Antunes, 
2014; Barros & Assaf, 2009; Barros & Managi, 2009a; Dike, 2013; Eller et al., 2011; 
Francisco et al., 2012; Hawdon, 2003; Ike & Lee, 2014; Ismail et al., 2013; Kashani, 2005b; 
Kim, Lee, Park, & Kim, 1999; Managi et al., 2006; Price & Weyman-Jones, 1996; 
Ramcharran, 2002; Sueyoshi & Goto, 2012a; Thompson, Dharmapala, Rothenberg, & 
Thrall, 1994; Wolf, 2009),  few have assessed the performance of IGOs  (Dike, 2013; Ike 
& Lee, 2014; Ramcharran, 2002) and they are the only ones who have examined the meta-
productive-efficiency, group efficiency, technology gap ratios and frontier differences of 
IGO members in terms of oil and gas production and supply using the metafrontier analysis 
2 
 
University of Ghana  http://ugspace.ug.edu.gh
(Battese, Rao, & O'Donnell, 2004; O’Donnell, Rao, & Battese, 2008a) or the global frontier 
difference (GFD) (Asmild & Tam, 2007). The GFD adequately compares the performance 
of the best-performing firms between different groups or the differences in the frontiers of 
different groups of firms. No oil efficiency study has tested the scale elasticity property 
based on the bootstrap algorithms (Simar & Wilson, 2002). Neither has any study used the 
Simar-Zelenyuk-adapted-Li test (SZAL) to statistically explore the significant differences 
in the distribution of meta-efficiency or average frontier differences between different IGOs 
and oil producing firms within these IGOs (Li, 1996; Simar & Zelenyuk, 2006). 
The purpose of this study is thus to contribute to the oil and gas efficiency literature by 
assessing meta-efficiency, group efficiency, technology gap ratios and frontier differences 
of oil-producing countries (OPCs) be it inter-IGOs or  intra-IGOs using a data envelopment 
analysis (DEA)-based  (Charnes, Cooper, & Rhodes, 1978b) metafrontier analysis (Battese 
et al., 2004; O’Donnell, Rao, & Battese, 2008b) and the GFD (Asmild & Tam, 2007). ‘Intra-
IGOs’, mean OPCs within one IGO whiles ‘inter-IGOs’, mean the 4 different groups of IGO 
in the oil industry. By this, we are able to determine the IGO-group-specific effect on 
member countries’ performance. Next for robustness check, the study tests the differences 
in the distribution of meta-efficiency levels and frontier estimates between OPEC and 
OAPEC, OPEC and FSU, OPEC and IEA, OAPEC and FSU, OAPEC and IEA, FSU and 
IEA using the SZAL (Li, 1996; Simar & Zelenyuk, 2006). Policy prescriptions are provided.  
 
1.1  Statement of Problem 
Despite the many oil and gas efficiency related studies (Zhou, Ang, & Poh, 2008), we 
identify some gaps in the recent literature. First, to the best of our knowledge, no paper has 
yet assessed the production efficiency or frontier differences between IGOs in the global oil 
3 
 
University of Ghana  http://ugspace.ug.edu.gh
industry. Though,  IGOs are explored in environmental policy (Biermann & Bauer, 2004), 
conflict policy (Dorussen & Ward, 2008), coastal zone management (Hayward & Cutler, 
2006) and several other sectors (Cao, 2009; Kalb, 2010). Oil efficiency studies have mainly 
focussed on policy reforms (Barros & Managi, 2009a; Kashani, 2005b; Price & Weyman-
Jones, 1996) and ownership and state involvement (Eller et al., 2011; Stevens, 2008; 
Sueyoshi & Goto, 2012a).   
Second, there are limited oil and gas efficiency and productivity studies that focus on inter-
country assessment (international benchmarking) or on oil-producing countries (OPCs).  
Majority of the studies where mainly on the performance of oil companies or the 
performance of oil firms in one country (Hawdon, 2003). For example, whereas Al-Obaidan 
and Scully (1995), Sueyoshi and Goto (2012a) and Wolf (2009) assessed the performance 
of oil companies, Barros and Assaf (2009), Barros and Managi (2009a) and Kashani (2005b) 
focused only on oil firms or blocks or units in a single country. However, OPCs and not oil 
companies, are usually the owners of natural resources including oil and gas reserves. 
Besides, it is the performance of OPCs which are more affected by regional or global shocks 
(Abdalla, 1995; Dike, 2013; Goldthau & Witte, 2011). Yet, only Hawdon (2003) evaluated 
the efficiency of OPCs although Hawdon (2003) just dwelled on gas distribution.  
Third, despite the several efficiency studies in the oil and gas industry  (Al-Obaidan & 
Scully, 1995; Barros & Antunes, 2014; Barros & Assaf, 2009; Barros & Managi, 2009a; 
Eller et al., 2011; Francisco et al., 2012; Hawdon, 2003; Ike & Lee, 2014; Ismail et al., 2013; 
Kashani, 2005b; Kim et al., 1999; Managi et al., 2006; Price & Weyman-Jones, 1996; 
Sueyoshi & Goto, 2012a; Thompson et al., 1994; Wolf, 2009), they all fail to adequately 
test the scale elasticity property of the industry before estimation. While different models 
exist to assess efficiency when the industry exhibits constant returns to scale (CRS) or 
4 
 
University of Ghana  http://ugspace.ug.edu.gh
variable returns to scale (VRS), previous studies have arbitrarily selected either or both scale 
elasticities in assessment. Wrong choice of the scale elasticity property will lead to 
misleading conclusions (Dyson et al., 2001; Simar & Wilson, 2002, 2011). As yet, only 
Hawdon (2003) has made some attempts to statistically test the returns to scale property in 
efficiency assessment in the oil industry. However, Hawdon (2003) relied on the t-test and 
Kolmogorov-Smirnov tests as proposed by Banker (1996) which do not provide consistent 
results for DEA estimations (Simar & Wilson, 2002).  
Finally, there are concerns about the methods used by previous studies to compare 
performance differences between groups of oil and gas firms. For example, whereas Kashani 
(2005a, 2005b) and Hawdon (2003) used t-test and Kolmogorov-Smirnov test to compare 
performance differences of oil and gas fields, Eller et al. (2011), Ike and Lee (2014) and 
Wolf (2009) relied on dummy variables in regression analysis for such comparison. These 
parametric and nonparametric statistics as used in previous studies are flawed when used in 
DEA estimations. First, they are point estimates that rely on the mean or median to the 
neglect of the entire distribution of scores (Zelenyuk & Zheka, 2006). Second, these tests, 
especially the parametric ones, require several statistical properties which DEA estimates 
do not possess (Simar & Wilson, 1998). Finally, they require a careful consideration of 
whether the groups under consideration are of a dependent sample or independent sample 
(Epure, Kerstens, & Prior, 2011; Kenjegalieva, Simper, Weyman-Jones, & Zelenyuk, 2009; 
Simar & Wilson, 2002). No paper in the oil and gas efficiency literature has used the Simar-
Zelenyuk-adapted-Li test (SZAL) to statistically explore significant differences in the 
distribution of efficiency or frontier estimates between different groups in the oil and gas 
industry (Li, 1996; Simar & Zelenyuk, 2006). This nonparametric test effectively compares 
the equality of distributions of efficiency estimates using kernel density estimations. 
5 
 
University of Ghana  http://ugspace.ug.edu.gh
1.2  Contributions of the Study  
The purpose of this whole study is to contribute to the oil and gas efficiency literature by 
assessing meta-efficiency, group efficiency, technology gap ratios and frontier differences 
of the 65 oil-producing countries (OPCs) belonging to the four IGOs under review using a 
metafrontier and GFD analysis from the period of 2000 to 2013, to explore the efficiency 
differences of the IGOs, among member countries, and whether the IGOs have any effect 
on member countries’ performance. This study makes several key contributions to policy, 
practice and research.  
On the basis of policy contributions, member states of the various IGOs in the international 
oil industry may be adequately informed about their (in) efficiencies in the production and 
supply of oil relative to other countries. Also the study compares, benchmarks and ranks the 
performances of the various oil related IGOs which can provide a useful insight into 
production and supply policy regulation. Since these IGOs will be given an empirically 
grounded assessment of their performance relative to similar IGOs in the industry, it would 
provide references for drafting rules governing decision-making on production and supply 
efficiency based on the outcomes and recommendations. Some useful and insightful 
managerial contributions are obtainable. It would enhance the decision making of OPCs 
who do not belong to any IGO to know which IGO to join based on the meta-efficiency 
levels and frontier difference estimates. OPCs that currently belong to IGOs may also be 
informed of their performance as compared to other member countries of that particular 
IGO and provide policy prescriptions and innovation. This, would help them put in measures 
to improve their performance and become more competitive. From this study, oil production 
policy makers and governments may be able to quantify the benefits of implementing 
efficiency measures.  
6 
 
University of Ghana  http://ugspace.ug.edu.gh
This study also makes five key contributions to research literature. First, it adds to the 
limited literature on inter-country oil and gas efficiency assessment which has only seen 
limited attention by Hawdon (2003) in gas distribution. Second, this is a premier oil and gas 
efficiency and productivity paper to apply the metafrontier and global frontier difference 
approach both of which cater for group heterogeneities. Estimation of technology gap ratios, 
group efficiency and meta-efficiency and global frontier difference. This is a novel 
empirical contribution in the oil and gas efficiency literature. Third, this is among the few 
studies to test the returns to scale property and the first in the international oil industry using 
Simar and Wilson (2002). Fourth and finally, this  study contributes to the few literature 
that employ the innovative Simar-Zelenyuk-adapted-Li (1996) test to delve deeper into 
comparing the frontier differences and the meta-efficiency of IGOs in the international oil 
and gas industry. 
 
1.3  Research Objectives 
The purpose of this study is to contribute to the oil and gas efficiency literature by assessing 
meta-efficiency, group efficiency, technology gap ratios and frontier differences of oil-
producing countries (OPCs) be it inter-IGOs or intra-IGOs over the period, 2000-2013. The 
specific objectives are: 
1. To test the scale elasticity property operating in the international oil industry of 
OPCs. 
2. To assess the intra-IGOs and inter-IGOs group efficiency, meta-efficiency and 
technology gaps of OPCs.   
3. To determine if a statistically significant difference exist in the distribution of the 
meta-efficiencies group efficiencies and technology indices of the four IGOs in the 
oil industry respectively. 
7 
 
University of Ghana  http://ugspace.ug.edu.gh
4. To evaluate the inter-group frontier differences of the four IGOs 
 
1.4  Research Questions 
The research seeks to answer the following questions: 
1. Does the international oil technology industry of OPCs exhibit constant or variable 
scale elasticity property?  
2. What are the intra-IGOs and inter-IGOs group efficiency, meta-efficiency and 
technology gap ratios of OPCs? 
3. Are there statistically significant differences in the distribution of the meta-
efficiencies, group efficiencies and technology indices of the four IGOs in the oil 
industry respectively? 
4. What are the inter-group frontier differences of the four IGOs? 
 
1.5  Limitations of the Study 
This study makes several key contributions to policy, practice and research, this 
notwithstanding there are some few challenges in terms of sample data and scope. Countries 
used for assessment were drawn from the list of members of the four IGOs under 
consideration. However, a few were eliminated from the sample because they produce either 
only oil or only gas. Whiles this is a limitation, majority of members of these IGOs produce 
both oil and gas. Additionally, in the sample data, a larger sample covering all oil producing 
countries in the world would have been appropriate but it is difficult to access due to data 
scarcity. The scope of the research covers the 14-year period from 2000 to 2013 as a result 
of data unavailability. Even though this period captures several dynamics in the international 
oil industry beyond this period would have allowed for more insights especially towards the 
8 
 
University of Ghana  http://ugspace.ug.edu.gh
new low price environment the oil industry is currently facing. Finally, there are several 
other IGOs like European Union (EU), Gulf Cooperation Council (GCC) and Organización 
Latinoamericana de Energia (OLADE) whose activities may influence the oil supply 
decisions of member countries. This notwithstanding, most of the members of these other 
IGOs are also members of the four IGOs that are currently under study. 
 
1.6  Thesis Structure 
The thesis is organized into six chapters. Chapter one is the introduction chapter that 
provides a research background, an overview of the problems being addressed in this thesis, 
the research contributions, objectives and questions as well as limitations of the study. This 
is the preliminary chapter that lays the foundations for this thesis. Chapter two is dedicated 
to review of relevant literature regarding operation of inter-governmental organizations. It 
begins with a theoretical review where theories that pertain to both inter-group performance 
and intra-group performance are adequately presented. This is then followed by an empirical 
review of the contemporary empirical works conducted on IGO-efficiency across various 
industries as well as the oil and gas industry. Chapter two ends with a conceptual framework 
that shows the process the research objectives are addressed. Chapter three is the context of 
the study. It is aimed at providing insights into how the international oil industry operates 
as well as an overview of the mechanisms for membership and operation of the IGOs under 
consideration. 
The research methodology is presented in chapter four. This chapter provides the research 
design, the sampling approaches and the mathematical justifications of the models to be 
used in this study. Research findings based on the objectives of this study are then organized 
in chapter five. Here, the theoretical and empirical literature reviewed in chapter two are 
9 
 
University of Ghana  http://ugspace.ug.edu.gh
used to support the research findings. Chapter six is the final chapter that provides a 
summary, conclusion and recommendations for policy, practice and further research. 
 
 
 
 
 
 
 
 
 
 
 
 
  
10 
 
University of Ghana  http://ugspace.ug.edu.gh
CHAPTER TWO 
LITERATURE REVIEW 
2.0  Introduction 
The section reviews theories on why membership of IGOs is expected to affect the 
efficiency of states in international oil industry. This consists of three main parts- a 
theoretical, an empirical review and a conceptual framework. The first part reviews the 
theories of intra-group performance that shows how state or firms gain or lose by belonging 
to a particular group. It also reviews theories on inter-group performance that shows why 
the performance of separate groups may differ. The second part reviews studies that have 
been done on efficiency of the oil industry. Finally, a conceptual framework that guides the 
entire research is presented.  
 
2.1  Theoretical Review 
2.1.1 Intra-Group Performance Theories 
Theories reviewed under this section provide an understanding of performance gains as a 
result of membership of a particular group. It will explain how units tend to benefit or lose 
by belonging to groups. Particularly, Social Network Theory (Mitchell, 1969; Tichy, 
Tushman, & Fombrun, 1979), Social Facilitation Theory (Triplett, 1898; Zajonc, 1965, 
1968) and Institutional Theory (DiMaggio & Powell, 1983; Meyer & Rowan, 1977) will be 
reviewed. 
  
11 
 
University of Ghana  http://ugspace.ug.edu.gh
2.1.1.1 Social Network Theory 
The effects of states’ membership and formation of IGOs and the importance of such 
governmental organisations have long been a topic of interest in world politics, economic 
development and academia. IGOs are formed to link actors (states) in acquiring information 
about interests and intentions because they create ties between states (Dorussen & Ward, 
2008). A Social Network consists of a set of actors (states) and their relations (ties) between 
the states (Wasserman & Faust, 1994). Social networks are important in sustaining and 
improving attitudes, behaviours, health and well-being of actors in these social networks 
(Miles, 2012).  
The theory is often credited to earlier works of Barnes (1954), Bott (1957) and Granovetter 
(1973). Its academic applications span a broad range of disciplines including sociology, 
operations research, social psychology, political science, mathematics, epidemiology, 
computer science and economics just to mention a few (Katz, Lazer, Arrow, & Contractor, 
2004; Marwell, Oliver, & Prahl, 1988). In operations management for example, complex 
logistic networks, transportation problems and some facility location decisions are based on 
social network theoretical considerations (Jackson, 2010), (Lovejoy & Handy, 2011; 
Lovejoy, Sciara, Salon, Handy, & Mokhtarian, 2013).  
The theory explains why individual groups or organization create these ties as an investment 
in the accumulation of social resources and capital (Katz et al., 2004). The basic idea of the 
theory is that the behaviour in a set of connection as a whole can be used to explain the 
behaviour of the actors (state) in the set (Mitchell, 1969; Tichy et al., 1979). States are said 
to be socially networked when they tend to think and behave similarly because they are 
connected (Garton, Haythornthwaite, & Wellman, 1997; Miles, 2012). These connections 
may include; communication ties (who gives information to who); formal ties (who reports 
12 
 
University of Ghana  http://ugspace.ug.edu.gh
to who); affective ties (who trusts who); material ties (who supports who); proximity ties 
(who is close to who); and cognitive ties (who knows who) (Katz et al., 2004). 
Social network theorists believe that the underlying structure of the group determines the 
type, access and flow of resources to actors in the network (Daly, 2012). Therefore the 
theory moves away from the adage “It is not what you know, but who you know” to a more 
accurate adage “Who you know defines what you know” (Cross & Parker, 2004; Daly, 2012; 
Newman, Barabasi, & Watts, 2006). This theory can be applied to various category of 
analysis from individuals, organization or nation state that have same attributes for 
reciprocation (Kadushin, 2004). 
There has, however, been concerns on the mode in which the concept such as distance, 
social structure and cohesion can be applied in real world systems (Embirbayer & Goodwin, 
1994). There is also debate over why people join networks. They may form these groups in 
order to maximize their personal preferences and desires (Katz et al., 2004). However, 
people may also join networks for more strategic and instrumental reasons (Kilduff & Brass, 
2010). 
It must be understood that this theory helps explain both the positive and negative 
consequences of membership of such social networks. These ties or network may provide 
positive utilization to countries in the oil industry by influencing access to resources, 
reducing transactional costs and building interest based on coalitions (Lauber, Decker, & 
Knuth, 2008). It also facilitates the flow of information between actors (Tindall & Wellman, 
2001). Countries may stand to benefit from membership of these groups through pooled 
collaborations in terms of technology and research as well as bargaining on global markets. 
However, because the ties between states in the group may be weak or strong (Granovetter, 
1973; Katz et al., 2004; Miles, 2012), membership of these bodies may not have the same 
13 
 
University of Ghana  http://ugspace.ug.edu.gh
level of success to all actors. Indeed, proponents of the theory believe that even strong 
membership may constrain an actor in maximizing performance (Daly, 2012; Lucas & 
Mayne, 2013), especially when the decisions of the group is not leading to improved 
performance. Eller et al. (2011) and Ike and Lee (2014), for example, have seen that 
production quotas of OPEC has over the years resulted in lower efficiencies of member 
states because of their inability to produce at truly efficient production levels. Finally, 
because individual actors may have overlapping and cross-cutting relationships with a 
multitude of groups (Katz et al., 2004), inter-group performance may differ 
 
2.1.1.2 Social Facilitation Theory 
Individuals respond to a range of incitements in the environment, significant of this is the 
presence of another actor or audience. This can have a positive or a negative effect on an 
individual’s performance and has been a subject debated for researchers in the field of social 
science and psychology for centuries (Miles, 2012; Zajonc, 1965). 
At the basic level, the social facilitation theory believes that the performance of various 
actors are not solely dependent on their own performance, but can also be largely influenced 
by other persons around the individual (Crawford, 1939; Miles, 2012). This theory was first 
identified by Triplett (1898). The term “social facilitation” was first introduced by the 
Allport (1920) and defined as an increase or decrease in response to sight and sound of 
others making the same movement. Crawford (1939) defined it as the measurement of 
individual activity resulting from the presence of others. According to the theory, social 
facilitation and interference, the mere presence of others is a source of generic and 
nondirective arousal that enhances the dominant responses of the performer (Markus, 1978). 
14 
 
University of Ghana  http://ugspace.ug.edu.gh
Social facilitation deals with the importance of social presence on individual or groups 
performance. It focuses on the changes in performance that occurs when individuals or 
groups are evaluated or observed by others (Aiello & Douthitt, 2001). It refers to 
performance enhancement and improvement effects engendered by the presence of others 
either as co-actors, more typically as observers or audience (Blascovich, Mendes, Hunter, 
& Salomon, 1999). 
When individuals or group performance or action is the focus of the attention of others, the 
reactions of the individuals are related to the meaning assigned by the social presence (Uziel, 
2007). This is because social presence increases drive and motivate greater effort (Triplett, 
1898). The presence of others strengthens the correct responses and has a beneficial effect 
on performance (Hunt & Hillery, 1973; Martens, 1969; Zajonc & Sales, 1966) 
However, Cottrell (1968) on the other hand proposed that the stimulation and drive for 
individual or group to perform is occasioned by the individual or group concern of others 
evaluating its performance. The theory is only concerned with a range of performance 
levels, instead of specific desired outcomes under the effects social presence or the 
environment (Aiello & Douthitt, 2001; Kelley & Thibaut, 1954).  
In the context of this study, it can be thought that other actors in the intergovernmental 
organization may have effect on the behavior of individual members to improve in their 
operations for efficiency and gains. In other words, when countries are operating alone, their 
performance may not be as good as if they are in a group. This is because the mere presence 
of another person is sufficient to influence an individual’s behavior. The power of others to 
influence an individual behavior is readily apparent in problems of imitation, conformity, 
competition, helping and aggression (Markus, 1978).  Whiles it may be true that the 
presence of others can improve performance, it is equally likely that performance of 
15 
 
University of Ghana  http://ugspace.ug.edu.gh
individual countries may be constraint by belonging to such groups. This is because group 
level distractions and negative influences of bad company may hamper the drive of a country 
towards optimal production levels. 
 
2.1.1.3 Institutional Theory  
Organisations in a particular industry tend to act and look same  (DiMaggio & Powell, 
1983). The concept of the theory is that the organizational structure and process help achieve 
their own right of effectiveness and efficiency in their desired outcomes through the goals 
and missions of the institution (Lincoln, 1995) 
Institutions have been defined as “regulative, normative, and cognitive structures and 
activities that provide stability and meaning for social behavior” (Scott, 1995). Institutions 
therefore exert three types of pressure on organizations; coercive, normative, and mimetic 
(DiMaggio & Powell, 1983). The theory posits that institutionalized activities occur due to 
influences on three levels: individual, organizational, and inter-organizational (Oliver, 
1997). Whereas managers consciously and unconsciously follow the laid down processes or 
norms, customs and traditions at the individual organization level (Berger & Luckmann, 
1991)  shared political, social and cultural belief systems guide the behaviors of a group of 
organizations (Miles, 2012). At the inter-organizational level, pressure from government, 
affiliated persons, and society define what is socially acceptable and expected of parties in 
the institution which drives them to look and act the same (DiMaggio & Powell, 1983). 
Most institutional theories see local actors whether individuals, organizations, or national 
states as being affected by institutions in their environments. Individuals and organizations 
are affected by societal institutions, and national-states by a world society to effectively 
operate in a particular manner (Greenwood, Sage, & Sage, 2008). The theories has been 
applied in the social sciences, and especially in political science (March & Olsen, 1983), 
16 
 
University of Ghana  http://ugspace.ug.edu.gh
economics (Alston, Eggerston, & North, 1996; Khalil, 1995; North, 1990) and in sociology 
(DiMaggio & Powell, 1983; Scott, 1995; Zucker, 1987). 
It can therefore be implied that the activities and ways of operation by individual countries 
in the intergovernmental organization will be very similar to that of other members. This is 
because countries will be more expected to follow the institutionalized behaviors of the 
organization. Therefore, if the institutionalized behaviors of a particular IGO is efficiency-
improving, it is expected that all members will also enjoy similar levels of efficiency. 
Conversely, where institutionalized activities do not lead to higher performance, then all 
members in that particular IGO will suffer. Institutional theory therefore believes that the 
performance of members in a particular IGO will be very similar. 
 
2.1.2 Inter-Group Performance Theories  
The theories on inter-group performance seeks to explain why the performance of different 
blocs of countries will differ. It is expected that the performance of individual IGOs will 
differ in the international oil industry. Therefore, two main theoretical views will be used to 
explain this possibility. These are the game theory  (Von Neumann & Morgenstern, 1944) 
and the resource-based theory (Penrose, 1959). 
 
2.1.2.1 Game Theory 
The theory explains decisions individuals or a group of players take in order to win a game 
when competing with one or more opponents (Von Neumann & Morgenstern, 1944). This 
theory describes the strategies employed by competitors in their choices of action that 
enhances their chances of gain or loss by considering the action being taking by their 
opponents (Miles, 2012). It basically examines the actions players make that decides the 
outcomes, gains or the optimal decision (Madhani, 2010; Rasmussen, 1989) 
17 
 
University of Ghana  http://ugspace.ug.edu.gh
The game-theoretical approach studies systems with multiple self-interested parties with the 
aim of predicting the likely outcomes of the system under rational behavior of the players 
with mutual and possibly conflicting interests (Trestian, Ormond, & Muntean, 2012; Yin et 
al., 2012). In other words, game theory studies games, which are mathematical models of 
relationships and interactions among multiple players, each trying to advance their self-
interest by choosing among a set of strategies (Weibull, 1997; Yin et al., 2012). It must be 
noted that, the theory provides useful mathematical tools necessary to understand the 
possible strategies that parties may follow when not only competing but also collaborating 
in games (Trestian et al., 2012; Weibull, 1997). The players seek to maximize their payoffs 
by choosing among alternative strategies that deploy actions depending on information 
available at a certain moment. Each player chooses strategies which can maximize their 
payoff  (Trestian et al., 2012). 
Although originally adopted in economics, in order to model the competition between 
companies, it has found wide application in other areas, such as biology, accounting, 
management, finance, marketing, sociology, politics, international relations, philosophy, 
computer science, operations research and engineering (Miles, 2012; Trestian et al., 2012; 
Weibull, 1997). For example, in operations research, Huang (Huang & Li, 2001) applied 
game theoretical perspective of Stackelberg game to model and solve advertising problems 
in manufacturer-retailer supply chain problems. Although early understanding of game 
theory was developed to analyse competitions in which one individual does better at 
another's expense, zero sum games (von Neumann & Morgenstern, 1947), many equilibrium 
concepts have been developed and incorporated into the theory to handle situations where 
both or all parties benefits (Nash, 1996; Weibull, 1997). Among them is the famous Nash 
equilibrium, in an attempt to capture this idea (Nash, 1996).  
18 
 
University of Ghana  http://ugspace.ug.edu.gh
The application of game theoretical ideologies in a variety of contexts, including policy-
oriented ones, has become an important and expanding research area, particularly where 
institutions, including government are involved (Eleftheriadou, 2008; Frisvold & Caswell, 
2000; O'Toole, 2004; Rigby, Dewick, Courtney, & Gee, 2013). This is probably as a result 
of the call by O'Toole (2004) for researchers in policy planning and implementation to 
recognize the importance of interaction between actors for successful policy 
implementation, and the use of game theory to understand the choices available to such 
policy actors. 
In the application of game theory to such governmental-backed intergovernmental 
institutions, both inter and intra-group dynamics of collaboration and competition among 
members can be explained. This has been summarized in Figure 2.1 that show various levels 
of collaboration among members of a particular IGO and other IGOs. This has been adapted 
from (Rigby et al., 2013) benchmarking options matrix. 
 
 
 
 
 
 
 
 
 
19 
 
University of Ghana  http://ugspace.ug.edu.gh
 Inter-Group Collaboration 
  Low High 
Low exchange of information within Low exchange of information within 
and between groups results in: group renders between group 
collaboration irrelevant leading to: 
 
 
1. Organizational learning not 
supported 1. Organizational learning not 
2. Information asymmetries not supported 
addressed 2. Information asymmetries not 
addressed 
High exchange of information within High exchange of information within 
group but low exchange between and between group results in: 
group results in: 
 
 
1. Organizational 
1. Organizational learning/innovation 
learning/innovation 2. Information asymmetries 
2. Information asymmetries not addressed 
addressed 
 
Figure 2.1: Outcomes of Inter and Intra Group Collaboration 
Source: adapted from (Rigby et al., 2013) 
 
In the bottom left box, where there is a high willingness to collaborate within a particular 
IGO but low level of cooperation between different IGOs, good intra-group efficiency can 
result and could benefit individual countries in terms of group learning which meets the 
objectives of good public policy. However, this will not address possible information 
asymmetries in policy that may foster achievement of Nash equilibrium or Pareto optimality 
between different IGOs. On the top left-hand and right-hand boxes, however, low levels of 
intra-group collaboration mean that, policy aimed enhancing inter-group efficiency will not 
be met sufficiently (Rigby et al., 2013). This is because, the low level of cooperation 
between members of a particular IGO will reduce the benefits of even high inter-group 
collaboration. (Rigby et al., 2013) believe that it is only when there is high intra-group and 
20 
 
Intra-Group Collaboration 
High Low 
University of Ghana  http://ugspace.ug.edu.gh
high inter-group collaboration that a true optimal point can be achieved. This is because, 
such collaboration will enhance organizational learning and reduce or remove such inter-
group information asymmetries. 
The theoretical views of game theorists are therefore very important in assessing and 
understanding the competition and collaboration between various IGOs in the international 
oil industry.  IGOs employ alternative strategies that may be a winning strategy with higher 
outcome in a particular situation it faces in the competitive environment. Individual member 
states tend to benefit or suffer in efficiency depending on not only the level of intra-group 
learning and cooperation, but also inter-group collaboration. 
 
2.1.2.2 Resource-based Theory (RBT)  
Groups or organisations want to be seen or identified as unique based on the special 
competences and resources they possess. In short, the resource-based theory (RBT) studies 
the differences in outcomes in respect of their resources (Peteraf & Barney, 2003). The 
theory can be attributable to the early works of Penrose (1959) that theorized about how a 
firm’s resources influence its growth; in particular, where growth is constrained when 
resources are inadequate (Barney, Wright, & Ketchen Jr, 2001). Theory defines the 
organization’s uniqueness and position in competitive situations in the environment  
(Hoopes, Madsen, & Walker, 2003). Its emphasis is on differences in efficiency (Peteraf & 
Barney, 2003). The focus of the theory is how the organization acts against competitors on 
their strength, competence and resources that shows performance differences in the 
environment (Barney, 1991; Miles, 2012; Wernerfelt, 1984).  
Value, rarity, inimitability, and non-substitutability are among the commonly cited 
characteristics that provide the core logic linking resources to competitive advantage 
21 
 
University of Ghana  http://ugspace.ug.edu.gh
(Sirmon, Hitt, Ireland, & Gilbert, 2011). The resource-based theory of strategy (RBT) 
hinges on the argument that firms with valuable, rare, special and inimitable resources have 
the potential of achieving superior performance (Amit & Schoemaker, 2012; Barney, 1991; 
Barney, 1995; Bharadwaj, 2000; Wiklund & Shepherd, 2003). RBT uses the internal 
characteristics and resources of firms to explain their heterogeneity in strategy and 
performance (Camisón & Villar-López, 2014). Basically, RBT assumes that there is 
underlying production heterogeneities or differences across firms (Barney, 1991; Dobbin & 
Baum, 2000; Peteraf, 1993). Thus production processes and resources are different from 
firm to firm. Therefore, firms endowed with such superior resources are able to produce 
more economically and/or better satisfy customer wants  (Peteraf, 1993). Heterogeneity in 
this context also implies that firms of varying capabilities are able to compete in the same 
marketplace and, at least, breakeven (Dobbin & Baum, 2000; Peteraf, 1993). Accordingly, 
the main assumption of RBT is that only firms with certain resources and capabilities with 
special characteristics will gain competitive advantage and, therefore, achieve superior 
performance (Camisón & Villar-López, 2014). 
This is therefore a theory of competitive advantage (Barney, 2001) since the central theme 
is that privately held knowledge (or resource) is the basic source of advantage in competition 
(Conner & Prahalad, 1996). A resource-based approach to firm management focuses on 
costly–to-copy attributes of the firm as sources of economic rent and therefore as the 
fundamental drivers of performance and competitive advantage (Barney, 1986; Conner & 
Prahalad, 1996; Rumelt, 1974). In resource-based theory, such resources may be financial, 
human, intangible, organizational, physical or technological (Dobbin & Baum, 2000). 
Indeed, (Miller & Shamsie, 1996) broadly separated these resource into knowledge based 
and those that are property-based resources. 
22 
 
University of Ghana  http://ugspace.ug.edu.gh
In early conceptions, the RBT emphasized how variation in firms’ access to key factor 
inputs could lead to variation in firm performance (Barney, 1991; Wernerfelt, 1984). 
However, subsequent extensions include a broader focus including competence-based view 
(Foss, 1996), commitment (Ghemawat, 1991),  dynamic capabilities (Teece & Pisano, 1994; 
Teece, Pisano, & Shuen, 1997), knowledge-based (Foss, 1996), relation-based (Dyer & 
Singh, 1998), and attention-based (Ocasio, 1997) approaches. The resource based view 
generally addresses performance differences between firms using asymmetry in knowledge 
and in associated competences or capabilities (Amit & Schoemaker, 2012; Conner, 1991; 
Henderson & Cockburn, 1994; Peteraf, 1993). 
This theory is mostly unique to the field of strategic management (Conner & Prahalad, 1996; 
Peteraf, 1993), although some extensions and applications have been made to related areas 
like neoclassical microeconomics and evolutionary economics (Barney, 2001). The RBT is 
undoubtedly consistent with and well rooted squarely in the policy research tradition 
(Barney, Ketchen, & Wright, 2011; Peteraf, 1993) and hence very important in explaining 
issues related to inter-organizational performance differences. 
This theory of the firm therefore believes that, the way a firm is organized, when combined 
with firm resources, can better enhance the positive relationship between resources and firm 
performance (Barney, 1995; Wiklund & Shepherd, 2003). Decades of empirical work have 
given ample support to the importance of these resource characteristics for firm performance 
(Crook, Ketchen, Combs, & Todd, 2008; Sirmon et al., 2011) and is widely acknowledged 
as one of the most prominent and powerful theories for describing, explaining, and 
predicting organizational relationships (Barney et al., 2011). Therefore, in the context of 
this study, when an IGO is considered as the unit of analysis, it stands to reason that the 
unique competencies and resources of members of the particular IGO will give it 
competitive advantages over other relevant IGOs in the international oil industry. Size of 
23 
 
University of Ghana  http://ugspace.ug.edu.gh
oil and gas reserves, human, technical and technological competencies of member states, 
political and economic bargaining powers are all relevant tangible and intangible resources 
that an IGO may use to out-compete others. RBT therefore believes that IGOs with such 
higher levels of resources will be more efficient. 
 
2.2 Empirical Review 
The empirical review presents an overview of peer-reviewed studies on efficiency, 
performance and productivity of various parties in the oil and gas industry. Extent of 
knowledge on efficiency of countries, oil firms, drilling blocks as well as intergovernmental 
organisations have been presented in a manner that will provide adequate understanding of 
the extent of literature. Review of frontier efficiency papers in the oil industry are based on 
Appendix B which provides a taxonomy of these papers reviewed. On the other hand, a 
taxonomy of papers on efficiency of intergovernmental organizations has been presented in 
Appendix A.  
 
2.2.1 Frontier Efficiency in the Oil Industry 
There has been quite a number of efficiency-related research in the oil and gas industry, 
substantially among the issues are ownership, privatization, environmental efficiency, 
international comparison, regulation and state intervention. For example, Al-Obaidan and 
Scully (1995); Ike and Lee (2014); Sueyoshi and Goto (2012a); Wolf (2009) have assessed 
ownership and efficiency of firms in the industry. The focus of these studies is to understand 
the differences in the efficiency of state-owned and privately owned oil firms. Closely 
related to this theme are papers on privatization, government regulation and government 
intervention (Hawdon, 2003; Kashani, 2005b; Price & Weyman-Jones, 1996). Their focus 
is to examine the influence of various levels of government involvement in the oil and gas 
24 
 
University of Ghana  http://ugspace.ug.edu.gh
industry and its effects on efficiency of players in the industry. Empirically, the role of 
government in the industry has been quite adverse, from the negative effects of state 
ownership (Eller et al., 2011; Ike & Lee, 2014) to the potential for efficiency gains when 
government influence is removed (Kashani, 2005b; Price & Weyman-Jones, 1996). 
Another important theme for assessment is the issue of environmental efficiency. Francisco 
et al. (2012); Ismail et al. (2013); Sueyoshi and Goto (2012a) have measured efficiency by 
incorporating harmful by-products of oil and gas exploration, distribution and consumption 
activities such as CO2 emissions, and transmission losses. For, Barros and Antunes (2014); 
Managi et al. (2006) their focus was on technological change, productivity change and 
comparison of Malmquist and Luenberger indices. Growth Accounting and Productivity 
Methods have also been investigated by Barros and Managi (2009b) and Thompson et al. 
(1994). Thompson et al. (1994) explored the integration of DEA and other methods like 
Assurance Region (AR) and Cone Ratio (CR) in assessing efficiency and performance of 
international oil companies. Indeed, there have been a number of conceptual papers 
dedicated to improving the methodology used in such benchmarking (Sueyoshi & Goto, 
2012a; Thompson, Dharmapala, Humphrey, Taylor, & Thrall, 1996). 
Finally, since this study does international comparison of the oil and gas supply activities 
of various countries, it is encouraging that some studies have previously attempted such 
international comparison. Hawdon (2003); Kim et al. (1999) have all attempted such 
comparison, except that all two studies only focussed on gas distribution to the neglect of 
crude oil. Similarly, these two papers that have attempted international comparison only 
focussed on the downstream oil and gas industry. Although the body of literature in the oil 
and gas industry is substantial, this is rarely done an international benchmarking and in a 
more aggregated context or mainly geared towards the upstream operations. However very 
few studies are dedicated to cross country and regional efficiency in the downstream. These 
25 
 
University of Ghana  http://ugspace.ug.edu.gh
includes Hawdon (2003) who analysed the efficiency of 33 countries gas industry, and Kim 
et al. (1999) who examined 28 natural gas transmission and distribution companies 
operating in 8 countries. 
The various approaches employed in estimating efficiency in the oil and gas industry 
extends a number of methods including parametric and non–parametric analysis. DEA, a 
non-parametric technique, and its variations have been largely used in most of the literature 
for example by Ismail et al. (2013); Sueyoshi and Goto (2012a) who used the method to 
measure environmental efficiency of 19 and 17 upstream oil firms respectively. Ike and Lee 
(2014) used the slacks-based measure which is a variant of the DEA non-parametric method. 
Whiles Barros and Assaf (2009); Hawdon (2003) have applied DEA together with the 
bootstrapping techniques. Other studies like Al-Obaidan and Scully (1995); Managi et al. 
(2006) have found support in the use of parametric efficiency techniques like SFA and 
Aigner and Chu deterministic frontier which require several assumptions which are 
sometime unrealistic in a real-world contexts (Chase, 2012).  
For some, a methodology that combines a number of techniques have been seen to provide 
a more complementary result. Ike and Lee (2014) combined DEA, Malmquist Productivity 
Index and Regression to assess 38 upstream oil companies. Eller et al. (2011) estimated 
efficiency differences of 78 upstream oil firms with DEA and SFA. Similarly, Kim et al. 
(1999); Lee, Kim, and Park (1996) applied Edgeworth Index and ANOVA, and Multilateral 
Tornqvist, Managerial Index, System Analysis, and non-parametric efficiency analysis  to 
examine 28 Natural Gas transmission and distribution companies operating in 8 countries. 
Whereas the diversity of methods used in these assessments are quite appreciable, none of 
the papers reviewed used methods that adequately cater for group differences 
(heterogeneities) in estimating the frontier efficiency. Methods like metafrontier analysis, 
global frontier differences are loudly missing in the literature. 
26 
 
University of Ghana  http://ugspace.ug.edu.gh
2.2.2 Group Performance and Efficiency 
The extant studies regarding group performance and efficiency of IGOs has been heavily 
skewed towards banking, energy efficiency, and health. For example, whereas Abu-Alkheil, 
Burghof, and Khan (2012) concentrated on bank efficiency of OAPEC and Gulf 
Cooperation Countries (GCC), Casu and Girardone (2004, 2006); Casu and Molyneux 
(2003); Claeys and Vander Vennet (2008); Marius Andrieş and Căpraru (2012), Košak, 
Zajc, and Zorić (2009); Mamatzakis, Staikouras, and Koutsomanoli-Filippaki (2008) looked 
at bank efficiency in the European Union (EU). Others worthy of mention are Behname 
(2012) who examined banking efficiency among OPEC countries and Drakos (2003) who 
considered same in the FSU and Central and Eastern European (CEE) countries. Just like 
banking, several studies can be cited for energy efficiency (Adetutu, 2014; Al‐Rashed & 
León, 2015; Filippini & Hunt, 2011; Goldthau & Witte, 2011) and health (Adler-Milstein, 
Ronchi, Cohen, Winn, & Jha, 2014; Al-Essa, Al-Rubaie, Walker, & Salek, 2015; Oderkirk, 
Ronchi, & Klazinga, 2013; Retzlaff-Roberts, Chang, & Rubin, 2004). 
Other research issues like education (Afonso & St Aubyn, 2005; Afonso & St. Aubyn, 2006; 
Krishnasamy & Ahmed, 2009) economy (Arestis, Chortareas, & Desli, 2006; Fare, 
Grosskopf, Norris, & Zhang, 1994; Krishnasamy & Ahmed, 2009), agriculture, (Arnade, 
1994; Gorton & Davidova, 2004; Vlontzos, Niavis, & Manos, 2014), insurance (Donni & 
Fecher, 1997), railways (Oum & Yu, 1994), postal service, environment and policy 
(Selowsky & Martin, 1997) have also been considered by other studies.  
Generally, although these studies see several differences in the level of efficiency of 
individual members of various IGOs (Oderkirk et al., 2013; Vlontzos et al., 2014) some 
particular IGOs have been seen to experience modest improvements in efficiency. For 
example Adetutu (2014) observed that some selected OPEC countries have modest energy 
efficiency arising from subsidy effect and artificially low energy prices. This 
27 
 
University of Ghana  http://ugspace.ug.edu.gh
notwithstanding, little is known about efficiency of these IGOs with respect to oil and gas 
production and supply efficiency. Even studies who purposely targeted oil-focussed IGOs 
like OPEC, OAPEC, FSU and IEA (Adetutu, 2014; Al‐Rashed & León, 2015; Behname, 
2012; Goldthau & Witte, 2011)  rather looked at issues like banking efficiency and energy 
efficiency which has nothing to do with oil and gas production and supply efficiency. 
Studies that have come close to examining supply efficiency include Dike (2013) who 
looked at security in energy exportation of OPEC countries and Ramcharran (2002) who 
examined efficiency and production responses to price changes in the international oil 
industry. 
When the focus of the review is shifted to which IGOs have attracted substantial research 
interest, interesting insights are revealed. To aid in this assessment, Figure 2.2 has 
graphically presented the distribution of research on the various IGOs. This is based on the 
taxonomy in Appendix A.  
ASEAN AU CEE EU FSU G20 G8 GCC IEA OAPEC OECD OPEC UN
IGO
  
Figure 2.2: Distribution of Efficiency Studies by IGO 
                 Source: Fieldwork (2015)   
28 
 
Frequency
0 2 4 6 8 10 12 14
University of Ghana  http://ugspace.ug.edu.gh
Quite evident in Figure 2.2 is that for the papers reviewed, OECD has attracted the most 
research interest with as many as 15 studies purposely studying this IGO individually or 
together with other IGOs. For example, Afonso and St Aubyn (2005); Çakır, Perçin, Min, 
and Gunasekaran (2015); Fare et al. (1994); Hori (2012); Krishnasamy and Ahmed (2009); 
Oderkirk et al. (2013) have all looked at various dimensions of efficiency in OECD 
countries. EU and OPEC follow with 12 papers and 7 papers respectively. Bosseboeuf, 
Chateau, and Lapillonne (1997); Casu and Girardone (2004); Casu and Molyneux (2003); 
Claeys and Vander Vennet (2008); Marius Andrieş and Căpraru (2012) looked at EU whiles 
Adetutu (2014); Behname (2012); Dike (2013); Goldthau and Witte (2011); Ramcharran 
(2002); Sari and Soytas (2009) focused their attention on OPEC countries. At the bottom 
are G20, G8 and ASEAN who recorded only one paper each. It must be clearly noted that 
FSU, OAPEC and IEA who are among the IGOs to be considered in this study are among 
those which have seen few research interest from the review. Insights on their efficiency 
will clearly add to the existing body of knowledge in these IGOs. From the taxonomy in 
Appendix A, it is also evident that most studies on IGOs concentrate on one particular IGO 
in their assessment. Gorton and Davidova (2004); Muldoon et al. (2011); Shahabinejad, 
Mehrjerdi, and Yaghoubi (2013); Taylor, d’Ortigue, Francoeur, and Trudeau (2010) have 
concentrated on intra-group performance assessment considering efficiency only in that 
particular IGO of interest.  A few studies like Abu-Alkheil et al. (2012); Aristovnik (2012); 
Drakos (2003); Selowsky and Martin (1997) have based their research arguments on 
samples comprising two or more IGOs. However, among these studies whose samples cut 
across various IGOs, there has been little inter-group comparison to identify and rank IGOs 
with respect to a particular phenomenon understudy. There is therefore room for this study 
to build literature by not only conducting intra-group assessment, but also an inter-group 
performance assessment which is lacking in literature. 
29 
 
University of Ghana  http://ugspace.ug.edu.gh
Finally, the focus is on the methods used in the assessment and whether these methods 
adequately cater for group heterogeneities. In assessing the performance and efficiency of 
groups (IGOs), several models relating to the issue of international comparisons and frontier 
efficiency have been applied. Abu-Alkheil et al. (2012) used DEA to determine banks 
efficiency in OAPEC and GCC countries between 2005 and 2008. Arestis et al. (2006)  
applied it to 26 OECD Countries from 1963 to 1992. Afonso and St Aubyn (2005) used both 
DEA and FDH to estimate the education and health efficiencies of 24 OECD countries. 
Others like Filippini and Hunt (2011) determined the energy efficiency of 29 OECD 
countries for 28 years period from1978 to 2006 using SFA. Košak et al. (2009); Mamatzakis 
et al. (2008) estimated bank efficiency in 5 New EU Member States from 1996 – 2006 and 
10 EU countries from 1998–2003 respectively, using the SFA approach. Regression-based 
estimation approaches like the Translog Cost Function (Adetutu, 2014; Claeys & Vander 
Vennet, 2008) and Auto Regressive Distributed Lag (ARDL) (Sari & Soytas, 2009) have 
also seen some considerable use. Whereas all these techniques have their own advantages 
and disadvantages, it is important that papers that compare various groups use models that 
can adequately cater for group differences in estimating the efficiency. From the review, 
only Krishnasamy and Ahmed (2009) used the Metafrontier approach to measure the 
efficiency and productivity of the economies of twenty six (26) OECD countries from 1980 
to 2008. Their paper does not even conduct inter-IGO comparison since the focus was only 
on OECD. Most of the reviewed literature employed in this study have applied a number of 
models, some in combination to estimate efficiency of groups or IGOs. But there is rare 
evidence regarding Global or Meta frontier analysis in the international oil and gas industry. 
Hence there is enough gap for this paper to situate in efficiency assessment the use of models 
that adequately cater for such group differences in the estimation of efficiency. 
 
30 
 
University of Ghana  http://ugspace.ug.edu.gh
2.3 Conceptual Framework 
The process of analysis and assessment of the objectives of this study will follow the Cooper 
Framework of Emrouznejad and De Witte (2010) which clearly shows the interrelated 
phases any DEA-based assessment should follow. This is illustrated in Figure 2.3. 
 
PHASES IN DEA PROJECT 
Concepts and 2 On Structuring Data 
Objectives 
1 
3    Operational Models 
DMU 
4 Performance 
6 Results and Comparison Models 
Deployment 
5 Evaluation 
Figure 2.3: The Cooper Conceptual Framework 
                 Source: Emrouznejad and De Witte (2010) 
31 
 
University of Ghana  http://ugspace.ug.edu.gh
This framework defines the issues and how to understand the processes of decision making 
units. The first and second phases (the concept and objectives, and the on structuring data) 
is the problem identification and research objectives captured in the first chapter of this 
study. The fourth and fifth phases show the outcome and documentation, which is dealt with 
by the analysis and discussions chapter of this work. Note that there is a link between the 
evaluation stage and the first phase. The evaluation must be closely linked to the concept 
and objectives of the study. The middle phases determines the model that is more suitable 
to best analyse the research issue. The interrelationship of the framework allows for 
feedback thus connecting phases. Note also that the first letter of each stage make up the 
acronym “COOPER”. On this background the Cooper Framework with respect to the 
current study are summarised below:  
 
2.4  Conclusion 
The chapter provided ample theoretical and empirical grounding for the study. 
Theoretically, established theories of Social Network Theory, Social Facilitation Theory 
and Institutional Theory were used to explain intra-group performance while game theory 
and resource-based theory were used for inter-group performance. Empirical work of 
frontier efficiency-related studies in the international oil industry as well as a general 
synopsis of efficiency studies that target IGOs were also presented. Based on these, the 
Cooper Framework is conceptually presented to aid in this research process. 
 
 
 
32 
 
University of Ghana  http://ugspace.ug.edu.gh
CHAPTER THREE 
CONTEXT OF STUDY 
3.0 Introduction 
Crude oil is a critical global resource in the economy of nations and an indispensable 
commodity in socio-economic, political and environmental priorities in the world today 
(Armaroli & Balzani, 2007; Asif & Muneer, 2007; Holdren, 2006; Igos et al., 2015). States 
establish agreement between them for socio-economic development but sometimes are 
unable to resolve the different  interest of the individual members which affect them; 
therefore the formation of supra-national political authorities – IGOs (Bennett & Oliver, 
2002; Park, 2015).  IGOs are important channels for enhancing socio cultural development, 
affinities, trust, improved economic development, trade, technology and information among 
member countries (Ingram, Robinson, & Busch, 2005; Park, 2015). This section will 
provide an overview of the major IGOs in the international oil industry. 
The demand and supply of oil affects the relationship between major oil exporting and 
importing countries (Yergin, 1991). The supply and sale of oil have created 
interdependencies  in the international oil industry and among nations (Strange, 1989). The 
study of IGOs have increased over the past centuries due to their importance to global 
economy and social interaction between states (Barnett & Finnemore, 1999; Cao, 2009; 
Volgy, Fausett, Grant, & Rodgers, 2008). IGOs may be categorized as political economic 
IGOs, social/cultural welfare IGOs, environmental IGOs, and governance and defence IGOs 
(Gomez & Parigi, 2013). The IGOs under study are political economic IGOs. These political 
economic IGOs address issues affecting trade, economic regulations and commence among 
member states (Alcacer & Ingram, 2013). These organisations bring together 
complementary skills and create platforms for innovation and creativity and make full use 
33 
 
University of Ghana  http://ugspace.ug.edu.gh
of the available resources to provide sustainable development for the member nations. This 
study also presents the relevance of IGOs in the international oil and gas industry.  
 
3.1 International Oil and Gas Supply and Demand 
Oil and gas have increased the multifaceted cross boarder relationship of countries in 
relations to production, consumption, regulatory influences, financial transactions and 
security of nations (Jones, 1995). Oil, one of the most important among other sources of 
energy, is critical to countries’ economies due to its universal use as fuel to feed 
manufacturing, industrial production, as well as the transportation sectors (Murphy & Hall, 
2011b).  Crude oil production and supply also have an impact on the global economy and 
countries’ socio economic development therefore countries are very interested in crude oil 
reserves. This section reviews the major countries in terms of proven reserves, production 
and consumption. 
 
3.1.1 Proven Oil and Gas Reserves 
Oil and Gas reserves constitute an important dimension that can determine the potential 
political power of a country in the international oil industry. This is simply the potential 
capacity of hydrocarbons buried inland that may be extracted in the future. Proven reserves 
are quantities of petroleum buried underground which by geological analysis and 
engineering data can be estimated with high confidence to be commercially recoverable 
from a certain time period, from known reservoirs and under current economic conditions 
(CIA, 2014). Therefore, the more reserves a country possess, the higher the potential of 
extracting them and the higher the political power this country has over other importing 
nations. Table 3.1 presents a summary of the top 11 countries in terms of crude oil proved 
reserves for 4 different years -2000, 2007, 2013 and 2014. Average values, also reported in 
34 
 
University of Ghana  http://ugspace.ug.edu.gh
the table, are for the entire period from 2000 to 2014. Regional statistics are also presented 
in the table. 
 Table 3.1: Crude Oil Proved Reserves (Billion Barrels) 
    2000 2007 2013 2014 Average 
1 Saudi Arabia 263.50 262.30 267.91 268.35 264.23 
       
2 Canada 4.93 179.21 173.11 173.20 142.48 
3 Venezuela 72.60 80.01 297.57 297.74 128.21 
4 Iran 89.70 136.27 154.58 157.30 126.09 
5 Iraq 112.50 115.00 141.35 140.30 119.65 
6 Kuwait 96.50 101.50 104.00 104.00 101.33 
7 United Arab Emirates 97.80 97.80 97.80 97.80 97.80 
8 Russia 48.57 60.00 80.00 80.00 60.38 
9 Libya 29.50 41.46 48.01 48.47 39.53 
10 Nigeria 22.50 36.22 37.20 37.14 32.25 
11 United States 23.17 22.31 33.40 36.52 25.02 
  World 1018.18 1318.00 1648.86 1655.56 1319.55 
 Middle East 675.64 739.20 802.16 803.60 738.34 
 North America 56.50 213.87 216.77 219.79 182.53 
 Central & South America 89.53 102.80 325.93 328.26 151.43 
 Africa 74.89 114.07 127.74 126.73 104.11 
 Eurasia 57.00 98.89 118.89 118.89 87.56 
 Asia & Oceania 43.99 33.37 45.36 46.01 39.76 
  Europe 20.64 15.80 12.02 12.28 15.81 
Source: EIA (2015) 
 
All values in Table 3.1 are measured in billions of barrels. Values from 2014 reveal that, 
out of the over 1.656 trillion barrels of crude oil reserves that have not been extracted in the 
world, Saudi Arabia possess more than 16% of these reserves making it the highest ranked 
nation. The country’s economy remains heavily dependent on petroleum since petroleum 
exports accounts for 85% of total export revenues in 2013 (OPEC, 2014). The country holds 
approximately 268 billion barrels of proved oil reserves. Although Saudi Arabia has about 
100 major oil and gas fields, more than half of its oil reserves are contained in eight fields 
in the northeast portion of the country (EIA, 2014). Looking at the average reserves since 
2000, Saudi Arabia (264.23bbls) is followed by Canada (142.48bbls), Venezuela 
(128.21bbls) and Iran (126.09bbls). Other notable mentions on this list are Libya and 
35 
 
University of Ghana  http://ugspace.ug.edu.gh
Nigeria, both of which are African countries, who placed 9th and 10th respectively. Finally, 
the United States has about 25.02 billion barrels of proven crude oil reserves buried 
especially in the Texas, North Dakota, Gulf of Mexico, Alaska and California (EIA, 2014). 
Close to half of the countries in this list are Middle Eastern countries – Saudi Arabia, Iran, 
Iraq, Kuwait and the UAE. It is therefore not surprising that most of the world’s proven 
reserves are found in this region. About 738.34 billion barrels of crude oil, on average, are 
buried in this location. Given the potential importance of this region, it is not surprising the 
several political instability in the region since ownership of land is a marker of a strong 
bargaining power in the world oil and gas supply (EIA, 2014). On the other side of the table 
is Europe with the smallest crude reserves of about 15.18 billion barrels. This means that 
Europe is expected to be a major importer of oil products from other regions. Values for 
natural gas reserves are presented in Table 3.2. 
Table 3.2: Proved Reserves of Natural Gas (Trillion Cubic Feet) 
    2000 2007 2013 2014 Average 
1 Russia 1700.00 1680.00 1688.00 1688.00 1683.73 
2 Iran 812.30 974.00 1187.00 1193.00 976.90 
3 Qatar 300.00 910.50 890.00 885.29 773.97 
4 Saudi Arabia 204.50 240.00 287.84 290.81 248.22 
5 United States 167.41 211.09 308.04 338.26 236.81 
6 United Arab Emirates 212.00 214.40 215.03 215.04 214.50 
7 Nigeria 124.00 181.90 182.00 180.74 165.40 
8 Venezuela 142.50 152.38 195.10 196.41 164.42 
9 Algeria 159.70 161.74 159.05 159.10 159.65 
10 Turkmenistan 101.00 100.00 265.00 265.00 147.07 
11 Iraq 109.80 112.00 111.52 111.52 111.03 
  World 5149.96 6190.88 6845.17 6972.52 6152.82 
 Middle East 1749.24 2566.04 2823.23 2812.83 2443.42 
 Eurasia 1977.00 2014.80 2177.80 2177.80 2039.65 
 Africa 394.57 485.81 514.81 605.96 475.86 
 Asia & Oceania 363.47 419.59 521.46 540.38 449.26 
 North America 261.34 283.59 393.43 422.06 315.03 
 Central & South America 222.66 240.75 268.92 277.62 256.15 
  Europe 181.69 180.30 145.52 135.87 173.46 
Source: EIA (2015) 
 
36 
 
University of Ghana  http://ugspace.ug.edu.gh
Unlike crude oil, majority of the world’s natural gas reserves are held by Russia (1683 cf). 
This is followed by Iran (976.90 Tcf), Qatar (773.97 Tcf), Saudi Arabia (248.22 Tcf) and 
United States (236.81 Tcf). Russia's economy is highly dependent on its hydrocarbons, since 
oil and natural gas revenues account for more than 50% of the federal budget revenues (EIA, 
2014). Most of the top crude oil reserve countries are still in the natural gas list, except that 
Qatar, and Turkmenistan are fresh additions when it comes to natural gas. Middle East 
continues to dominate natural gas reserves with about 2443.42 Tcf of natural gas which 
represents about 39.7% of the world’s value. Europe is also at the bottom of this list when 
regional blocks are compared. Figure 3.1 shows the trend of world oil and gas reserves from 
1980 to 2014. Note that Natural gas reserves are converted from trillions of cubic feets to 
billions of barrel equivalent in order to ensure comparison. The graph clearly shows that 
both resources have seen an upward spiral in volumes over the years. It is possible that the 
invent of more advanced technologies have increased the probability of discovery. Oil 
reserves are also seen to mostly out-number equivalent gas reserves globally over this 
period. 
1800
1600
1400
1200
1000
800
600
400
200
0
YEAR
Oil Reserves (Billion Barrels)
Natural Gas Reserves (Billion Barrels Equivalent)
 
Figure 3.1: Oil and Gas Reserve Trends 
  Source: EIA (2015) 
37 
 
BILLIONS OF BARREL EQUIVALENTS
University of Ghana  http://ugspace.ug.edu.gh
3.1.2 Oil and Gas Supply 
The supply and sale of oil and gas have created interdependencies  in the international oil 
industry and among nations (Strange, 1989). This affects the relationship between major oil 
exporting and importing countries (Yergin, 1991). The growth of most countries’ economies 
is highly related to the supply of oil (Murphy & Hall, 2011a). Table 3.3 presents a summary 
of the major oil supplying countries for the years 2000, 2007, 2013 and 2014. Also presented 
in the table are average values for the entire period from 2000 to 2014. Regional statistics 
are also reported. It must be noted that total oil supply takes into consideration crude oil, 
natural gas and other hydrocarbons from oil and gas extraction. The values of all the supplies 
are in thousands of barrels per day.  
Table 3.3: Total Oil Supply (Thousand Barrels Per Day) 
    2000 2007 2013 2014 Average 
1 Saudi Arabia 9475.75 10748.62 11701.51 11623.70 10729.72 
2 United States 9057.78 8469.40 12342.77 14020.82 9640.82 
3 Russia 6723.64 9938.18 10757.91 10847.10 9422.16 
4 China 3377.53 3953.86 4561.12 4598.05 3984.01 
5 Iran 3765.39 4039.03 3194.30 3376.61 3890.54 
6 Canada 2753.14 3448.52 4073.07 4383.32 3375.41 
7 Mexico 3460.09 3500.29 2915.07 2811.93 3337.80 
8 United Arab Emirates 2572.33 2947.50 3443.71 3473.71 2921.25 
9 Venezuela 3460.76 2681.99 2684.55 2684.55 2813.37 
10 Norway 3354.61 2564.89 1845.05 1904.38 2639.58 
11 Kuwait 2200.73 2603.43 2798.64 2767.20 2519.99 
  World 77725.45 85130.47 91014.47 93201.09 84960.21 
 Middle East 23484.22 25286.09 27472.12 27836.48 25482.40 
 North America 15271.01 15418.22 19330.91 21216.06 16354.03 
 Eurasia 8184.70 12687.58 13782.08 13905.19 11909.59 
 Africa 7989.70 10485.50 9296.44 8707.23 9447.15 
 Asia & Oceania 8316.46 8537.09 9199.47 9257.79 8697.19 
 Central & South America 7313.03 7270.57 8120.31 8408.39 7513.66 
  Europe 7166.34 5445.43 3813.14 3869.95 5556.19 
Source: EIA (2015) 
 
Values of the average oil supply for the period show that of the 84960.21 thousand barrels 
of oil supplied daily, Saudi Arabia supplies the highest volumes of 10729.72 thousand b/d. 
38 
 
University of Ghana  http://ugspace.ug.edu.gh
The United States follows in second place with 9640.82 thousand b/d even though the US 
is a major importer of oil  (EIA, 2013). Studying the averages, China who is also another 
substantial importer of oil comes forth with 3984.01 thousand b/d, after Russia (9422.16 
thousand b/d), whiles Iran (3890.54 thousand b/d) and Canada (3375.41 thousand b/d) 
follow thereafter. When the regional dynamics are carefully examined, it is obvious that the 
Middle East region as a whole has the highest supply volumes with 25482.40 thousand b/d, 
followed by North America (16354.03 thousand b/d), Eurasia (11909.59 thousand b/d), 
Africa (3944.15 thousand b/d) and at the bottom is Europe (5556.19 thousand b/d). Total 
supply may have substantial portions going to domestic markets, therefore the export 
volumes of both crude oil and gas productions are examined in Tables 3.4 and 3.5 
respectively. 
Table 3.4: Exports of Crude Oil (Thousand Barrels Per Day) 
    2000 2007 2013 2014 Average 
1 Saudi Arabia 6444.00 6968.74 7477.62 7657.93 6888.47 
2 Russia 3150.00 5171.58 4892.29 4807.16 4633.31 
3 Iran 2309.13 2617.63 2207.42 1401.93 2296.23 
4 Norway 3070.00 2012.96 1446.17 1324.11 2251.34 
5 Nigeria 2069.18 2120.22 2402.46 2410.57 2162.14 
6 United Arab Emirates 1870.00 2289.44 2365.56 2427.59 2113.94 
7 Canada 1590.77 1958.06 2338.78 2469.65 1901.58 
8 Venezuela 2094.30 2224.72 1253.00 1357.94 1777.36 
9 Iraq 2071.66 1617.92 2175.50 2427.72 1740.57 
10 Mexico 1763.67 1792.67 1364.73 1280.22 1712.24 
11 Kuwait 1316.81 1645.49 1776.96 1824.15 1520.90 
  World 40326.98 43699.35 43654.98 44104.95 42519.52 
 Middle East 16242.62 16771.29 18153.22 17806.84 16622.4 
 Africa 5941.996 7602.513 7395.989 8142.392 7134.349 
 Eurasia 3826.214 6861.941 7204.251 7119.682 6211.378 
 Europe 5475.625 3524.847 2513.65 2465.82 3893.751 
 North America 3457.058 3879.033 4037.696 4149.29 3767.773 
 Central & South America 3105.631 3409.887 2921.627 2997.834 3038.199 
  Asia & Oceania 2277.831 1649.84 1428.556 1423.092 1851.666 
Source: EIA (2015) 
 
39 
 
University of Ghana  http://ugspace.ug.edu.gh
Most oil producing countries depend on the export of the product for the growth of their 
economy. Table 3.4 presents the total crude oil export of major oil producing countries for 
2000, 2007, 2013 and 2014 and statistics from the regional blocks. The values of the average 
crude oil exported are in thousands of barrels per day. Of the 42519.52 thousand barrels 
exported throughout the world daily, Saudi Arabia is the highest exporter, distributing about 
6888.47 thousand b/d especially to the Asia and Pacific areas  (OPEC, 2015). It supplies 
about 4417 thousand barrels per day to the Asia and Pacific region, 1251 thousand b/d to 
North America and 952, 273 and 191 thousand barrels daily to the Europe, Middle East and 
Africa respectively (OPEC, 2015) The remainder goes to several other locations globally. 
The country’s export accounted for 85% of the total earning in 2013 (EIA, 2014). Russia’s 
exports of 4633.31 thousand b/d accounts for about 50% of the country’s GDP. Iran comes 
third with 2296.23 thousand b/d. The world crude oil export is dominated by the Middle 
East with countries like Saudi Arabia and Iran exporting (9184.78 thousand b/d). They are 
followed by Africa making up 7134.349 thousand b/d. The Asia and Oceania region is the 
least exporter of crude oil (1851.666 thousand b/d).  
  
40 
 
University of Ghana  http://ugspace.ug.edu.gh
Table 3.5: Exports of Natural Gas (Billion Cubic Feet) 
    2000 2007 2013 2014 Average 
1 Russia 6590.49 8232.99 7176.01 7801.08 7378.55 
2 Canada 3575.50 3843.83 3117.96 2911.72 3530.07 
3 Norway 1727.26 3011.66 3878.65 3629.89 2932.40 
4 Algeria 2213.54 2061.34 1771.40 1518.55 2009.02 
5 Qatar 495.82 1536.20 4400.25 4432.03 1977.09 
6 Netherlands 1462.68 1965.92 2133.34 2375.57 1937.41 
7 Turkmenistan 1380.82 1744.56 1624.49 2147.15 1486.00 
8 Indonesia 1279.11 1295.71 1241.32 1105.36 1289.86 
9 Malaysia 780.11 1107.48 1157.27 1250.15 1018.13 
10 United States 244.00 822.00 1619.00 1572.00 915.07 
11 Australia 362.05 688.36 924.79 1150.00 631.13 
  World 22978.03 33262.21 37409.38 38301.79 31107.92 
 Eurasia 8571.66 10927.49 9809.41 11077.96 9724.80 
 Europe 4050.77 6232.14 7961.59 7892.69 6182.49 
 North America 3827.98 4716.43 4739.82 4484.85 4459.25 
 Asia & Oceania 2881.70 3848.66 4233.10 4465.83 3665.44 
 Africa 2444.15 3883.52 3561.41 2996.87 3225.82 
 Middle East 841.56 2487.51 5619.39 5767.19 2851.40 
  Central & South America 360.21 1166.45 1484.64 1616.40 1049.77 
Source: EIA (2015) 
 
Table 3.5 shows the major exporters of natural gas. Gas represents a substantial source of 
energy production in the world. Most countries producing this product export it for 
improvement of their economies and for social economic development. This table shows 
the top gas exporting countries in the years 2000, 2007, 20013 and 2014. The values are 
reported in billions of cubic feet. The average total world export stands at 31107.92 bcf with 
Russia being the highest exporter (7378.55 bcf) after which Canada (3530.07) Norway 
(2932.40 bcf) and Algeria (2009.02 bcf) follow by Qatar (1977.09 bcf) from Middle East 
comes 5th with the United States (915.07 bcf) and Australia (631.13bcf) placing 10th and 11th 
respectively. 
The export of gas by regions is dominated by Eurasia accounting for (31107.92 bcf) 
followed by Europe (9724.80 bcf) and the North America (4459.25 bcf). Africa, Middle 
41 
 
University of Ghana  http://ugspace.ug.edu.gh
East, Central and South America represent the least exporting regions with less than 3500 
bcf each. 
 
3.1.2 Oil and Gas Consumption 
Increase in oil and gas consumption is highly linked with economic growth  (Murphy & 
Hall, 2011b). About half of economic changes in most countries can be explain by oil 
consumption (Cleveland et al., 1997). Countries with high GDP have been noted to have a 
high correlation with increases in oil consumption (Murphy & Hall, 2011a). Table 3.6 
reveals the pattern of the world’s largest consumers of petroleum in 2000, 2007, 2012 and 
2013. All values represented here are in thousands of barrels per day.   
Table 3.6: Total Petroleum Consumption (Thousand Barrels Per Day) 
    2000 2007 2012 2013 Average 
1 United States 19701.1 20680.4 18490.2 18961.1 19759 
2 China 4795.71 7479.92 10175.1 10480 7139.47 
3 Japan 5480.14 5009.22 4697.33 4556.81 5000.42 
4 Russia 2578.5 2885.1 3445.1 3493 2891.1 
5 India 2147.44 2888.06 3617.85 3660 2791.68 
6 Germany 2766.76 2406.69 2389.13 2435.08 2576.51 
7 Brazil 2121.28 2296.55 2922.93 3003 2343.19 
8 Canada 2007.72 2389.48 2402.8 2374.45 2236.61 
9 Korea, South 2135.33 2240.48 2321.62 2328.3 2195.31 
10 Saudi Arabia 1537.1 2094.33 2881.65 2961 2111.7 
11 Mexico 2095.93 2172.79 2101.38 2090.45 2092.78 
  World 76927.6 86788.2 90391.8 91253.2 84523.5 
 Asia & Oceania 20871.9 25474.6 29761.9 30123.7 25082.8 
 North America 23812.5 25255.2 23006.5 23438.2 24051.1 
 Europe 15924.3 16245.8 14409.8 14233.2 15716.9 
 Middle East 4896.88 6523.07 7996.98 8083.49 6463.56 
 Central & South America 5191.71 5918.85 6963.32 7085.22 5952.34 
 Eurasia 3719.74 4259.88 4644.3 4688.6 4162.05 
  Africa 2510.57 3110.9 3608.9 3600.71 3094.73 
 
 
42 
 
University of Ghana  http://ugspace.ug.edu.gh
Two of the largest economies in the world top with United States, the world largest 
economy, leading with 19759.04 thousand barrels per day, followed China’s 7139.47 
thousand barrels per day. China’s average consumption levels is about 36.13% of that the 
United States. However, a closer look at the values by year show that China’s consumption 
level is gradually closing up to that of the US. Given the populous nature of these two 
countries, it is not exactly surprising that they have higher consumption levels. Not also 
surprising is the fact that Russia and India are all among the top consumers of petroleum 
due to their land mass and population. The regional representation by oil consumption is 
largely dominated by Asia and Oceania region accounting for 25082.8 thousand barrels per 
day. This is followed by North America with 24051.1 thousand barrels per day. Europe 
consumption of 15716.9 thousands barrels per day is also not surprising given this region’s 
relatively lower possession of oil and gas reserves as seen in earlier sections of this chapter.  
In Table 3.7, the major consumers of natural gas around the world are presented. Again the 
United States is the highest in this regard accounting for 23323.79 bcf out of the world’s 
total average of 108458.90 bcf. Iran accounts for 3985.98 bcf after Russia (14563.48 bcf). 
United Kingdom, Germany and Italy all Europeans are all among the substantial consumers 
of gas, albeit in the bottom half with an average of 19168.41 bcf. 
  
43 
 
University of Ghana  http://ugspace.ug.edu.gh
Table 3.7: Dry Natural Gas Consumption (Billion Cubic Feet) 
    2000 2007 2013 2014 Average 
1 United States 23333.00 23104.00 25538.00 26168.00 23323.79 
2 Russia 13058.78 15180.86 15710.94 15598.99 14563.48 
3 Iran 2220.96 3992.01 5553.95 5555.58 3985.98 
4 Japan 2913.91 3748.09 4471.94 4492.35 3574.33 
5 United Kingdom 3373.29 3244.04 2752.06 2734.90 3209.28 
6 Canada 2991.36 3050.51 3541.28 3654.78 3176.84 
7 Germany 3098.11 3142.61 3001.21 3123.44 3162.80 
8 Italy 2498.36 2998.14 2645.62 2474.49 2762.09 
9 Saudi Arabia 1759.04 2628.14 3507.84 3532.56 2631.55 
10 China 867.05 2489.71 5073.92 5760.12 2615.09 
11 Ukraine 2779.29 2906.42 1861.10 1659.81 2505.40 
  World 87236.66 105545.38 119696.29 121357.10 108458.90 
 North America 27722.55 28179.40 31500.87 32103.43 29876.56 
 Eurasia 19471.53 22543.42 22243.47 21944.56 21334.42 
 Europe 17394.37 19873.13 18685.59 18511.95 19168.41 
 Asia & Oceania 10484.35 16871.03 22788.93 23626.78 16358.48 
 Middle East 6822.15 10674.04 14740.02 15078.16 10733.84 
 Central & South America 3303.69 4387.37 5245.73 5518.78 4353.89 
  Africa 2038.03 3016.99 4491.68 4573.43 3179.43 
Source: EIA (2015) 
 
North America is the highest with respect to regional consumption of 29876.56 bcf of natural 
gas. Asia and Oceania (16358.48 bcf), Middle East (10733.84 bcf), Central and South America 
(4353.89 bcf) and Africa (3179.43 bcf) are among the least consumers of natural gas. 
 
3.2 Intergovernmental organizations in the oil industry 
Countries in the international oil industry in order for them  to stay productive and 
competitive  have coalesced into intergovernmental organisations (IGOs), with a goal of 
developing fast-based approach in delivering high quality project outcomes, fostering 
innovations and helping develop solutions for operational and economic issues at regional 
and global levels (Belyi & Talus, 2015; Cissokho et al., 2013; Dorussen & Ward, 2008; 
Escobar & Le Chaffotec, 2015; Holland, 1998). These organisations bring together 
complementary skills and create platforms for innovation, cooperation and creativity, and 
44 
 
University of Ghana  http://ugspace.ug.edu.gh
make full use of the available resources to provide sustainable development for the member 
nations (Dorussen & Ward, 2008; Holland, 1998). The participation of States in IGOs is to  
shape their policy in various ways; for example education (Bradley & Ramirez, 1996; 
Meyer, Ramirez, & Soysal, 1992; Schafer, 1999), environmental protection (Frank, 1997, 
1999; Frank, Hironaka, & Schofer, 2000), science (Finnemore, 1993) etc. As the oil and gas 
industry continues to play a major role in meeting the world’s growing economic demands, 
there have been a number of such IGOs with policy direction towards the oil industry. In 
the international Oil and Gas industry the four IGOs at the forefront of forging global policy 
are OPEC, OAPEC, FSU and IEA. Without IGOs, it will be difficult to monitor 
performance over time, set up international regulations and restrictions for countries as 
IGOs help to create and bind international administrations (Biermann & Bauer, 2004). This 
section presents a brief overview of each of these IGOs. 
 
3.2.1 Organization of the Petroleum Exporting Countries (OPEC) 
Organization of the Petroleum Exporting Countries (OPEC) established in September 1960, 
is a permanent intergovernmental organization headquartered in Vienna, Austria, created at 
the Baghdad Conference (OPEC, 2012). Originally made up of 5 members1, the five 
founding members were later joined by nine2 other members. Currently, OPEC comprises 
of 12 members, namely Algeria, Angola, Ecuador, Iran, Iraq, Kuwait, Libya, Nigeria, Qatar, 
Saudi Arabia, United Arab Emirates (UAE) and Venezuela. The main objective of this IGO 
is to co-ordinate and unify petroleum policies among member countries in order to secure 
fair and stable prices for petroleum producers (OPEC, 2012). The organization aims to 
ensure an efficient, economic and regular supply of petroleum to consuming nations; and a 
                                                          
1 OPEC founding members - (5) Iran, Iraq, Kuwait, Saudi Arabia, and Venezuela. 
2 OPEC other members - (9) Qatar, Indonesia, Libya, UAE, Algeria, Nigeria, Ecuador, Angola, Gabon. 
45 
 
University of Ghana  http://ugspace.ug.edu.gh
fair return on capital to those investing in the industry (OPEC, 2012). Membership is 
automatic for the five founding members, but opened for any other country with a 
substantial net export of crude petroleum, which has fundamentally similar interests to those 
of founding member countries. A new member state is accepted into the organization by a 
majority of three-fourths of full member countries by the Conference (OPEC, 2012). The 
Organization functions through three organs: The Conference, Board of Governors; and the 
Secretariat. The organization is regulated by statute approved in the conference of 1961 
which is the highest decision body(Goldthau & Witte, 2011). The membership of OPEC is 
diversely represented in various characteristics. From the regional stand point they can be 
grouped into (a) American countries (Ecuador & Venezuela), (b) North Africa (Algeria & 
Libya), (c) South and West Africa (Angola & Nigeria) and (d) Middle East (Iran, Iraq, Saudi 
Arabia, Qatar & UAE) (Al-Rashed & León, 2015). OPEC is one of the most important or 
linchpin in the global oil market accounting for crude oil production of 41 per cent (that is 
30.7m b/d) of the world’s average production and with proven crude oil reserves as at 2014 
at 1,206billion barrels, which is 80.8 per cent of the 1,492.9 billion barrels of the World 
proven crude oil reserves (BP, 2014; OPEC, 2015; Sari & Soytas, 2009). Additionally, 
OPEC’s export revenue in 2014 was $730 billion (EIA, 2015) With a natural gas proven 
reserve of 95,129bn standard cubic meters, OPEC represents 47.3 per cent of the world 
proven gas reserves (OPEC, 2015). The organization phases a challenge of member states 
not respecting the output policies established (Goldthau & Witte, 2011) OPEC continues to 
actively engage in international cooperation and dialogue which has become an important 
industry event an effort in exchanging views and outlooks with other energy stakeholders 
(OPEC, 2014). 
46 
 
University of Ghana  http://ugspace.ug.edu.gh
3.2.2 Organization of Arab Petroleum Exporting Countries (OAPEC) 
Organization of Arab Petroleum Exporting Countries (OAPEC) is a regional inter-
governmental organization concerned with the development of the petroleum industry by 
fostering cooperation among its members (OAPEC, 2015). OAPEC was establish on 9th 
January 1968 in Beirut by three3 Arab states with the headquarters in Kuwait. By 1982 the 
membership of the Organization had risen to eleven Arab oil exporting countries, namely 
Saudi Arabia, Kuwait, Libya, United Arab Emirates, Qatar, Bahrain, Algeria, Syrian, Iraq, 
Egypt, and Tunisia.  
OAPEC contributes to the effective use of the resources of member countries through 
sponsoring joint ventures. The main goals of OAPEC are: member states’ cooperation in 
various forms of the economic activity, development of close links between them, 
safeguarding the legitimate interests of its members under fair and reasonable terms in the 
oil industry (OAPEC, 2015). The Organization is guided by the belief of building an 
integrated petroleum industry as a cornerstone for future economic integration among Arab 
countries (OAPEC, 2015).   
The Organization functions through four organs (OAPEC, 2015):  The Council of Ministers, 
which is the supreme authority responsible for its policies and rules. It appoints the Secretary 
General and Assistant Secretaries. The next is the Executive Bureau which assists the 
Ministerial Council in supervising the Organization’s affairs and recommendations to the 
Council on matters related to articles of the Agreement and the execution of the 
Organization’s activities. The Executive Bureau is composed of one representative from 
each of the member countries.  The chairmanship rotates annually in the order followed by 
the Ministerial Council. OAPEC’s oil reserves have been estimated at about 713 billion 
                                                          
3 OAPEC founding members - (3) Saudi Arabia, Kuwait & Libya. 
 
47 
 
University of Ghana  http://ugspace.ug.edu.gh
barrels about 55.2% of the proven world’s reserves in 2014 (OAPEC, 2015). The production 
of crude oil, despite OAPEC holding over half of the proven oil reserve, is relatively 
moderate at 22.9m b/d about 31.2 per cent of the worlds of 73.4m b/d. oil product 
consumption of OAPEC countries rose from 3.8m 6.8m in 2014. OAPEC now represents 
an international specialised organisation which supports the cooperation between oil 
producers and the implementation of common projects connected to the regional integration. 
 
3.2.3 The Former Soviet Union (FSU) 
The Post-Soviet States, also collectively known as the Former Soviet Union (FSU) are the 
15 independent states4 that emerged from the Union of Soviet Socialist Republic (USSR) in 
its dissolution in December 1991 (Minescu, Hagendoorn, & Poppe, 2008). The dissolution 
of the Soviet Union took place as a result  of general economic stagnation, even regression 
the inter-republic economic connections, leading to even more serious breakdown of the 
post-Soviet economies (Easterly & Fischer, 1994). Most of the formerly Soviet states began 
the transition to a market economy in 1990-1991 and made efforts to rebuild and restructure 
their economic systems, with varying results  (Podkorytova & Raskina, 2014). The process 
triggered a severe transition decline, with Gross Domestic Product (GDP) dropping by more 
than 40% between 1990 and 1995 (Podkorytova & Raskina, 2014)).This decline in GDP 
was much more intense than the 27% decline that the United States suffered in the wake of 
the Great Depression between 1930 and 1934 (Podkorytova & Raskina, 2014). The 
reconfiguration of public finance in compliance with the principles of market economy 
resulted in dramatically reduced spending on health, education and other social programs, 
leading to a sharp increase in poverty. Although the FSU is not a traditional IGO like OPEC 
                                                          
4 FSU members - (15) Armenia, Azerbaijan, Belarus, Estonia, Georgia, Kazakhstan, Kyrgyzstan, Latvia, 
Lithuania, Moldova, Russia, Tajikistan, Turkmenistan, Ukraine, Uzbekistan 
48 
 
University of Ghana  http://ugspace.ug.edu.gh
and OAPEC, the close historical links, the share of oil and gas reserves and production and 
collaboration in trade have made their influence critical in such comparison. The FSU holds  
a sizeable quantity of natural resources in oil and gas that can be economically produced to 
meet the global demand for energy (Aguilera, 2012). The member states of FSU dependent 
on this enormous natural resource endowment to build their economy (Aguilera, 2012).  The 
distribution of gas shows that the FSU has the largest natural gas resource by region in the 
world, holding 3,716 (TCFG) and 30% of stranded gas reserve, followed by the Middle East 
and USA. The three regions together have about 70% of the global gas reserves(Aguilera, 
2012). The FSU is second in oil reserve endowment after the Middle East with 118,886mb 
(OPEC, 2015). The FSU holds about 16.7 % of regional distribution of conventional oil 
reserves and with global production of 12,646.7 bd representing 17.2% (OPEC, 2015). 
Together the FSU region and the Middle East controls about 2/3 of the global conventional 
oil (Rogner et al., 2012).  In view of their importance OPEC and EIA recognize their 
influence in their annual statistical bulletin. 
 
3.2.4 The International Energy Agency (IEA) 
The International Energy Agency (IEA) was establish as an autonomous organization of the 
OECD after an agreement of the International Energy Program (IEP) in 1974 with 
membership of sixteen countries and its secretariat in Paris (Colgan, Keohane, & Van de 
Graaf, 2012). Currently IEA membership stands at 295. They work to ensure reliable, 
affordable and clean energy for its member countries and most importantly for oil importing 
countries and beyond. The IEA was founded in response to the 1973-1974 oil crisis, the 
IEA’s initial role was to help countries co-ordinate a collective response to major disruptions 
                                                          
5 IEA members - (29) Australia, Austria, Belgium, Canada, Czech Republic, Denmark, EC, Finland, France, 
Germany, Greece, Hungary, Ireland, Italy, Japan, South Korea, Luxembourg, Netherlands, NZ, Norway, 
Poland, Portugal, Slovakia, Spain, Sweden, Switzerland, Turkey, UK, US 
49 
 
University of Ghana  http://ugspace.ug.edu.gh
in oil supply through the release of emergency oil stocks to the markets (IEA, 2014). While 
this continues to be a key aspect of its work, the IEA has evolved and expanded. It is at the 
heart of global dialogue on energy, providing authoritative statistics, analysis and 
recommendations. Today, the IEA’s four main areas of focus are: Energy security: 
Promoting diversity, efficiency and flexibility within all energy sectors; Economic 
development: Ensuring the stable supply of energy to IEA member countries and promoting 
free markets to foster economic growth and eliminate energy poverty;  Environmental 
awareness: Enhancing international knowledge of options for tackling climate change; and  
Engagement worldwide: Working closely with non-member countries, especially major 
producers and consumers, to find solutions to shared energy and environmental concerns 
(IEA, 2014). An important part of the Agency’s program involves collaboration in the 
research, development, and demonstration of new energy technologies to reduce excessive 
reliance on imported oil, increase long-term energy security, and reduce greenhouse gas 
emissions. The IEA is made up of 29 member countries. Membership is based on four basic 
criteria. Before becoming a member country of the IEA, a candidate country must 
demonstrate that it has (a) as a net oil importer, reserves of crude oil and/or product 
equivalent to 90 days of the prior year’s average net oil imports to which the government 
(even if it does not own those stocks directly) has immediate access should there be 
activation of the Co-ordinated Emergency Response Measures (CERM) – which provide a 
rapid and flexible system of response to actual or imminent oil supply disruptions (b) a 
demand-restraint programme for reducing national oil consumption by up to 10% (c) 
legislation and organisation necessary to operate, on a national basis, the CERM, and (d) 
Legislation and measures in place to ensure that all oil companies operating under its 
jurisdiction report information as is necessary (IEA, 2014). 
50 
 
University of Ghana  http://ugspace.ug.edu.gh
To fully understand the energy markets, requires bringing on board the concerns, challenges 
and the solutions being considered by all the important actors in the industry (IEA, 2014). 
Collaborating with partner countries and others gives a wide array of activities, such as, 
holding workshop jointly on specific issues such as energy efficiency, policies on 
emergency response, co-operating on in-depth surveys of specific energy sectors in partner 
countries. 
The IEA also works with other international organisations and forums in the energy field. It 
engages in an active discussions with producer countries and other IGOs in the oil and gas 
industry particularly at the International Energy Forum (IEF) (IEA, 2014). In addition, the 
IEA collaborates with the International Renewable Energy Agency (IRENA) and engages 
with partner countries and other international agencies to provide all stakeholders including 
business leaders a true global perspective of the world’s energy system. 
 
3.3 Conclusion 
This chapter describes the relevance of IGOs and their roles in the international oil and gas 
industry which informs the global production and supply of the natural resources. The 
chapter also presents some statistics of major producer and consumer countries globally. 
 
 
51 
 
University of Ghana  http://ugspace.ug.edu.gh
CHAPTER FOUR 
METHODOLOGY 
4.0 Introduction 
This chapter looks at research design adopted for the study and data and data sources. This 
is followed by an overview of the methods of assessment of oil and gas producing countries. 
The aim of the chapter is to provide an overview of the analytical framework and techniques 
that guide the research. 
 
4.1 Research design 
The study is quantitative in nature which provides an objective assessment and testing of 
theories and the relationship between variables which can be measured through statistical 
methods (Creswell, 2013). The quantitative approach to research adopted in this study is 
guided by a positivist philosophical paradigm. This paradigm provides an objective view to 
data collection and estimation procedure since it detaches itself from research subjects 
(Creswell, 2013; Wilson, 2010). The study uses a panel data to draw inferences to answer 
the research questions. By employing panel data to explore efficiency of oil producing 
countries (OPCs) of the four IGOs in the international oil industry, this study provides 
findings that are not biased by cross-sectional or firm-specific (time series) restrictions. The 
study covers the period from 2000 to 2013. The study relies on a secondary data sourced 
from the US Energy Information Administration (EIA) database on OPCs and World Bank’s 
World development indicators which is assessed using various Data Envelopment Analysis 
(DEA) techniques. The adopted research design is experimental in nature. This enables 
comparison of groups with high standard of internal validity as this study seeks to examine 
52 
 
University of Ghana  http://ugspace.ug.edu.gh
the efficiency and the frontier differences within and between the various groups of OPCs 
(Bhattacherjee, 2012; Creswell, 2013). 
4.2 Data for modelling 
The study is based on population of all OPCs under the period of review from 2000 to 2013. 
Although 2014 data is available, there are several missing values for some variables and is 
therefore not included. Although various countries may specialise in either oil resources or 
natural gas exploration and extraction, to provide a holistic assessment, the study is 
interested in countries who hold both oil and gas resources. The governments of most of the 
OPCs own oil blocks and are involved in either of the two segments, that is, the upstream 
or downstream segments of the oil and gas industry. The upstream is for the exploration and 
production of oil and gas from the reserves whilst the downstream deals more with 
processing, bulk storage, distribution, and marketing. This study is interested in only the 
upstream segment.  
Whiles all oil and gas producing countries are eligible for selection, based on the research 
objectives, the population of the study comprises the 65 oil and gas producing countries of 
the four IGOs - OPEC (12 countries), IEA (28 countries), FSU (15 countries) and OAPEC 
(10 countries). Of these 65 members of the four IGOs understudy, 62 countries produce oil 
and 54 countries extract gas. Only 53 members are producers of both oil and gas. This is 
clearly presented in Table 4.1. Therefore, the study is based on the 53 OPCs who produce 
both oil and gas. It is important to note that some OPCs belong to two IGOs, but when the 
duplicate OPCs, especially in OPEC and OAPEC are excluded from the sample, the 
effective number of IGOs become 46 unique countries. 
  
53 
 
University of Ghana  http://ugspace.ug.edu.gh
Table 4.1 : Membership distribution of IGOs 
IGO Membership Oil Producing Gas Processing Oil & Gas 
OPEC 12 12 12 12 
IEA 28 27 22 21 
FSU 15 13 10 10 
OAPEC 10 10 10 10 
Total 65 62 54 53 
 
The primary data source for this study is the US Energy Information Administration (EIA). 
This institution gathers data on the oil and gas activities of all countries globally. 
Information on Production, Reserves and Consumption of Petroleum, Natural Gas, 
Electricity and other renewables collated from the period from 1980 to 2014 are available 
in this database. Additionally, the World Development Indicators database of World Bank 
is used to gather mostly labour statistics.  
 
4.3 Efficiency modelling and other considerations 
In recent years, considerable methods have been used to examine and compare frontier 
efficiency performance of different countries and groups. The two widely used methods to 
estimate frontier efficiency of decision making units (DMUs) are the nonparametric 
constant return to scale  (CRS) Data Envelopment Analysis (DEA) (Charnes, Cooper, & 
Rhodes, 1978a), which was later extended by (Banker, 1984), and the parametric Stochastic 
Frontier Analysis (Aigner, Lovell, & Schmidt, 1977) which was later extended to variable 
return to scale (VRS). This study relies on the DEA approach because unlike the SFA, it 
does not require several impractical model specifications on the production frontier as seen 
in the SFA approach (Jacobs, 2001). DEA is a nonparametric mathematical programming 
method based on a linear programming model used to estimate and compare the efficiency 
54 
 
University of Ghana  http://ugspace.ug.edu.gh
of organisations or DMUs given their available resources (inputs) to create a set of outcomes 
(outputs) relative to other units (Ramanathan, 2007). DEA is a technique which depends on 
input-output data to construct a production frontier which is used as the benchmark to 
evaluate other units in the sample.  
DEA was introduced by Charnes et al. (1978a) to compare efficiencies of organisations.  
Their approach was based on a constant return to scale (CRS) production frontier which 
holds the assumption of full proportionality of inputs and output relationship. Banker, 
Charnes, and Cooper (1984) extended this methodology using the variable returns to scale 
(VRS) frontier, which holds that the proportionality assumption is not always upheld. In the 
model, the DMUs with the best efficiency in converting inputs into output are identified as 
the best and the other DMUs are ranked relative to the most efficient DMUs. Arguably, one 
key benefit of this approach for relative assessment of DMUs is because it can estimate 
efficiencies even when a particular DMU has multiple inputs and outputs. 
Often the interest in evaluating the efficiency and performance of DMUs is constrained by 
substantial differences in geographical boundaries or group dynamics of the DMUs under 
consideration. The metafrontier and global frontier approaches allow for better handling of 
these group specific heterogeneities in the efficiency assessment. This study relies on these 
two frontier frameworks to measure and compare the efficiency of countries in different 
IGOs in the oil and gas industry. The next section presents an overview of these frontier 
approaches.  
 
4.3.1 Metafrontier analysis 
Metafrontier in DEA has been attributed to the works of Battese et al. (2004); Battese and 
Rao (2002b); O’Donnell et al. (2008b). It was later extended by De Witte and Marques 
(2009). However the idea is based on the concept of meta-production function by Hayami 
55 
 
University of Ghana  http://ugspace.ug.edu.gh
and Ruttan (1971) (Battese et al., 2004; O’Donnell et al., 2008b). The meta-production 
function has some advantages but the lack of comparable data and the presence of inherent 
differences across groups was the major limitation of this approach (Battese et al., 2004). 
Therefore, the metafrontier approach fills in these disadvantages by allowing comparison 
across heterogeneous groups (Battese & Rao, 2002a; O’Donnell et al., 2008b). Oil 
producing countries are in different groups and regions and are faced with different 
production capabilities such as the quality of physical and human capital, economic 
infrastructure and resource endowment. Metafrontier, most importantly, envelops all group 
frontiers. In other words, this frontier combines all firms irrespective of which group they 
belong to. It then measures efficiencies relative to the metafrontier and the group frontier 
and can be decomposed into two components: A component that measures the distance from 
an input – output point to the group frontier, and another component that measures the 
differences between the group frontier and the metafrontier. Therefore, given y and x 
nonnegative real output and input vectors, the meta-technology set is defined as:                
                         T  (x, y) : x  0; y  0; x  can  be  used  to   produce    y               (1) 
 The output-oriented technical efficiency (Meta efficiency) of a DMU relative to the 
M
metafrontier (TE0 (x, y))  based on a CRS assumption, can be defined as: 
TEM0 (x, y)  Max 
        s.t :
n
          yrj j yro           r 1  ,...,  p                                          (2) 
j1
n
         xij j  xio            i 1  ,...,  m
j1
                 j  0                 j 1  ,...,  n
56 
 
University of Ghana  http://ugspace.ug.edu.gh
M
Where TE0 represent the technical efficiency with respect to metafrontier of a DMU. A 
M
firm is said to be technically efficient if TE0 = 1. In the model in Eqn. (2), xij  denotes the 
amount of the ith input used by the jth DMU. y rj  is the amount of the r
th output produced by 
the jth DMU. Note that, in the sample, we have m number of inputs and p number of outputs 
for a set of n number of DMUs. Also j  is the intensity weight defining the convex 
combination of the best practice units that are compared with the jth unit.   is equivalent to 
TE M0 which measures the maximum percentage of expansion of the output of a particular 
DMU necessary to make that DMU efficient. 
The existence of sub-groups in the meta-technology can be defined and assessed. Here, we 
now assume that the set of DMUs can be divided into K groups (where K > 1). The 
production technology defined in Eqn. (1) can be redefined for the kth group as:  
 
 T
k  (x, y) : x  0; y  0;   x can be used by DMUs in group k to produce y         
(3) 
k
In a similar way, a group-specific technical efficiency, TE0 (x, y) , can be formulated for a 
DMU this time relative to its group frontier. This is defined in Eqn. (4) as: 
TE k0 (x, y)  Max 
        s.t :
n
         y k krj . j  yro           r  1  ,...,  p                                          (4) 
j1
n
         x k ij .k j  .xio            i  1  ,...,  m
j1
                 k j  0                 j  1  ,...,  n
57 
 
University of Ghana  http://ugspace.ug.edu.gh
Based on these two efficiency scores- meta efficiency and the group efficiency, a technology 
gap ratio can be computed. According to Battese et al. (2004) the output oriented technology 
gap ratio of a group k , TGR
k
0 (x, y) , can be defined as the ratio of the Meta technical 
efficiency to the group technical efficiency.  
k TE
M (x, y)
TGR0 (x, y) 
0                                                         (5) 
TEk0 (x, y)
The ratio takes value between zero and one and measures the diversion from the metafrontier 
(available technology irrespective of group) due to membership of a particular group k. In 
other words, how far back, or close, is a particular firm due to membership of a particular 
group k. This shows that the meta technical efficiency can be decomposed as follows: 
TEM0 (x, y)  TGR
k
0 (x, y)TE
k
0 (x, y)                                      (6) 
The meta-efficiency estimated relative to the meta frontier which represents the existing 
state of knowledge irrespective of group can therefore be decomposed into the product of a 
group-specific technical which represents the existing state of knowledge and the physical, 
social and economic environment that characterizes that particular group k and the 
technological gap ratio (meta-technology ratio) for group-k which provides a measure of 
how close the group-k frontier is to the metafrontier (O’Donnell et al., 2008a). 
To illustrate the Metafrontier approach, a hypothetical data of 9 DMUs who belong to three 
IGOs (OPEC, IEA and FSU) have been presented in Table 4.2. Each firm uses 1 input 
(reserves) to produce 2 outputs (Oil and Gas).  
 
 
58 
 
University of Ghana  http://ugspace.ug.edu.gh
Table 4.2: Data of Hypothetical OPCs and IGOs 
 
       OIL       GAS       
DMUs IGO PRODUCED PRODUCED    R   E   S  ERVES 
O1 OPEC 9 3 1 
O2 OPEC 6.5 4 1 
O3 OPEC 6 2 1 
I1 IEA 7.5 3.5 1 
I2 IEA 5 4 1 
I3 IEA 5.5 5.5 1 
F1 FSU 4 5 1 
F2 FSU 7 3 1 
F3 FSU 2 3 1 
 
To graphically illustrate this dataset, the outputs have been normalised by the inputs. This 
can be depicted in Figure 4.1. 
6
I3 OPEC  
FSU  
F1 IEA  
5 Meta Frontier 
I2 O2
4
I1
F3 F2 O
3
1 
O3
2
1
0
0 1 2 3 4 5 6 7 8 9 10
OIL
 
Figure 4.1: Graphical Illustration of Hypothetical Data 
59 
 
GAS
University of Ghana  http://ugspace.ug.edu.gh
Here, while round bullets represents the IEA group, the group specific frontiers of OPEC 
and FSU are square and triangular bullets respectively. Finally, all DMUs can be enveloped 
into a meta frontier represented by the black dashed line. The output oriented group 
efficiency of a DMU can be estimated as a radial projection of the DMUs on the frontier 
and the current location of the DMU. For example, the group-specific efficiency of DMU 
O3 can be estimated as 9/6 = 1.5. Where 9 is the projected location and 6 is the current 
location. Although these values are picked on the X-axis, values on the Y-axis can also be 
used. The efficiency score of the same DMU using the Y-axis can be estimated as 3/2 = 1.5. 
This holds because of the proportionality condition. Note also that, for DMU O3, its meta 
efficiency score will also be 1.5 since at the target point, the group frontier is the same as 
the meta frontier. The technology gap ratio (TGR) of DMU O3 can therefore be estimated 
based on notations in Equation (5) as:  1.5/1.5 =1. Which means that O3 experiences no 
difference in production capabilities by belonging to the OPEC group. It therefore makes 
100% use of all technological spillovers. The VRS linear programming model for DMU O3 
for its group specific frontier and metafrontier are presented below: 
Output-oriented VRS model for DMU O3 relative to OPEC frontier 
TEOPECO3 (x, y)  Max 
       s .t
    9o1  6.5o2  6o3  6
    3o1     4o2  2o3  2    
      o1       o2    o3 1   
      o1       o2    o3 1 
                              λ j  0
 TEOPECO3 (x, y)  1.5
 
 
  
60 
 
University of Ghana  http://ugspace.ug.edu.gh
Output-oriented VRS model for DMU O3 relative to Meta frontier 
TE METAO3 (x, y)  Max 
       s .t
    9o1  6.5o2  6o3  7.5I1  5I 2  5.5I 3  4F1  7F 2  2F 3  6
    3o1     4o2  2o3  3.5I1  4I 2  5.5I 3  5F1   3F 2  3F 3  2    
      o1       o2    o3        I1    I 2       I 3     F1    F 2    F 3 1  
      o1       o2    o3        I1    I 2       I 3     F1    F 2    F 3 1 
                                                                                                 λ j  0
 TE METAO3 (x, y)  1.5
The group-specific efficiencies as well as the meta efficiencies and TGR for all DMUs are 
presented in Table 4.3. For output orientation as in this case, the inverse of the TGR is used 
in order to ensure that the values are bounded by 0 and 1. 
Table 4.3 : Group and Meta Efficiencies 
 Meta Efficiencies OPEC Frontier IEA Frontier FSU Frontier TGR 
TEM (x, y) TEOPEC (x, y) TE IEA( ) ( ) ( (x, y) ) (TE FSU (x, y) ) 
DMUs IGO 
O1  OPEC 1 1 - - 1 
O2  OPEC 1.091 1 - - 0.92 
O3  OPEC 1.5 1.5 - - 1 
I1  IEA 1.065 - 1 - 0.94 
I2  IEA 1.245 - 1.2222 - 0.98 
I3  IEA 1 - 1 - 1 
F1  FSU 1.1 - - 1 0.91 
F2  FSU 1.179 - - 1 0.85 
F3  FSU 1.833 - - 1.6667 0.91 
 
Based on the values in Table 4.3, DMU O2 for example is producing at 91% of its potential 
due to membership of the OPEC frontier. 
 
4.3.2 Global-frontier differences 
Whereas the metafrontier approach is able to provide a group specific efficiency assessment 
based on the group frontier as well as the metafrontier, it is sometimes necessary that best 
performing in two different groups are compared to see which group outperforms the other. 
61 
 
University of Ghana  http://ugspace.ug.edu.gh
Whereas the metafrontier approach is an intra-group performance indicator, the global 
frontier shift differences measure is an inter-group performance indicator that compares 
groups of DMUs. The global frontier shift index is a component of the global Malmquist 
productivity index and is the primary model used in comparing the frontier differences 
between various groups (Asmild & Tam, 2007). This examines the overall rather than the 
individual changes in the frontier of various groups (Asmild, Hollingsworth, & Birch, 
2013). This means that this model can draw conclusions about performance differences for 
the entire sample. It is better since aggregating other performance indices can be problematic 
using other methods in sparsely populated data set and unbalanced panel (Otsuki, 2013). 
The global frontier shift index therefore performs better than traditional frontier shift 
indices. It can also cater for overlapping frontiers.  
In estimating the global frontier differences, it is important to define a technology index for 
each group k (TI
k ) , which is a geometric mean of the efficiency scores of all firms 
belonging to a particular group k. This is defined as: 
1
  j
k  TI (x j , y j ) 
k
 TE (x
K
j , y
K
j )                                       (7)
 
 j1,n 
 K1,,k 
The global frontier shift index or global technical change or the global frontier difference 
between different groups k  and k is defined as: 1 2
62 
 
University of Ghana  http://ugspace.ug.edu.gh
k
TI 2 (x j , y j )k ,k
GFD 1 2 
k
TI 1 (x j , y j )
1
  j
 k 
TE 2 (x
K
j , y
K
j )                                      (8) 
 j1,n 
 K1,,k 
             
1
  j
 k
TE 1 (xK K

  j , y j )
 j1,n 
 K1,,k 
This is the ratio of the geometric mean of the efficiencies of all firms relative to k  frontier 2
to the geometric mean of the efficiencies of all firms relative to k  frontier. The efficiency 1
scores are computed using similar model formulation to that already presented in equation 
(4). The GFD >1 indicate that group 1 is, on average better that group 2 frontier. When GFS 
= 1 then the indication is that group 1 frontier is not better than group 2 frontier. Finally, if 
GFS < 1, then group 1 frontier is worse than group 2. Frontier shift is used where changes 
in time are being assessed. However, where the frontiers are of different groups, then the 
term global frontier differences is preferred. 
To illustrate the Global Frontier Differences approach, we use the same dataset and graph 
shown in Table 4.2 and Figure 4.1 respectively. Now efficiencies scores of all the 9 DMUs 
will estimated using each of the group frontiers. For example, the efficiency of DMU O3 
relative to the IEA frontier will be 7.5/6 = 1.25. O3’s efficiency relative to FSU’s frontier 
will be 7/6 = 1.167. This process is done for all DMUs. The mathematical programming 
models for the efficiency of DMU O3 relative to the IEA and FSU frontiers are presented 
below: 
 
 
63 
 
University of Ghana  http://ugspace.ug.edu.gh
Output-oriented VRS model for DMU O3 relative to IEA frontier 
TE IEAO3 (x, y)  Max 
       s .t
    7.5I1  5I 2  5.5I 3  6
    3.5I1  4I 2  5.5I 3  2    
         I1    I 2       I 3 1     
         I1    I 2       I 3 1 
                              λ j  0
 TE IEAO3 (x, y)  1.25
Output-oriented VRS model for DMU O3 relative to FSU frontier 
TE FSUO3 (x, y)  Max 
       s .t
    4F1  7F 2  2F 3  6
    5F1   3F 2  3F 3  2    
         F1    F 2    F 3 1     
         F1    F 2    F 3 1 
                              λ j  0
 TE FSUO3 (x, y)  1.16667
The efficiencies for all the firms are therefore presented in Table 4.4 
Table 4.4: Efficiencies for Global Frontier Difference Analysis 
OPEC Frontier IEA Frontier FSU Frontier 
TEOPEC( (x, y) IEA FSU) (TE (x, y) ) (TE (x, y) ) 
DMUs IGO 
O1 OPEC 1 0.8333 0.7778 
O2 OPEC 1 1.0476 0.92 
O3 OPEC 1.5 1.25 1.1667 
I1 IEA 1.0154 1 0.902 
I2 IEA 1 1.2222 1.0455 
I3 IEA 0.7273 1 0.8364 
F1 FSU 0.8 1.1 1 
F2 FSU 1.1379 1.0714 1 
F3 FSU 1.3333 1.8333 1.6667 
Geometric Mean 1.0334 1.1248 1.0104 
64 
 
University of Ghana  http://ugspace.ug.edu.gh
Therefore, based on the geometric means in Table 4.4, and the notations in Equation (8), the 
GFD between IEA and OPEC can be estimated as 1.1248/1.0334 = 1.09. This shows that 
the IEA frontier is about 9% better on the average than the OPEC frontier. A matrix of the 
GFD between the three groups is presented in Table 4.5. 
Table 4.5: Global Frontier Differences 
 OPEC IEA FSU 
OPEC 1   
IEA 1.09 1  
FSU 0.98 0.90 1 
 
Therefore, for the hypothetical data, whereas IEA frontier is 9% better than OPEC, FSU 
frontier is 2% worse than the OPEC frontier. Indeed, the FSU frontier is 10% worse than 
the IEA frontier. 
 
4.3.3 Bootstrapping 
DEA nonparametric efficiency scores have an inherent bias since efficiency scores can be 
affected by changes in the data and presence of the outliers and do possess the necessary 
statistical properties for making inferences (Gitto & Mancuso, 2012; Simar & Wilson, 2000; 
Simar & Wilson, 2015). Although parametric efficiency estimators like stochastic frontier 
analysis (SFA) permits statistical inference (Atkinson & Cornwell, 1994), when the sample 
observations are few these parametric statistics may not be applicable (Lothgren & 
Tambour, 1999). To correct the deficiencies of bootstrapping, Simar and Wilson (1998); 
(2000) proposed a bootstrap method which allows for statistical inferences on distance 
function estimates (efficiency scores) estimated using the DEA approach along the lines of 
the bootstrap approach which was introduced by Efron (1979). Bootstrap procedures offer 
realistic assessment for improving the bias of the efficiency estimates and for building 
confidence intervals for the efficiency scores (Simar & Wilson, 2015). The idea of 
65 
 
University of Ghana  http://ugspace.ug.edu.gh
bootstrapping introduced by (Efron, 1979) is to use empirical distribution of resampling data 
to generate a new dataset to estimate and reproduce repeated efficiency score from the 
observed data. This is the process of regenerating the original data repeatedly and each time 
estimating the efficiency scores. The re-generation of the data is perform by resampling with 
replacement of the original data (Hoff, 2006). It is a computer intensive approach (Simar 
& Wilson, 1999).  

The aim of the bootstrap algorithm is to mimic the distribution of DEA scores TE j (x, y) , 
in order to approximate the true unknown efficiency score TE j (x, y) . However, since the 
true score is unknown, the difference between the true score and the estimated score 
 
TE j

 (x, y)TE
j (x, y)  is also unknown. However, appropriate bootstrap approximation 
 
  
provides the opportunity to proxy TE
j (x, y) TE j (x, y)  to the bootstrap counterpart, 
 
 j* j  , where; TE j*TE (x, y) TE (x, y) (x, y)  is the bootstrap estimate which is completely 
 
known after the resampling procedure. The approximation between the two distributions is 
appropriate when the homogenous bootstrap is used (Kneip, Simar, & Wilson, 2008). Once 
the distributions are mimicked, statistical properties like the bias, bias-corrected score and 
confidence intervals for each DMU are easily derived. 
  
It is given by the following bootstrapped counterpart, the bias for a DMU is the difference 
between the average of the bootstrap samples for the DMU and the original estimated 
efficiency score, such that: 
   
BIAS (TE j )  E(TE j*) TE j
B                        (9) 
j*
                    B1TEb TE
j           j  1,    ,n
b1
66 
 
University of Ghana  http://ugspace.ug.edu.gh

j*
Where TEb is the bootstrapped efficiency and B is the number of bootstrap samples. After 
estimating the bias, the bias-corrected DEA efficiency scores can be estimated as: 
   
TE j**  TE j  BIAS (TE j )
 B                              (10) 
j*
          2TE j  B1TEb           j  1,    ,n
b1
It must be recognized that DEA efficiencies are corrected unless the ratio between absolute 
value of bias and standard deviation is greater than 0.25: 
 
 
  BIAS (TE j )  
     0.25                  j 1,    ,n                     (11) 
st d (TE j )
where, 
1/ 2
   1 B  j j* j* st d (TE )    (TEb - TE )
2
b  ,         b 1,.  , B        (12) 
B1 b1 
 

j*
and TEb  is the mean of the bootstrapped efficiency scores. As the bootstrap distribution 
 
TE j*

 (x, y) TE
j (x, y)  is completely known, the relative 
*
  and 
*
  quartiles, for a given 
 
level of probability could be easily found. These are good proxies for the quartiles of the 
 j j 
unknown distribution TE (x, y)TE (x, y)  and can be used in the bootstrap 
 

approximation as Prb*α TE j(x,y)TE j(x,y) a*α 1α . Hence, the 1  percent 

k *
confidence interval is given by the lower bound at the level TE   and the upper at the 

k *
level TE   .  
67 
 
University of Ghana  http://ugspace.ug.edu.gh
4.3.4 Testing returns to scale 
Without statistical testing, it is uncertain whether the production technology set that operates 
in an industry exhibits a constant or variable return to scale (CRS) or a variable one (VRS) 
in the measurement of efficiency in a nonparametric DEA model (Camanho & Dyson, 
2005). It is therefore imperative to investigate the returns to scale characteristics of the 
international oil industry in order to use the appropriate models relating to the appropriate 
underlying technology. There have been few studies on statistical testing of returns to scale 
estimators in DEA especially in the oil and gas industry. Testing of the returns to scale 
hypothesis in DEA-models was earlier discussed in the works of Banker (1993, 1996). 
However, the semiparametric nature of this returns to scale approach by Banker (1993) has 
been seen by Simar and Wilson (2002) to be problematic. Therefore, the bootstrap method 
proposed by Simar and Wilson (2002) is used to test the scale elasticity of the oil and gas 
industry. Simar and Wilson (2002) introduce the DEA bootstrap method for testing the 
returns to scale in order to know whether the technology shows constant return to scale 
everywhere on the frontier or otherwise.     
The basic idea of the test of returns to scale is along the lines of Färe and Grosskopf (1985) 
by computing ratios of the CRS technical efficiency score to the VRS technical efficiency 
score (TEcrsj (x, y) TE
vrs
j (x, y) 1)  for each observation j 1,...,n in the technology set T
crs vrs
. With this idea, if the ratio is equal to 1 (i.e.  TE j (x, y) TE j (x, y) 1) , then the estimated 
technology is assumed to exhibit CRS at the benchmark point of that particular DMU; 
otherwise, the estimated technology is assumed to exhibit a variable returns to scale at this 
TEcrs vrspoint. This means that, if for any particular observation j (x, y) TE j (x, y) 1 , then 
CRS may not hold for this observation. However, without a formal statistical testing 
procedure, it is quite difficult to determine whether this is due to non-constant returns to 
68 
 
University of Ghana  http://ugspace.ug.edu.gh
scale or merely due to sampling variation  (Simar & Wilson, 2002). Therefore, the null (Ho) 
and alternative (H1) hypotheses for the Simar and Wilson (2002) test of returns to scale are 
stated as follows: 
H0:  T is globally CRS (i.e. the technology set exhibits CRS for all observations) 
H1:  T is VRS (i.e. the technology set does not exhibit CRS for all observations) 
To develop the test statistic to be used in testing the hypotheses above, consider first an 
estimator of scale efficiency sx, y  of a particular DMU, defined as 
  
s TEcrs(x, y) TEvrsj j (x, y)   the mean of ratios of all observations can be used as the test 
statistic for this test.  This is defined as: 

n TE crs (x, y) 
Sˆ crs
j
1  n
1`                                                (13) 
j1 TE
vrs
j (x, y)


which is an estimator of the mean scale efficiencies for all observations. If the null 
hypothesis (H0) is true, then the CRS technical efficiencies for all observations is expected 
to be the same as the VRS scores, and Sˆ crs1 1.  Otherwise S
ˆ crs
1  1which signifies that the 
mean of ratios is significantly less than unity. Other test statistics can be used but the mean 
of ratios is employed just like Tortosa-Ausina, Grifell-Tatjé, Armero, and Conesa (2008) 
because the mean of ratios has an intuitive geometric interpretation (Simar & Wilson, 2002). 
Finally, in order to statistically test the hypotheses, there is the need for an appropriate 
critical value or p-value. However, since the statistical distribution of the scale test is 
unknown, the bootstrap methodology developed in Simar and Wilson (1998) can be 
extended in order to generate an empirical distribution of the scale test from which critical 
values, confidence intervals and p-values can be estimated. 
 
69 
 
University of Ghana  http://ugspace.ug.edu.gh
4.3.5 Testing of differences in the distribution of efficiency scores 
Testing of differences in efficiency among various groups is an important area of concern 
for empirical studies (Simar & Zelenyuk, 2006). One of the issues of concern is that DEA 
estimated efficiency scores are restrictive efficiency scores distributed across population of 
given set of DMUSs and do not provide good basis for statistical tests (Simar & Zelenyuk, 
2006; Trinh & Zelenyuk, 2015). Therefore, traditional tests of differences like t-test, anova, 
Mann Whitney test and Wilcoxon’s tests, will not provide the appropriate statistical power. 
In examining whether the differences in the efficiencies scores of the IGOs in the 
international oil and gas industry are statistically significant, or whether due to only 
estimation variations, a test of difference in the distribution of the efficiency scores is 
important. The study employs the Simar-Zelenyuk-adapted-Li test (SZAL) to statistically 
explore significant differences in the distribution of efficiency or frontier estimates between 
different IGOs in the oil and gas industry (Li, 1996; Simar & Zelenyuk, 2006). This 
nonparametric test effectively compares the equality of distributions of efficiency estimates 
using kernel density estimations. The non-parametric kernel density estimator, is largely 
gaining more significance in research (Banker, Cooper, Seiford, Thrall, & Zhu, 2004; Simar 
& Zelenyuk, 2006) and is very useful since there is no distributional assumptions imposed 
on the efficiency scores across the groups.  
In comparing the density of distribution of efficiency scores between two random groups 
for which the random samples TE A, j : j 1,...,n andTE B, j : j 1,...,n, representing the 
efficiencies of the two subgroups A and B in a population. Now, given that f l denotes the 
density of the distribution of the efficiency  TE l l  A, B  our null and alternative 
hypotheses would be  
70 
 
University of Ghana  http://ugspace.ug.edu.gh
H 0 : f
A
A TE  f B TE B ,
                                                       (14) 
H a : f A TE A  f BB TE 
The true technical efficiency scores in each subgroup, TE A, j : j 1,...,n  and
TE B, j : j 1,...,n, are independently and identically distributed (i.i.d.) within each 
subgroup with densities f .  and f . , respectively. The bootstrap algorithm for the Simar A B
and Zelenyuk adapted Li-test in comparing the distributions of efficiency scores can be 
summarized as follows: 
1. For each DMU estimate the TE(x, y)  using the DEA approach, therefore obtaining 
a sequence of estimated efficiency scores for all DMUs{ TE j j 1,,n  }. 
2. Smooth the original estimates of the efficiency scores using the smoothing rule: 
 j j j j j
TE* (x j j
TE(x , y )   ,       i f  TE(x , y ) 1
, y )    
TE(x
j , y j ),               otherwise            
Based on this, split the sample estimates into two sample estimates into two 
subsamples of DEA estimates, A and B thus obtaining: 
{  TE*A, j j 1,,n   }                                                                (A1) b A
{  TE*B, j j 1,,n   }                                                                (A2) b B
3. Next, estimate the Li (1996) test statistic using the subsamples in (A1) and (A2) and 
bandwidth h*  min{ h*A ,h
*
B  }, where h
*
A  and h
*
B  are obtained using same optimal 
rule applied to (A1) and (A2) respectively. 
4. Resample from the largest subsample out of (A1) or (A2) in order to obtain the 
bootstrap analogues of (A1) and (A2) and call them: 
{  TE**A, j j 1,,n   }                                                               (A3) b A
{  TE**B, j j 1,,n   }                                                              (A4) b B
71 
 
University of Ghana  http://ugspace.ug.edu.gh
5. Estimate the bootstrapped Li test statistic using (A3) and (A4) and 
**
hb  min{ h
** **
b,A ,hb,B  }, where h
**  and h**  are obtained using the same optimal rule b,A b,B
applied to (A1) and (A2) in step 3 to (A3) and (A4) respectively. 
6. Repeat steps 4 and 5 Bb 1,, B  times to obtain B bootstrap estimates of the Li 
statistic that will mimic the distribution of the original estimate of the Li statistic 
under the null hypothesis. 
 
4.4 Modelling inputs and outputs 
Three inputs and two outputs are selected for the efficiency estimation process. Oil reserves, 
gas reserves and total labour force employed were chosen as inputs to generate natural 
resources of oil and gas. The two outputs, oil and gas quantities are physically generated 
from the oil and gas reserves using human resources. These variables are selected because 
the issue under consideration is how oil producing countries are converting their resource 
inputs into maximum outputs obtainable. Table 4.6 presents a summary of the selected 
variables to be used in the analysis. Selection is mainly guided by literature. 
  
72 
 
University of Ghana  http://ugspace.ug.edu.gh
Table 4.6: Model Variable Selection 
 
Purpose Variable Unit  
Crude Oil Billions of Reason  Very important asset and is regarded as accurately reported 
Proved barrels (Eller et al., 2011) 
Reserves 
Used by: (Eller et al., 2011; Ike & Lee, 2014; Kashani, 2005b; 
Sueyoshi & Goto, 2012a; Thompson et al., 1996; Wolf, 
2009) 
Proved Trillion of Reason Very important asset and is regarded as accurately reported  
Reserves of Cubic (Eller et al., 2011) 
Inputs 
Natural Gas Feet Used by: (Eller et al., 2011; Ike & Lee, 2014; Kashani, 2005b; 
Sueyoshi & Goto, 2012a, 2012b; Thompson et al., 1996; 
Wolf, 2009) 
Employees Millions Reason  This is a proxy for labour as a factor of production 
Used by: (Eller et al., 2011; Ike & Lee, 2014; Kashani, 2005b; 
Sueyoshi & Goto, 2012a, 2012b, 2014; Thompson et al., 
1996; Wolf, 2009) 
     
Total Oil Thousand Reason  The input resources must be accounted for and the maximum 
Supply Barrels obtainable output from oil reserves using best practice is oil 
Per Day production (Ike & Lee, 2014) 
 
Used by: (Barros & Assaf, 2009; Barros & Managi, 2009a; Eller et al., 
2011; Ike & Lee, 2014; Kashani, 2005b; Sueyoshi & Goto, 
2012a, 2012b; Thompson et al., 1996; Wolf, 2009) 
Outputs  
Gross Billion Reason  The input resources must be accounted for and the maximum 
Natural Gas Cubic obtainable output from gas reserves using best practice is gas 
Production Feet production (Ike & Lee, 2014) 
 
Used by: (Barros & Assaf, 2009; Barros & Managi, 2009a; Eller et al., 
2011; Ike & Lee, 2014; Kashani, 2005b; Sueyoshi & Goto, 
2012a, 2012b; Thompson et al., 1996; Wolf, 2009)   
 
 
4.4.1 Inputs 
The three inputs used for generating oil and gas outputs for all the oil producing countries 
are oil and gas reserves estimated differently and the labour force employed. 
 
4.4.1.1 Oil and Gas reserves 
Oil and gas reserves are commercially identified volumes of oil and gas that can be 
recovered in the future. Oil reserves are volumes of an estimated quantity of crude oil 
identified in a specific area through geographic analysis and data from demonstrated 
engineering surveys (EIA, 2013). The resource are converted into outputs that generates 
73 
 
University of Ghana  http://ugspace.ug.edu.gh
significant revenue (Wolf, 2009). It is the most important asset and is regarded as more 
accurately reported by all countries than other inputs (Eller et al., 2011). These two reserves 
are measured separately. Whereas oil reserves are measured in billions of barrels (bbls), 
natural gas reserves are measured in trillions of cubic feet (tcf). 
 
4.4.1.2 Labour force 
Labour is a critical resource in the production process. It is therefore a key resource in all 
production efficiency estimation models, including DEA. The oil industry is both labour 
and capital intensive process. Although labour can be measured in numbers, wages and 
salaries as well as labour hours, there is no comprehensive database or source for this 
variable. However, time series datasets of the total labour force for all countries are available 
in the World-Bank’s World Development Indicator database. This labour force statistic will 
not be an appropriate proxy as labour for the oil industry of these countries since it includes 
persons operating in all other industries of the country. Therefore, labour force of these 
countries are estimated in relation to the industry’s contribution to the total GDP of the 
country at a particular time. Therefore, oil and gas labour force for a country j at time t is 
estimated as a function of the total labour force of that country at time t given an average of 
the oil rents and the natural gas rents. Note that the oil and gas rents are estimated as a 
percentage of the GDP and can be seen as a proxy of the contribution of the oil and gas 
industry towards the GDP of the country. This is can be expressed as: 
 GR jt OR jt 
OLF jt  LF                                                (15)  jt 
  2 
OLF jt   = Oil and Gas Labour force for country j at time t. 
LF jt   = Labour force for country j at time t. 
74 
 
University of Ghana  http://ugspace.ug.edu.gh
GR   = Natural Gas Rents for country j at time t as a percentage of GDP. jt
OR   = Oil Rents for country j at time t as a percentage of GDP. jt
Whereas natural gas rents are the difference between the value of natural gas production at 
world prices and total costs of production, oil rents are the difference between the value of 
crude production at world prices and total costs of production (World Bank, 2011). 
 
4.4.2 Outputs 
The two outputs are the oil and gas quantities physically generated from the critical inputs 
of oil producing countries. Oil and gas production is the quantity of oil and gas that have 
been recovered in a given time period. This is primarily output from operations of drilling 
from the oil and gas reserves as an end product of the upstream industry activities (Wolf, 
2009). By this, increasing the production output is an essential pointer for improved 
economic performance of oil producing countries (OPCs). Oil and gas producing countries 
are looking forward to meeting the growing demands on improving the socioeconomic 
activities of their countries essentially from oil and gas production generated revenue. Hence 
production in thousands of barrels per day and billion cubic feet for oil and gas respectively 
are good estimates for measuring performance of not just output generated revenue, but also 
for benchmarking and appropriate technology use.  
 
4.5 Other DEA consideration 
Two DEA models input and output orientations are used in the efficiency assessment.  The 
adapted model for this study is the output-orientation DEA model for the four IGOs, and 
sample of 65 oil producing countries belonging to them. In the output orientation, the 
proportional output expansion level is observed to provide the best practice technology 
75 
 
University of Ghana  http://ugspace.ug.edu.gh
being applied whilst the input is held constant. Technically efficient oil producing countries 
will be required to increase their output for a given set inputs. More specifically the output-
oriented technical efficiency measure is appropriate in respect of a group(s). This has been 
applied to most oil and gas related efficiency studies (Barros & Assaf, 2009; Eller et al., 
2011; Ike & Lee, 2014; Ismail et al., 2013). This is because oil producing countries are more 
concerned with the maximum production output for supply to the various areas (oil 
consuming countries) since pricing decisions are left with market conditions. Apart from 
that, output-orientated measure is more inclined to industries with critical demand schedules 
(Zhou et al., 2008). For all estimations, R software version 3.1.1 used with the rDEA, 
Benchmarking and Nonparaeff packages. MaxDEA and EMS were also used for 
confirmatory purposes. 
 
4.6 Conclusions 
The discussion of this chapter emphasized and provided a clear and precise description on 
the methodology and the adopted assumptions in the assessment of this study. It also showed 
the data source as well as the quantitative, experimental and positivist research approaches 
employed in the data collection and analysis process.  
  
76 
 
University of Ghana  http://ugspace.ug.edu.gh
CHAPTER FIVE 
ANALYSIS AND DISCUSSIONS 
5.0 Introduction 
This section provides empirical results aimed at answering the research questions and 
achieving the objectives of the study. The chapter is divided into five sections. Whereas the 
first presents a description of the data used, the final section presents a summary of the 
chapter. The remaining three sections provides empirical results that directly answer the 
research questions. Whereas section two presents answers to research question 1, section 
three presents answers to research questions 2 and 3. Finally, section four provides answers 
to questions 4 and 5. For each of these research questions addressed, discussions are made 
with respect to literature and practice. 
 
5.1 Data Description 
The data for the analysis was primarily sourced from the US Energy Information 
Administration’s (EIA) database of oil and gas statistics of oil producing countries (OPCs) 
in the international oil industry. Whereas data for several countries are available, data for 
the 52 OPCs under consideration were gathered from the period 2000 to 2013. Annual oil 
reserves, annual gas reserves, daily oil production and gas production levels were gathered 
from this source. The data needed for the estimation of labour force of these OPCs was from 
the World Banks’s World Development Indicators database. Generally, three inputs and two 
outputs are used for the efficiency estimation. Whereas the oil reserves, gas reserves and 
labour force are the inputs, production levels of oil and gas are the two outputs. Descriptive 
statistics of the variables for the pooled dataset covering the 14-year period from 2000 to 
2013 is presented in Table 5.1. The number of observations (N), mean, standard deviations, 
minimum and maximum values for each variable and for each IGO are presented. Also 
77 
 
University of Ghana  http://ugspace.ug.edu.gh
included in the table are the results of a test of differences using one-way ANOVA to test 
differences in the variables between the four IGOs understudy. 
Table 5.1: Descriptive Statistics of Pooled Data from 2000 to 2013 
ANOVA 
 N Mean Std. Dev. Minimum Maximum 
  F Sig. 
Oil Reserves FSU 140 8.53 18.30 0.01 80.00 71.32 .000 
IEA 280 9.09 34.50 0.01 180.02   
OAPEC 138 56.18 81.33 0.12 267.91   
OPEC 165 78.76 75.83 2.12 297.57   
Overall 723 33.87 63.28 0.01 297.57   
Gas Reserves FSU 140 202.98 497.81 0.10 1700.00 25.07 .000 
IEA 280 25.88 54.29 0.02 334.07   
OAPEC 138 147.93 232.06 2.30 910.52   
OPEC 165 245.66 301.78 0.25 1187.00   
Overall 723 133.63 296.81 0.02 1700.00   
Labour Force FSU 140 13621277 21562330 1734672 77074406 19.91 .000 
IEA 280 23798984 33942132 1931619 159144632   
OAPEC 138 6427296 6786129 303604 27742106   
OPEC 165 10870575 12351137 332190 54196350   
Overall 723 15561972 25022196 303604 159144632   
Oil Production FSU 140 1174.84 2785.94 0.21 10757.91 25.12 .000 
IEA 280 856.94 2157.16 3.19 12342.77   
OAPEC 138 2163.83 3010.28 47.40 11840.68   
OPEC 165 2875.40 2556.94 392.72 11840.68   
Overall 723 1628.59 2679.15 0.21 12342.77   
Gas Production FSU 140 2870.49 6462.89 0.00 24317.91 1.46 .223 
IEA 280 2338.43 5694.76 0.18 30005.00   
OAPEC 138 1680.55 2006.22 68.51 7104.88   
OPEC 165 2308.40 2188.92 38.85 8169.49   
 Overall 723 2309.03 4749.99 0.00 30005.00   
 
Overall, the dataset is made up of 723 observations over the 14-year period for the 52 OPCs 
who belong to the four IGOs in the international oil industry. In Table 5.1, the range between 
the minimum and maximum values for all variables are high showing the possibility of 
differences in the sizes of the oil and gas industries of the countries understudy. This view 
is supported by the size of the deviations from the sample means. For example, the mean of 
the oil production output of 1628.59 thousands of barrels of crude per day has an even larger 
standard deviation of 2679.15.  
78 
 
University of Ghana  http://ugspace.ug.edu.gh
Shifting attention towards the individual IGOs, for oil reserves, OPEC holds the highest oil 
reserves (M=78.76, SD=75.83). OAPEC follows in second place (M=56.18, SD= 81.33), 
followed by IEA (M= 9.09, SD= 34.5) and FSU (M= 8.53, SD= 18.30). There is evidence 
of significant differences in the oil reserves capacities of the various IGOs (F=71.32, 
p<0.001). The reserve endowment ranking seem to change a little when gas reserves are 
considered. Although, OPEC member states possess the largest amounts of gas reserves 
(M=245.66, SD= 301.78), they are closely followed by FSU member states (M= 202.98, 
SD= 497.81) who were seen to possess the least oil reserves.  In third place for gas reserves 
is OAPEC (M= 147.93, SD= 232.06). Finally, IEA members hold the least gas reserves on 
average (M= 25.88, SD= 54.29). Similarly, significant differences were identified in the 
average gas levels of member states (F=25.07, p<0.001).  
For the levels of labour force employed, IEA has the highest number of workers 
(M=23798984, SD=33942132), FSU comes second (M=13621277, SD=21562330) and then 
OPEC (M=10870575, SD=12351137). The F-statistic of 19.91 with p-value less than 0.1% 
also indicates significant differences among the labour force of the IGOs. Although it is 
possible that higher resource endowment can lead to higher production levels and hence 
higher efficiencies compared to less endowed IGOs, it is equally possible that higher 
resource endowment can lead to lower efficiencies if the production capabilities of these 
countries are not streamlined adequately to ensure higher production levels. 
The oil production outputs in thousands of barrel per day of the four IGOs are also 
significantly different (F=25.12, p<0.001). OPEC member states produce the highest levels 
of oil outputs (M=2875.40, SD=2556.94). This is encouraging since they also are the highest 
endowed IGO in terms of oil reserves. OAPEC is the second largest producer of oil outputs 
(M=2163.83, SD=3010.28) followed by FSU. IEA has the lowest oil production levels 
(M=856.94, SD=2157.16). It is also clear that with gas production levels, FSU leads the 
79 
 
University of Ghana  http://ugspace.ug.edu.gh
pack (M=2870.49, SD=6462.89). FSU, however, is closely followed by the remaining IGOs. 
Indeed, there are no statistically significant differences in the average gas production levels 
of the various IGOs (F=1.46, p=0.223). 
Summary of the annual statistics of the data have been attached in Appendix C. A test of 
differences for each variable over time is also included in Appendix C. The One way Anova 
test of differences for each variable are not statistically significant, giving an indication of 
little variation in the industry over time, and providing justification for using a pooled 
frontier for efficiency estimation. 
The final stage in the data description process in DEA estimation is the isotonicity test. The 
isotonicity property of DEA requires a positive correlation between all inputs and outputs 
(Cooper, Seiford, & Zhu, 2011; Thanassoulis, 2001). This means that, its expected 
consumption of more inputs will lead to higher outputs. The inputs and outputs correlation 
test in a nonparametric frontier analysis is necessary for robust analysis. The correlations 
between the inputs and outputs are presented in Table 5.2.   
Table 5.2: Correlations between Inputs and Outputs 
Oil Gas Oil Gas Labour 
  Production Production Reserves Reserves Force 
Oil Production 1     
Gas Production .673** 1    
Oil Reserves .749** .157** 1   
Gas Reserves .526** .618** .261** 1  
Labour Force .435** .764** -.032 .280** 1 
** p < 0.01 
 
From Table 5.2, all inputs from the table are positively and significantly associated with 
both outputs. For example, whereas the correlation between oil production and oil reserves 
is 0.749, the correlation between gas production and labour force is 0.764. All correlations 
are significant at the 1% level therefore the isotonciity characteristic of DEA which requires 
80 
 
University of Ghana  http://ugspace.ug.edu.gh
that an output should not decrease with an input increase (Dyson et al., 2001; Honma & Hu, 
2008; Wanke, Barros, & Faria, 2015) is not violated. The intuition for the positive 
associations is that employing more inputs is expected to lead to higher production levels. 
It is not surprising that there is a stronger correlation between oil production and oil reserves 
as compared to the correlation between oil production and gas reserves. This is because oil 
production emanates from oil reserves. Same remark is evident for gas production as well. 
Finally, there are relatively weaker correlations among the inputs. Even the correlation 
between the oil reserves and labour force is not statistically significant (r = -0.032, p > 0.05). 
The weak or no correlations among the inputs is also an encouraging sign in DEA 
estimations since it provides evidence of the discriminatory power of the inputs used (Dyson 
et al., 2001). This means that the inputs actually measure different dimensions in the 
production process. 
 
5.2 Scale Elasticity in the International Oil Industry 
The first objective is to test the scale elasticity properties of the industry. The assumption 
on returns to scale of the underlying technology is one of the important a priori assumptions 
necessary to employ DEA (Badunenko, 2008). While the production frontier can exhibit 
either constant return to scale or variable return to scale, previous researchers have failed to 
statistically test it. The consequence is that research finding from previous study may be 
biased since the appropriate production technology may not have been assumed. Here, 
Simar and Wilson’s (2002) bootstrap-based scale elasticity test  is used. The elasticity 
property of the frontier is first tested on annual bases, thereby assuming heterogeneity in the 
production technology over time (Canhoto & Dermine, 2003). Subsequently the data for the 
entire period is pooled and tested to see the scale elasticity based on the assumption of 
technological homogeneity (Canhoto & Dermine, 2003). This procedure is adopted by 
81 
 
University of Ghana  http://ugspace.ug.edu.gh
Gómez-Calvet, Conesa, Gómez-Calvet, and Tortosa-Ausina (2014). In here, the mean of 
ratio is used following Tortosa-Ausina et al. (2008). The intuition behind this test is that as 
long as the difference between the CRS and VRS efficiency score is small we will not have 
the statistical confidence to reject the null hypothesis of CRS (Mahlberg & Url, 2010). 
Therefore, if the p-value for the test statistic is greater than the significance level we will 
fail to reject the null hypothesis of CRS, otherwise we will reject the null hypothesis in 
favour of a VRS frontier. Result of this test are presented in Table 5.3. The unbiased test 
statistic is from bootstrap DEA estimates based on 2000 replications.  
Table 5.3: Scale Elasticity Tests (Simar & Wilson, 2002) using the Mean of Ratios 
Year S test p-value Decision 
2000 0.8307 0.22 CRS 
2001 0.8351 0.27 CRS 
2002 0.8362 0.29 CRS 
2003 0.8671 0.42 CRS 
2004 0.8585 0.25 CRS 
2005 0.8535 0.28 CRS 
2006 0.8529 0.19 CRS 
2007 0.8134 0.19 CRS 
2008 0.7966 0.24 CRS 
2009 0.7995 0.29 CRS 
2010 0.8309 0.22 CRS 
2011 0.8371 0.29 CRS 
2012 0.8337 0.33 CRS 
2013 0.8251 0.25 CRS 
Overall/Pooled 0.8070 0.051 CRS 
 
From Table 5.3 the result of the test values for each year from 2000 to 2013 as well as the 
p-values provides a statistical indication that size does not matter in the international oil 
industry of OPCs. This is because the p-values of each of the 14 years is greater than the 5 
percent significance level. By implication, we fail to reject the null hypotheses and conclude 
that the production frontier exhibits constant scale elasticities for each year. Additionally, 
test results based on the pooled frontier also provide same conclusions since the p-value of 
0.051 is larger than the 0.05 significance level. 
82 
 
University of Ghana  http://ugspace.ug.edu.gh
This provides both statistical and empirical justification to adopt CRS in all DEA efficiency 
estimations in this work. This means that OPCs are of similar production capacities 
irrespective of the size of the country’s oil and gas resource endowment. Even the pooled 
data for the period also indicate same. The justification for the finding is probably because 
of the capital intensive nature of oil and gas exploration and production activities. Indeed, 
Szilas (1985) underscores that investment in the oil and gas industry, no matter the level of 
involvement, requires heavy monetary and logistical commitments. This means that, 
previous study by Hawdon (2003) on inter-country efficiency in the gas industry is 
questionable since he failed to adequately test for the returns to scale property. The reliance 
of Hawdon (2003) on t-test and Kolmogorov-Smirnov test for the returns to scale resulted 
in mixed findings. This current work therefore provides basis for the adoption of CRS by 
subsequent papers that aim to assess efficiency of OPCs. 
 
5.3  Meta and Group Analysis of Inter- and Intra-IGO Performance 
This section provides results and details of meta and group frontier analysis as well as 
technological gap ratio estimated using the metafrontier approach of DEA.  The metafrontier 
analysis provides a meaningful model for the comparison of efficiencies among and between 
groups of countries (Barnes & Revoredo-Giha, 2011). These results aid in achieving the 
second and third objectives of this study. Whereas objective two is aimed at assessing the 
intra and inter group performance of countries in each IGO, objective three aims to provide 
empirical comparison of the performance of various IGOs using the Simar Zenlenyuk 
adapted Li Test (SZAL). Whereas the intra group assessment looks at the performance of 
countries in a particular IGO, inter group assessment compares the performance of different 
blocks of IGOs.  
83 
 
University of Ghana  http://ugspace.ug.edu.gh
The metafrontier approach provides a curve enveloping all the group frontiers constructed 
for assessing DEA scores for each group and presents an empirical application using cross-
country data (O’Donnell et al., 2008b). Metafrontier analysis ensures that heterogeneous 
countries or groups can be examined based on their distance from an identical and common 
frontier (Assaf, Barros, & Josiassen, 2010). Therefore, the metafrontier approach fills in 
these disadvantages by allowing comparison across heterogeneous groups (Battese & Rao, 
2002a; O’Donnell et al., 2008a). The metafrontier provides benchmarking for all sample in 
the data set separately from the frontier in which they are members (Kounetas, Mourtos, & 
Tsekouras, 2009). The frontier analysis can be used by decision makers to pinpoint non 
performing units among the whole in other to orientate policy prescriptions for them. The 
section is divided into 2 parts. The first part considers only intra-group performance 
evaluation. The second part then considers similarities and differences in the performance 
of different IGOs (inter-group performance) in the international oil industry. 
 
Intra-Group Performance Evaluation 
Tables 5.4, 5.5, 5.6, and 5.7 presents results of metafrontier analysis for FSU, IEA, OAPEC 
and OPEC respectively. For each table, bootstrapped meta-efficiency scores, bootstrapped 
group efficiency scores and bootstrapped technological gap ratios (TGR) are reported. The 
meta-efficiency scores measure efficiency relative to the metafrontier which represents the 
existing state of knowledge in the industry irrespective of the particular group a country 
belongs. This is decomposed into the groups-specific efficiency and the TGR. Whereas the 
group efficiency scores represent the existing state of knowledge, the physical, social and 
economic environment that characterizes a particular group, the TGR provides a measure of 
how close the group frontier is to the metafrontier (O’Donnell et al., 2008a). In other words, 
the group efficiencies allow for an investigation of the relationship between different 
84 
 
University of Ghana  http://ugspace.ug.edu.gh
groups-specific technologies and is used to explain the differences in the production 
opportunities attributed to the resource endowments of a particular group. The value of the 
TGR explains the technological progress or regress between the group and the industry as a 
whole (Battese & Rao, 2002a). Therefore, a score closer to 1 signifies lower discrepancy 
between the group-specific frontier and the meta frontier (Ahmed & Krishnasamy, 2013). 
The TGRs of countries in each IGO have therefore been ranked in order of importance. 
Table 5.4: Intra-Group Analysis of FSU Member States  
Country IGO Meta Efficiencies Group Efficiencies TGR Rank 
Azerbaijan FSU 3.2583 1.4763 0.4531 4 
Belarus FSU 2.0023 1.1233 0.5610 2 
Georgia FSU 24.7591 7.2097 0.2912 9 
Kazakhstan FSU 4.3545 1.3011 0.2988 8 
Kyrgyzstan FSU 24.6788 8.7513 0.3546 6 
Russia FSU 2.8920 1.3199 0.4564 3 
Tajikistan FSU 68.0373 9.6844 0.1423 10 
Turkmenistan FSU 1.3249 1.2247 0.9244 1 
Ukraine FSU 4.9945 1.6374 0.3278 7 
Uzbekistan FSU 2.7007 1.1421 0.4229 5 
Geomean   6.0057 2.2911 0.3815   
      
ANOVA F 62.548 70.715 99.771  
 Sig. .000 .000 .000  
      
Kruskal Wallis 𝝌𝟐 125.773 116.754 120.802  
 Sig. .000 .000 .000  
      
 
The intra group analysis of the 10 member states of the FSU listed in Table 5.4 shows their 
meta-efficiency, group efficiency, TGR and the ranking of the TGR based on a dataset from 
2000 to 2013. Starting with the group efficiencies, it can be observed that no country has 
efficiency score exactly equal to 1. This is as a result of the bootstrap based bias correction 
of efficiency scores and is consistent with Hawdon’s (2003) assertion that in real market 
systems probability of a unit being exactly 100 percent efficient is zero. This 
notwithstanding, best performing countries in this IGO include Belarus (1.1233), 
Uzbekistan (1.1421), and Turkmenistan (1.2247). However, most other countries in the FUS 
85 
 
University of Ghana  http://ugspace.ug.edu.gh
have high inefficiency levels. For example Georgia, Kyrgyzstan and Tajikistan all have 
group efficiency score greater than 1. Even worse is when the meta-efficiency scores are 
considered. Except Turkmenistan, all other members of this FSU have meta-efficiency 
scores greater than 2.00.  
The disparities between the meta-efficiencies and group efficiencies give an indication that 
countries in this IGO are not producing using the best state of knowledge in the industry. 
Even the TGRs for all members except Turkmenistan (0.9244) and Belarus (0.5610) are 
below 50 percent. The average TGRs of all countries in the IGO is about 0.3815 signifying 
that on average countries in FSU produced using only 38.15 percent of the existing state of 
knowledge in the industry. Ranking of TGRs show that Turkmenistan, Belarus, Russia, and 
Azerbaijan are among the best countries in the FSU to capitalize on the existing global 
technological spill overs whilst Tajikistan, Georgia, Kazakhstan and Ukraine are trailing. 
Also reported in the table are result of Anova and Kruskal Wallis test of differences. Based 
on the results there are significant differences in the performance of individual countries in 
this IGO for all three indicators. This shows that not all countries are equally good or poor 
in this industry. Summary of the Tukey HSD pairwise comparison test as reported in 
Appendix D, for example shows that whereas Turkmenistan has the significantly highest 
TGR, the TGRs of Azerbaijan and Belarus are not statistically significantly different.  
Turkmenistan’s higher TGR of 0.9244 is far above that of the other countries in this FSU. 
The closest country is Belarus with TGR of 0.5610. Turkmenistan’s higher TGR may be a 
payoff from their government’s continual policy of modernizing and expanding 
infrastructure in the oil and gas industry by increasing investment to develop its resources 
and the implementation of market reforms to facilitate intensive production and supply of 
more gas (EIA, 2013). Belarus is the next best performing country in the FSU. The 
Belarusian government after the collapse of the Soviet Union took control of the country’s 
86 
 
University of Ghana  http://ugspace.ug.edu.gh
development and ensured high level of government’s involvement in the oil sector 
(Ghedrovici & Ostapenko, 2013). The huge state support for the sector possibly has 
accounted for their level performance. The strong industry that was maintained after the 
union collapse could also be a factor to their TGR of 0.5610. Another reason is their close 
links with the developed EU market that has advanced the level of orientation and the quality 
of market reforms (Gaytaranov, 2013; Ghedrovici & Ostapenko, 2013). The country’s 
performance is also due to their high level of human capital (Podkorytova & Raskina, 2014). 
Russia has the third largest TGR in the IGO. Russia’s performance can be associated with 
the policy of allowing the participation of private and foreign investors to manage and 
extract large oil and gas prospects (EIA, 2013). The investment include allowing 
international oil companies like Lukoil, Surgutneftegas, Novatek and Tatneft to invest and 
participate in the industry with tax incentives. Also, the use of advance technology and 
improved recovery techniques has accelerated oil production output from existing oil 
deposits and contributed to their supply improvement (EIA, 2013). This notwithstanding, 
Russia needs more efforts in the industry in order to move their TGR above the 0.50 mark. 
The low performance of most countries in the FSU may be attributable to their historical 
developments. Members of this IGO are states that were formed after the collapse of the 
Soviet Union (Minescu et al., 2008; Podkorytova & Raskina, 2014). The collapse generated 
diverse legal interpretation regarding boundary issues (EIA, 2013). The new independent 
states started the development of the large deposits of untapped oil and gas resource 
endowment with diverse and sometimes adversarial approaches individually (EIA, 2013). 
Therefore, the initial rivalries among states and the individual approach towards oil and gas 
development may have contributed to the poor performances of states. This view can also 
be seen based on the ideas of the theory of social networks which explains the importance 
of good ties among states as investment in the accumulation and management of social 
87 
 
University of Ghana  http://ugspace.ug.edu.gh
resources and capital (Katz et al., 2004). In other words, if states have strong ties they are 
bound to help themselves improve their social capital. However, when states have 
adversarial ties, as is seen in the FSU, the consequence may be detrimental to the wellbeing 
of these states. In addition when countries are operating alone, their performance may not 
be as good as if they were in a group as indicated in the theory of social facilitation (Markus, 
1978) because the external presence is not sufficient to influence their behaviors. Further to 
this, because these countries are not well institutionalised or do not follow any 
institutionalized behaviors as “regulative, normative, and cognitive structures and activities 
that provide stability and meaning for social behavior” (Scott, 1995) their performance are 
affected negatively base on institutional theory views. Their low performance may in part 
also be attributable to the fact that most FSU countries lack innovative technology, the 
infrastructure and investment to support the oil and gas sectors (EIA, 2014). 
  
88 
 
University of Ghana  http://ugspace.ug.edu.gh
Table 5.5: Intra-Group Analysis of IEA Member States   
Country IGO Meta Efficiencies Group Efficiencies TGR Rank 
Australia IEA 2.5025 2.3667 0.9457 4 
Austria IEA 2.3571 2.2408 0.9506 3 
Canada IEA 1.4636 1.1617 0.7938 16 
Czech Republic IEA 2.0217 1.9230 0.9512 2 
Denmark IEA 1.3178 1.1886 0.9020 11 
France IEA 1.2901 1.1651 0.9031 10 
Germany IEA 2.9925 2.3907 0.7989 15 
Greece IEA 1.8586 1.7283 0.9299 5 
Hungary IEA 1.8614 1.7059 0.9165 8 
Italy IEA 3.0119 2.7236 0.9043 9 
Japan IEA 1.3818 1.2702 0.9192 7 
Netherlands IEA 1.3123 1.1573 0.8819 13 
Norway IEA 1.2524 1.1629 0.9285 6 
NZ IEA 1.7235 1.2943 0.7510 18 
Poland IEA 3.9728 3.5796 0.9010 12 
Slovakia IEA 2.2430 2.1672 0.9662 1 
Spain IEA 2.4080 1.3765 0.5716 20 
Turkey IEA 2.5924 1.5230 0.5875 19 
UK IEA 1.4797 1.1413 0.7713 17 
US IEA 1.4600 1.2827 0.8786 14 
Geomean   1.9140 1.6255 0.8493   
      
ANOVA F 30.269 34.467 46.729  
 Sig. .000 .000 .000  
      
Kruskal Wallis 𝝌𝟐 98.333 98.453 60.007  
 Sig. .000 .000 .000  
      
      
 
The intra group analysis of the 20 member states of the IEA listed in Table 5.5 shows their 
meta-efficiency, group efficiency, TGR and the ranking of the TGR based on a dataset from 
2000 to 2013. Starting with the group efficiencies, it can be observed that the efficiency 
score of all the countries in the IEA are quite close to 1 and are mostly similar to each other. 
Similarly, no country has a score of exactly 1 as result of the bootstrap based bias correction 
of efficiency scores. However, best performing countries in this IGO include UK (1.1413), 
Netherlands (1.1573), Canada, Norway and France having score slightly above (1.16). This 
notwithstanding, some countries in this IGO seem to have efficiency scores that are quite 
different from the rest. These countries are Poland (3.5796), Italy (2.7236), Germany 
89 
 
University of Ghana  http://ugspace.ug.edu.gh
(2.3907), Australia (2.3667) and Austria (2.2408). In addition, the meta-efficiency scores of 
all the countries do not seem far apart from their group efficiency scores. This gives an 
indication that countries in this IGO may be producing using the best state of knowledge in 
the industry. Even the technology gap ratio for all members except Spain (0.5716), Turkey 
(0.5875) and New Zealand (0.575) are above 0.77. The average technology gap ratio of all 
countries in the IGO is about 0.8493 signifying that, on average countries in IEA produced 
using about 84.93 percent of the existing state of knowledge in the industry. Ranking of 
TGRs show that Slovakia, Czech Republic, Austria, and Australia are the best producers in 
the IGO whilst New Zealand, Turkey and Spain are trailing. Also reported in the table are 
result of Anova and Kruskal Wallis test of differences. Based on the results there are 
significant differences in the performance of individual countries in this IGO for all three 
indicators. This shows that all the countries are not performing at the same level of 
efficiency in this industry. Summary of the Tukey HSD pairwise comparison test as reported 
in Appendix D, for example shows that all the countries have significantly higher TGRs 
than Spain and Turkey.  
For this IGO, all members have relatively high TGRs. Even Spain which places 20th in the 
rankings has a TGR which is greater than 0.5. The performance of the countries in this IGO 
is instructive. Because most of the states are very close to the frontier after the bootstrap 
estimates. This possibly may be as a result of the strong ties that exist between the countries 
and the fact that the IGO membership are mostly drawn from the industrial regions of 
Western Europe and North America (Colgan, Keohane, & Van de Graaf, 2011). The IEA is 
part of a much stronger OECD with a framework of achieving high sustainable economic 
growth for it member countries in the process of economic development (Bamberger, Scott, 
Agency, & Development, 2004). As stated in the aim of IEA, the organization promotes 
rational policy through co-operative relations with industry and other international 
90 
 
University of Ghana  http://ugspace.ug.edu.gh
organizations (Bamberger et al., 2004). Social network theory suggests that the structure of 
a group determines the access and flow of resources in the network (Daly, 2012). The IEA’s 
clear concern for compliance monitoring, execution and evaluation of efficiency policies in 
line with the continuous transfer of policy to member countries also might have contributed 
to the performance observed (Jollands et al., 2010). It can therefore be implied that the 
activities and ways of operation by individual countries in the IEA will be very similar to 
that of other members and is consistent with the theory that organizational structure and 
process helps achieve effectiveness and efficiency in their desired outcomes as predicted by 
institutional theory. Therefore, it is expected that all members will also have close to similar 
levels of efficiency. The IEA policy for the development and technology transfer remains a 
key factor for the cooperation among member states (Colgan et al., 2011). This 
notwithstanding, as Taylor et al. (2010) posits, the relatively lower performance of countries 
like Spain (0.5716) and Turkey (0.5875) scores may be attributable to the differences in 
production infrastructure across countries.  
Table 5.6: Intra-Group Analysis of OAPEC Member States    
Country IGO Meta Efficiencies Group Efficiencies TGR Rank 
Algeria OAPEC 2.4920 1.1606 0.4657 9 
Bahrain OAPEC 1.3046 1.1399 0.8738 3 
Egypt OAPEC 3.5822 1.8823 0.5255 8 
Iraq OAPEC 4.2290 2.4220 0.5727 5 
Kuwait OAPEC 1.1972 1.1299 0.9438 1 
Libya OAPEC 2.4402 1.2840 0.5262 7 
Qatar OAPEC 1.2729 1.1669 0.9167 2 
Saudi Arabia OAPEC 1.4135 1.0949 0.7746 4 
Syria OAPEC 2.5099 1.4184 0.5651 6 
Tunisia OAPEC 2.6699 1.2394 0.4642 10 
Geomean   2.1083 1.3469 0.6389   
      
ANOVA F 14.469 10.065 421.286  
 Sig. .000 .000 .000  
      
Kruskal Wallis 𝝌𝟐 117.784 67.857 125.137  
 Sig. .000 .000 .000  
      
 
91 
 
University of Ghana  http://ugspace.ug.edu.gh
Next IGO under consideration is the OAPEC whose membership comprises only Arab oil 
and gas producing countries. Results of the intra group analysis of the 10 member states of 
OAPEC shown in Table 5.6 are based on a dataset from 2000 to 2013. For their group 
efficiencies, it can be observed that the efficiency scores of most of the countries in the 
group are moderately high and not very far from the absolute efficiency score of 1. Judging 
by means of the group efficiency scores, best performing countries include Saudi Arabia 
(1.0949), Kuwait (1.1299), Bahrain (1.1399), Algeria (1.1606) and Qatar (1.1669). This 
notwithstanding three countries in this IGO seem to have efficiency scores that are quite 
high. These countries are Iraq (2.4220), Egypt (1.8823) and Syria (1.4184). In addition, the 
meta-efficiency scores are quite high and quite away from the absolute efficiency score of 
1. There are quite small differences between the meta-efficiencies and group efficiencies 
giving an indication that countries in this IGO may be producing, to a large extent, close to 
the best state of knowledge in the industry. Even the technology gap ratio for all members 
except Tunisia (0.4642) and Algeria (0.4657) are above 0.525. The average technology gap 
ratio of all countries in the IGO is about 0.6389 signifying that on average countries in 
OAPEC produced using 63.89 percent of the existing state of knowledge in the industry. 
Ranking of TGRs, together with the TukeyHSD results in Appendix D, show that Kuwait 
and Qatar are the best performing countries in the IGO followed by Bahrain, and then Saudi 
Arabia. Tunisia and Algeria are trailing. Anova and Kruskal Wallis test of differences show 
some significant differences in the performances of individual countries in this IGO for all 
three indicators. 
The results show the performance of members to be moderate to high for almost all the 
member countries. OAPEC is made up of Arab countries having a distinct characteristic of 
being close in same location in the Middle East (Colgan et al., 2011). The objective of 
organization to support member countries in the effective use of their resources by 
92 
 
University of Ghana  http://ugspace.ug.edu.gh
sponsoring joint venture is very informative. This may have been a contributory factor to 
the performance of its member countries. Which is evident in the TGRs in Table 5.6. 
Although some member states have quite high TGRs, a few other states have lower than 0.5 
TGRs, this shows some level of dispersion in the performance of member states. This can 
be viewed through the lenses of the institutional theory. It is possible that the regulative, 
normative, and cognitive structures and activities that are required to provide stability for 
behavior of member states (Scott, 1995) may not be widely followed by all states. This is 
because the theory posits that, if members strictly followed the organizational structures and 
ideologies, we expect members to look and act the same  (DiMaggio & Powell, 1983). This 
is especially true because some members of the OAPEC group, such as Iraq, Libya, Saudi 
Arabia, Kuwait, Algeria, and Qatar, are also members of other similar IGOs like OPEC. 
This means that there can be the possibility of conflicts in institutional guidelines of the 
various IGOs, hence states may not follow the laid down policies in the group. This may be 
the reason for the differences in performances of states in this IGO. 
Table 5.7: Intra-Group Analysis of OPEC Member States     
Country IGO Meta Efficiencies Group Efficiencies TGR Rank 
Algeria OPEC 2.4920 1.0820 0.4342 10 
Angola OPEC 1.5269 1.1925 0.7810 4 
Ecuador OPEC 1.4769 1.1941 0.8085 3 
Iran OPEC 8.3536 2.5220 0.3019 12 
Iraq OPEC 4.2290 2.6755 0.6326 6 
Kuwait OPEC 1.1972 1.1282 0.9424 1 
Libya OPEC 2.4402 1.2646 0.5182 9 
Nigeria OPEC 5.5772 3.2812 0.5883 7 
Qatar OPEC 1.2729 1.1826 0.9290 2 
Saudi Arabia OPEC 1.4135 1.0934 0.7735 5 
UAE OPEC 2.1083 1.1957 0.5671 8 
Venezuela OPEC 5.4166 2.0236 0.3736 11 
Geomean   2.5228 1.5201 0.6026   
      
ANOVA F 118.877 77.475 218.370  
 Sig. .000 .000 .000  
      
Kruskal Wallis 𝝌𝟐 1473694 118.339 153.835  
 Sig. .000 .000 .000  
      
93 
 
University of Ghana  http://ugspace.ug.edu.gh
The final IGO under consideration is OPEC which has membership comprising oil and gas 
producing countries from the Middle East, Africa and South America. This is one of the 
foremost and most influential IGOs in the international oil industry (Ike & Lee, 2014) 
therefore their performance dynamics are very important. Results from the analysis of the 
12 member countries are detailed in Table 5.7. For their group efficiencies, it can be 
observed that the efficiency scores of most of the countries in the group are slightly high 
although not very far from the absolute efficiency score of 1. In this group, Algeria (1.0820), 
Saudi Arabia (1.0934), and Kuwait (1.1282) are among the best performers in terms of 
group efficiency. Other countries like Qatar, United Arab Emirates, Ecuador, and Angola 
have scores ranging from 1.1826 to 1.1957. For other countries like Venezuela (2.0236), 
Iran (2.5220), Iraq (2.6755), and Nigeria (3.2812), their group efficiency scores are very 
high showing high levels of inefficiency by these countries within the IGO. Meta-efficiency 
scores of OPEC countries seem also high. However, there are some countries with similar 
group efficiency scores and meta-efficiencies. Kuwait for example has a group efficiency 
score of 1.1282 whiles its meta- efficiency score is 1.1972. These two scores are not far 
apart signifying that for some countries in the IGO, they are producing using similar states 
of knowledge as available in the industry. For some other members of this IGO, however, 
their meta-efficiency scores are far apart from their group efficiencies. Algeria, for example 
which had the best group efficiency score of 1.0820 has a meta-efficiency score of 2.4920. 
Meaning that it does not adequately use the best available state of knowledge in its 
production efforts. A better sense of the disparities between the group efficiencies and the 
meta-efficiencies can be gained by reference to the technology gap ratios. For all member 
states, except Venezuela (0.3736), Iran (0.3019) and Algeria (0.4342), TGRs are above 0.50. 
The average technology gap ratio of all countries in the IGO is about 0.6026 signifying that 
on average countries in OPEC produced using about 60.26 percent of the existing state of 
94 
 
University of Ghana  http://ugspace.ug.edu.gh
knowledge in the industry. Ranking of TGRs show that Kuwait, Qatar, Ecuador, Angola 
and Saudi Arabia are the best producers in the IGO whilst Libya, Algeria, Venezuela and 
Iran are trailing. Anova and Kruskal Wallis tests of differences also reveal significant 
differences in the scores of these countries. 
OPEC member countries, unlike OAPEC, are a diverse set with many shared characteristics 
(Al-Rashed & León, 2015). These countries are at different stages of social and economic 
development with different economic structures (Al-Rashed & León, 2015). This may be 
the reason for the large variations in the TGR estimates of member countries as shown in 
Table 5.7. Countries like Kuwait, Qatar and Ecuador have scores above 0.8, whiles Angola, 
Saudi Arabia and Iraq have scores above 0.6. Iran, Algeria and Venezuela have TGRs below 
0.5. As an institution, it is expected that the member countries follow similar processes as 
laid down rules by OPEC. This is expected to result in similar performance levels as the 
proponents of the institutional theory postulate  (Lincoln, 1995). Differences in the 
performance is an indication that the members do not well follow the institutional policies 
and frameworks available. This may be the reason for the large differences in their TGR 
scores. For OPEC, evidence abound in literature to show that some members have exhibited 
loose adherence to announced production cutbacks (OPEC, 2002). 
From the regional point of view, OPEC countries can be categorised as belonging to one of 
four geographical regions: sub-Saharan African countries (Angola and Nigeria), North 
African countries (Algeria and Libya), Middle East countries (Iran, Iraq, Kuwait, Saudi 
Arabia, Qatar and UAE) and South American countries (Ecuador and Venezuela). This 
diversity in membership may also be the reason for the differences in TGRs since as the 
proponents of Social Network Theory argue, when members do not share similar social 
bonds performance may be different (Barnes, 1954; Bott, 1957; Granovetter, 1973). The 
main purpose of OPEC is to protect the collective interests of individual member states and 
95 
 
University of Ghana  http://ugspace.ug.edu.gh
to ensure balance in their international oil market transactions (Desta, 2003). But there 
seems to be a lack of co-ordination on broader policy issues impacting on the common 
interests of members (Arena, 2008).  This has resulted in most countries hardly adhering to 
the organization’s policy leading to them adopting their own development in dealing with 
multilateral transaction (Desta, 2003). Hence the large variation their performance and 
efficiency score. This is true of the social network theory which argues that there are both 
positive and negative consequences of membership in social network. The theory explains 
that states are said to be socially networked when they tend to think and behave similarly 
because they are connected. However, because the ties between states in the group may be 
weak members may not have the same level of success. Again issues of internal collective 
action challenges and the rise of new producers can explain the differences in their 
performance (Goldthau & Witte, 2011). The practice of supply restrictions of production 
output may be the reason for low performance of some of its members (Desta, 2003; Ike & 
Lee, 2014) like Venezuela and Angola. 
Inter-Group Performance Evaluation 
This part analysis the inter-group performances of the IGOs in this study. Tables 5.8 
summarizes the results of metafrontier analysis with respect to the four IGOs under study. 
The average meta-efficiencies, group efficiencies and technology gap ratios of each IGO is 
first presented in Table 5.8. Values in this table are then presented graphically in Figure 5.1 
for better conceptualization of deductions. IGOs are also ranked based on their TGRs to 
identify the best performing ones in order of importance. 
  
96 
 
University of Ghana  http://ugspace.ug.edu.gh
Table 5.8: Inter-IGO Metafrontier Results 
IGO Meta Efficiencies Group Efficiencies TGR Rank 
FSU 6.0057 2.2911 0.3815 4 
IEA 1.9140 1.6255 0.8493 1 
OAPEC 2.1083 1.3469 0.6389 2 
OPEC 2.5228 1.5201 0.6026 3 
 
Results of the inter-group performance comparison of the four IGOs under consideration as 
detailed in Table 5.8 are based on the dataset from the period 2000 to 2013. The group 
efficiency scores of member states of each of these IGOs as computed are quite high. Since 
the group efficiencies are computed relative to different production frontiers, it is not 
appropriate to compare the group’s specific efficiency scores of these different groups 
(Canhoto & Dermine, 2003; Dietsch & Lozano-Vivas, 2000). It can however be noted that, 
whereas OAPEC member states are producing at about 74% (1/1.3469 = 0.74) of their 
potential capacity, FSU states are only producing at 44% (1/2.2911 = 0.44) of their potential 
production capacity. IEA and OPEC states are producing at 62% and 66% of their potential 
production capacities respectively. The meta-efficiency scores, can however, be compared 
since they are all based on the same pooled frontier. Meta-efficiencies are quite high 
indicating more inefficiencies since most scores are away from the efficiency score of 1. 
The IEA (1.9140) is the only IGO with a meta-efficiency score that is quite similar to their 
group efficiency score.  The differences between the meta-efficiencies and group 
efficiencies of OAPEC and OPEC are fairly high but not too far away from each other. 
Again, scores for FSU seem to show quite high differences between the two efficiency 
scores. This is an indication that, as compared to other IGOs, IEA member states are 
producing using the best state of knowledge in the industry since their TGR at 0.8493. 
OAPEC and OPEC are producing, to some extent, close to the best state of knowledge in 
97 
 
University of Ghana  http://ugspace.ug.edu.gh
the industry with TGRs slightly above 0.60. FSU could only manage 0.3815 of the state of 
knowledge available in the international oil industry.  Also, the TGRs of all IGOs except 
FSU (0.3815) are above 0.6026. This means that given the inputs IEA is producing at 
84.93% close to the available state of production technology in the international oil industry. 
OAPEC and OPEC are producing at 63.89% and 60.26% respectively. 
 
7.00
6.00
5.00
4.00
3.00
2.00
1.00
0.00 FSU IEA OAPEC OPEC FSU IEA OAPEC OPEC FSU IEA OAPEC OPEC
Meta Efficiencies Group Efficiencies TGR
 
Figure 5.1: Distribution of Metafrontier Scores of the IGOs 
 
Figure 5.1 presents a graphical depiction of the distribution of the metafrontier scores of the 
IGOs being considered. The graphical view allows for easy understanding of how different 
scores of specific IGOs are from other ones. The Figure shows the meta-efficiencies, group 
efficiencies as well as the technological gap ratio of these IGOs. From Figure 5.1, it is clearly 
seen that the meta-efficiency score, on average, of FSU is higher than all other IGOs, 
signifying higher levels of inefficiencies by member state relative to the best possible 
technology in the industry as compare with member states of other IGOs. IEA has the closest 
98 
 
EFFICIENCY SCORES/TGR
University of Ghana  http://ugspace.ug.edu.gh
meta-efficiencies, on average, to the efficiency score of 1.  When each IGO’s meta-
efficiencies are compared with the group efficiency scores, it is clear that unlike other IGOs, 
the meta and group efficiencies of IEA states are quite similar. Disparity between the meta-
efficiency and group efficiency scores of FSU is much more pronounced. The result is that 
FSU has the lowest TGR. IEA, however, has clearly, the highest technology gap ratio, 
followed closely by OAPEC and OPEC. What is not obvious in these inter-group 
comparison is whether differences in the meta-efficiencies and TGRs of these IGOs are 
statistically significant. This is the basis for the pairwise comparisons presented in Table 
5.9. 
To achieve the third objective of this study, a pairwise comparison of the scores of these 
IGOs are presented based on a dataset from 2000 to 2013. The performance of the four IGOs 
are first compared using traditional point estimate comparison statistical techniques. 
Independent t-test and Mann Whitney U tests are used to conduct pairwise comparison of 
the means and ranks of the various IGOs. Whiles these tests are well known, it only 
compares point estimates and neglect the distribution of the entire dataset (Li, 1996; Simar 
& Zelenyuk, 2006). To cater for this weakness, the SZAL test, which uses kernel density 
estimators to compare the distribution of the scores are used here. Notice that in Table 5.9, 
test statistics are presented together with p-values in parenthesis. Notice also that group 
efficiencies are not compared in this table since the group efficiency scores of the various 
IGOs are based on separate production frontiers.  
  
99 
 
University of Ghana  http://ugspace.ug.edu.gh
Table 5.9: Pairwise Comparisons of Inter-IGO Performance 
    T-test  Mann Whitney  SZAL 
FSU - IEA Meta Eff. 6.288 (0.000)***  7462 (0.000)***  27.75 (0.000)*** 
 TGR -21.997 (0.000)***  3080 (0.000)***  67.38 (0.000)*** 
       
FSU - OAPEC Meta Eff. 6.114 (0.000)***  4390 (0.000)***  16.77 (0.000)*** 
 TGR -9.854 (0.000)***  3129 (0.000)***  33.28 (0.000)*** 
       
FSU – OPEC Meta Eff. 5.706 (0.000)***  6865 (0.000)***  17.45 (0.000)*** 
 TGR -8.602 (0.000)***  5326 (0.000)***  19.14 (0.000)*** 
       
IEA – OAPEC Meta Eff. -2.464 (0.015)*  18015 (0.261)  4.199 (0.000)*** 
 TGR 11.055 (0.000)***  7804 (0.000)***  30.78 (0.000)*** 
       
IEA – OPEC Meta Eff. -6.021 (0.000)***  18624 (0.001)**  2.992 (0.001)** 
 TGR 12.365 (0.000)***  8887 (0.000)***  35.77 (0.000)*** 
       
OAPEC - OPEC Meta Eff. -3.654 (0.000)***  9746.5 (0.031)*  2.88 (0.002)** 
 TGR 1.219 (0.224)  10399 (0.194)  14.80 (0.000)*** 
       
***p <0.001.   **p<0.01.  *p<0.05  
values in parenthesis ( ) are the p-values 
 
First pairwise comparison in between the scores of FSU and IEA. This is a comparison of 
the two extremes, since previous results from Table 5.8 and Figure 5.1 revealed that FSU 
has the highest meta-inefficiency scores of 6.0057 and lowest TGR of 0.3815 whiles IEA 
had the best meta-efficiency score and TGR of 1.9140 and 0.8493 respectively. Statistical 
comparisons for all three estimators reveal significant differences in the meta-efficiencies 
and TGRs of FSU and IEA at the 0.1% significance level. IEA member states significantly 
outperform their FSU counterparts on both meta-efficiency and TGR. FSU states are next 
compared with OAPEC member states. Conclusions on both the meta-efficiencies and the 
TGR are similar to that revealed when FSU and IEA were compared. OAPEC states, on 
average, outperform FSU states on both indicators and on all three estimators of difference. 
OPEC member states also significantly outperform FSU states on both meta-efficiency and 
TGR. Results from the metafrontier analyses therefore reveal that FSU is the least 
performing IGO in the international oil industry. 
100 
 
University of Ghana  http://ugspace.ug.edu.gh
Next, results for IEA are compared with that of OAPEC. From Figure 5.1, IEA is seen to 
have lower meta-inefficiencies but higher TGR on average. The question is however, 
whether these differences are statistically significant. Starting with the meta-efficiency 
scores, mixed results are observed between the results of the parametric t-test and its 
nonparametric counterpart- Mann Whitney test. Whereas results from the t-test reveal that 
IEA states have significantly lower average meta-inefficiencies than OAPEC states (t = -
2.464, p < 0.05), Mann Whitney shows no statistically significant differences in the ranks 
on these two IGOs (U = 18015, p = 0.261). It is amidst these disparities in conclusions that 
the utility of the SZAL test is seen. SZAL results show statistically significant differences 
in the distribution of meta-efficiency scores of IEA and OAPEC states (l = 4.199, p< 0.001). 
Results of the SZAL test is more reliable since it compares all members of one group against 
all members in the other. It is also based on nonparametric techniques which are important 
because of the nonparametric nature of DEA estimation technique. For the TGR, all three 
estimation techniques observe significant differences in the TGRs of IEA and OAPEC 
states. IEA states therefore significantly outperform OAPEC counterparts. This can be 
graphically observed with reference to Figure 5.2 which shows the kernel density 
distribution of the TGRs of the four IGOs. From Figure 5.2 shows it is clear that whereas 
the distribution of scores for IEA seem to gain more density towards the score of 1.0, that 
of OAPEC seem to peak between 0.4 and 0.6. 
101 
 
University of Ghana  http://ugspace.ug.edu.gh
 
Technology Gap Ratios 
 
Figure 5.2: Distribution of Technological Gap Ratios 
 
IEA states are then statistically compared with OPEC states. Results are quite 
straightforward. There are statistically significant differences in the means, ranks and 
distributions of these two IGOs for both the meta-efficiency scores and TGRs. IEA has 
significantly lower meta-inefficiencies than OPEC and higher TGRs than OPEC on average. 
IEA states therefore significantly outperform OPEC states on both meta-efficiency scores 
and TGRs. 
From deductions made, it is clear that FSU is the worst performing IGO based on the 
metafrontier analyses, whiles IEA is the best IGO. What is not clear is whether any 
differences exist between the performances of OAPEC and OPEC. Both IGOs were seen in 
Table 5.8 to have meta-efficiencies slightly below 2.60 and TGRs slightly above 0.6. 
Although OAPEC has larger TGR of 0.6389 compared to OPEC’s of 0.6026 and smaller 
meta-efficiency of 2.1083 compared with OPEC’s of 2.5228, there question remains 
whether there are statistically significant differences. Just like previous comparisons, scores 
102 
 
University of Ghana  http://ugspace.ug.edu.gh
of OAPEC and OPEC are statistically compared on all three estimators of differences. For 
the meta-efficiency scores, all three statistical approaches revealed significant differences 
between the scores of OAPEC and OPEC. OAPEC therefore has lower meta-inefficiencies 
compared with OPEC states on average. The result is not that straightforward when the 
TGRs are compared. For both point estimators, no statistically significant differences were 
observed between the scores of these two IGOs. There are no statistically significant 
differences in the means (t = 1.219, p = 0.224) and ranks (U = 10399, p = 0.194) of TGRs 
of OAPEC and OPEC member states. This notwithstanding, there is a statistically 
significant difference between the distribution of OAPEC and OPEC states (l = 14.80, p < 
0.001). This can be inferred from Figure 5.2 where it is seen that whereas OAPEC gathers 
greater mass between 0.4 and 0.6 as well as between 0.8 and 1.0, OPEC states seem to be 
distributed relatively evenly across a wider range of scores. There are even quite a number 
of OPEC states with scores lower than the 0.4 mark whiles only few OAPEC members fall 
in this sector. OAPEC states therefore seem to outperform OPEC states in this regard. 
Statistical tests therefore reveal that IEA is the best performing IGO followed by OAPEC. 
OAPEC is closely followed by OPEC states. FSU is, however, in a distant fourth place on 
the ranking. 
The inadequate level of cooperation among governments of FSU countries probably 
underscores the reasons for its inefficiencies. There is no agreement coupled with 
differences in political and economic interest in the development of the oil resource in their 
respective countries (Aguilera, 2012). These essential ties, cooperation or networks  as 
expounded by social network theory provides positive utility to countries in the oil industry, 
by influencing access to resources, reducing transactional costs and building interest based 
on coalitions (Lauber et al., 2008). However, absence of higher levels of such cooperation 
may be detrimental to the group interest since differences in the individual goals may not 
103 
 
University of Ghana  http://ugspace.ug.edu.gh
engender group performance.  Another issue that probably impacts on the higher 
inefficiency of members in this IGO is the concentration on internal use of the oil and gas 
produced by the FSU countries (EIA, 2013). Excessive government subsidies on the oil and 
gas supplied for domestic market can be a disincentive to higher production levels since 
economic benefits from higher production levels may not be realised by the producing 
organisations. Additionally, it is possible that because most oil produced in this region is 
heavy (EIA, 2013; Goldemberg, 2000), it is a contributory factor to their inefficiencies. 
Heavy oil requires enhanced oil recovery techniques. This stands to reason that since higher 
production technology is required for exploring heavy oil, as social network theory explains, 
countries stand to benefit from collaborations in terms of technology and research as well 
as reducing the transactional cost. Therefore the poor performance can be associated to lack 
of collaboration in technology and research in the industry (Goldemberg, 2000).  
The progress of IEA member countries may be explained by the distinct policy of 
collaboration by the IGO with groups across the international oil and gas industry 
(Bamberger et al., 2004). The performance of the group members can be attributed to the 
framework that allows for consultation with oil companies by member countries, to establish 
permanent basis for consulting in an appropriate manner for the request for information 
from individual oil companies on all important aspect of the oil industry. The agency allows 
participating countries to endure to promote cooperate relations with oil producing countries 
and consuming country. This probably has created ample opportunities for promoting 
dialogue and solution for members’ unique challenges. For OAPEC, the organization’s 
assistance in facilitating access to a more scientific and new technological developments in 
the industry could also account for their performances (OAPEC, 2015). OAPEC maintains 
its international collaboration with similar organizations by organizing and participation in 
expert scientific seminars particularly those concerned with regional and international 
104 
 
University of Ghana  http://ugspace.ug.edu.gh
energy affairs (OAPEC, 2015) to improve on the group’s performance which is consistent 
with social facilitation theory. OPEC states are not far behind OAPEC on average. Their 
results are much more mixed in nature since whereas some OPEC states are performing 
quite well, other OPEC states seem to suffer because of the membership of this IGO. It is 
quite difficult to believe that OPEC states do not follow organizational guidelines and there 
is absence of policies by the IGO that governs the oil and gas exploration and production 
activities of member states like the FSU. Indeed, the several production quotas that critically 
affect the world oil prices is ample evidence of standardised organisational policies. Policies 
that consider global supply and associated prices have always been critical targets for OPEC 
members (Wolf, 2009). It stands to reason that, although there exist such policies for OPEC, 
and member states follow them, although sometimes loosely, these policies may not be ones 
that adequately benefit member states. Ike and Lee (2014) for example observe that 
production quotas of OPEC significantly affect the performance of OPEC states. Similarly, 
Desta (2003) believes that the practice of supply restrictions of production output may be 
the reason for low performance of some of its members. 
 
5.4 Global Frontier Differences Analysis of Inter-IGO Performance 
Section 5.4 provides empirical reasons to help achieve objectives four and five of this study. 
Whiles objective four evaluates the inter-group frontier differences of the four IGOs, 
objective five determines whether there are statistically significant differences in the 
distribution of the technological indices (TI) between the various IGOs in the international 
oil industry. First, technology indices are estimated for all the IGOs. This is then followed 
by the frontier differences with respect to each individual IGO. Finally, results for the 
statistical comparison of the TIs for the various IGOs are presented. 
105 
 
University of Ghana  http://ugspace.ug.edu.gh
The technological index, as expressed in equation (7) of the forth chapter of this work, 
measures the geometric mean of the efficiencies of all the observations relative to the 
frontier of the particular group under consideration (Asmild & Tam, 2007). In an output 
orientation, where inefficient DMUs have scores greater than 1, and superefficient DMUs 
have scores less than 1, when the TI of a particular group’s frontier is greater than one, it 
means that, on average that frontier is better than most of the observations. However, if the 
TI is less than 1, it means that the frontier is worse than most observations since on average 
DMUs are superefficient (score < 1 in output oriented model). In short, higher TIs signify 
better frontier on average. The average TI for each IGO is presented in Table 5.10 for each 
year from 2000 to 2013. The average for the pooled dataset is also included.   
Table 5.10: Technology Indices for the IGOs 
Tech Indices Tech Indices Tech Indices Tech Indices 
 (OPEC)  (IEA) (FSU) (OAPEC) 
2000 0.75527 1.96932 0.62074 1.02384 
2001 0.74587 2.00257 0.62470 0.99897 
2002 0.75488 2.03378 0.63499 0.97886 
2003 0.84372 2.05208 0.65695 1.00207 
2004 0.78535 1.97307 0.61539 0.95344 
2005 0.76513 1.91421 0.60085 0.91389 
2006 0.76211 1.93629 0.60752 0.90892 
2007 0.76679 2.15787 0.64191 0.97376 
2008 0.74015 2.17450 0.64533 0.97248 
2009 0.77802 2.12043 0.64587 0.95175 
2010 0.76870 2.10701 0.65830 0.95040 
2011 0.82907 2.24633 0.72576 1.04132 
2012 0.80028 2.14188 0.70450 0.98054 
2013 0.84220 2.24973 0.72976 1.03601 
Pooled 0.78055 2.07384 0.64950 0.97680 
 
It is obvious from Table 5.10 that, for each year, IEA has had the highest TI. The TI for IEA 
has consistently been greater than 1, signifying that most observations (especially from the 
other IGOs) have been inefficient relative to the IEA frontier. Although the score fluctuates 
above 2.00 mark, it declined below 2.00 in 2004, 2005, 2006. It however regained 
momentum in 2007, and has maintained TI scores above 2.10 mark since then. However, 
106 
 
University of Ghana  http://ugspace.ug.edu.gh
even in the periods that IEA’s indices were lower 2.00 mark, its TI was still better than all 
the other IGOs. IEA’s worse TI of 1.9142 is even higher than OAPEC’s best of 1.04132. 
OAPEC seems to have technological indices fluctuating around 1.00. It has in effect had 
periods where its frontier is better than most observations, on average and periods where its 
frontier has not been that good on average. Whiles the TI of OPEC ranges from 0.7459 to 
0.8438, the TIs of FSU have had the lowest levels of the technological indices having 
achieved their highest index of 0.7298 in 2013. FSU’s largest TI is even lower than the 
lowest of OPEC’s TIs.  For the pooled dataset, on average, whereas IEA has the highest 
average of 2.07384 and OAPEC has a score of 0.97689, OPEC and FSU have scores of 
0.78055 and 0.64950 respectively. These scores have been graphically presented in Figure 
5.3. 
From Figure 5.3, it is obvious how far apart IEA’s scores are from the other IGOs. IEA 
technological indices appear to fluctuate, however on the whole it seems to be on an upward 
spiral. The scores for the FSU, OPEC and OAPEC seem quite close. They seem to be 
fluctuating around the same scores but there is no real progress in the indices over the period. 
Another observation is that, at no point do two curves intersect. Performance of these IGO’s 
seem to have remained stable over the years.  
107 
 
University of Ghana  http://ugspace.ug.edu.gh
Tech Indices (OPEC) Tech Indices (IEA)
Tech Indices (FSU) Tech Indices (OAPEC)
2.5
2
1.5
1
0.5
0
YEAR
 
Figure 5.3: Distribution of Technological Indices 
Whereas the TI will give an indication of how good the frontier is, it provides no real 
indication of how well a particular group’s frontier is as compared to that of another specific 
group. This is where the global frontier differences (GFD) approach of Asmild and Tam 
(2007) gains utility. The TI are used in the estimation of the global frontier differences 
between the different groups. Results of the global frontier difference are presented 
subsequently. GFD measures and provides the overall conclusions about whether one group 
is superior to the other (Asmild & Tam, 2007). This approach provides an overall estimation 
of the differences between two frontiers or more importantly between groups. The frontier 
can differentiate between the efficiencies of two frontiers or two groups without considering 
shift overtime. The GFD tells by how much a particular IGO is far away from or close to 
another IGO. Mathematical notations of the index are presented in equation (8) of the fourth 
chapter of this work. Results are presented in Tables 5.11, 5.12, 5.13 and 5.14.  
 
108 
 
TECHNOLOGY INDEX
University of Ghana  http://ugspace.ug.edu.gh
Table 5.11: Global Frontier Differences against OPEC Frontier 
 YEAR OPEC IEA FSU OAPEC 
2000 1.0000 2.6075 0.8219 1.3556 
2001 1.0000 2.6849 0.8375 1.3393 
2002 1.0000 2.6942 0.8412 1.2967 
2003 1.0000 2.4322 0.7786 1.1877 
2004 1.0000 2.5123 0.7836 1.2140 
2005 1.0000 2.5018 0.7853 1.1944 
2006 1.0000 2.5407 0.7972 1.1926 
2007 1.0000 2.8142 0.8371 1.2699 
2008 1.0000 2.9379 0.8719 1.3139 
2009 1.0000 2.7254 0.8301 1.2233 
2010 1.0000 2.7410 0.8564 1.2364 
2011 1.0000 2.7095 0.8754 1.2560 
2012 1.0000 2.6764 0.8803 1.2252 
2013 1.0000 2.6713 0.8665 1.2301 
Pooled 1.0000 2.5145 0.8010 1.2582 
 
Table 5.11 presents GFDs between OPEC and all other IGOs. In other words, it presents 
how the other IGOs view the OPEC frontier over the period. First using the pooled results, 
on average, IEA and OAPEC are 151.45% ([2.51451]100)  and 25.82% 
([1.25821]100)  better than OPEC frontier, whereas FSU on average is 20% 
([10.80]100)  worse than OPEC. These overall dynamics are observed across all periods 
under review. IEA and OAPEC have consistently been better than OPEC from the results, 
whiles FSU always underperform the OPEC frontier. 
Observing results from Table 5.12 which assesses how the other IGOs view the IEA frontier, 
it is readily seen from the GFDs that the IEA frontier is far better than that of all other IGOs 
under study. This is because, for the entire study period, all three other IGOs reported GFDs 
lower than 1 relative to the IEA frontier. Overall, FSU recorded the worse overall 
performance with respect to the IEA frontier. FSU frontier was, on average, 68.68% 
([10.3132]100)  worse than IEA’s frontier. This was closely followed by OPEC which 
had a GFD of 0.3763 signifying that OPEC frontier was on average 62.37% 
109 
 
University of Ghana  http://ugspace.ug.edu.gh
([10.3763]100)  worse than IEA’s frontier. The closest group to the state of technology 
employed by IEA was OAPEC, which had a frontier 52.91% worse than IEA’s.  
Table 5.12: Global Frontier Differences against IEA Frontier 
YEAR OPEC IEA FSU OAPEC 
2000 0.3835 1.0000 0.3152 0.5199 
2001 0.3725 1.0000 0.3119 0.4988 
2002 0.3712 1.0000 0.3122 0.4813 
2003 0.4112 1.0000 0.3201 0.4883 
2004 0.3980 1.0000 0.3119 0.4832 
2005 0.3997 1.0000 0.3139 0.4774 
2006 0.3936 1.0000 0.3138 0.4694 
2007 0.3553 1.0000 0.2975 0.4513 
2008 0.3404 1.0000 0.2968 0.4472 
2009 0.3669 1.0000 0.3046 0.4488 
2010 0.3648 1.0000 0.3124 0.4511 
2011 0.3691 1.0000 0.3231 0.4636 
2012 0.3736 1.0000 0.3289 0.4578 
2013 0.3744 1.0000 0.3244 0.4605 
Pooled 0.3763 1.0000 0.3132 0.4709 
 
 
Table 5.13: Global Frontier Differences against FSU Frontier 
YEAR OPEC IEA FSU OAPEC 
2000 1.2167 3.1725 1.0000 1.6494 
2001 1.1940 3.2057 1.0000 1.5991 
2002 1.1888 3.2029 1.0000 1.5415 
2003 1.2843 3.1237 1.0000 1.5253 
2004 1.2762 3.2062 1.0000 1.5493 
2005 1.2734 3.1858 1.0000 1.5210 
2006 1.2545 3.1872 1.0000 1.4961 
2007 1.1945 3.3617 1.0000 1.5170 
2008 1.1469 3.3696 1.0000 1.5069 
2009 1.2046 3.2831 1.0000 1.4736 
2010 1.1677 3.2007 1.0000 1.4437 
2011 1.1423 3.0951 1.0000 1.4348 
2012 1.1360 3.0403 1.0000 1.3918 
2013 1.1541 3.0828 1.0000 1.4197 
Pooled 1.2014 3.1928 1.0000 1.5035 
 
110 
 
University of Ghana  http://ugspace.ug.edu.gh
Results from Table 5.13 show the GFDs relative to the FSU frontier. It shows how the other 
IGOs view the FSU frontier.  From the results, it is evident that FSU frontier is the worse 
frontier among the groups under study. All other IGOs scored values greater than 1. IEA 
reported the strongest performance with an overall average of 3.1928, whereas OAPEC had 
an overall GFD of 1.5035. For OPEC, its score of 1.2014 on average means that OPEC’s 
frontier is about 20.14% better than FSU’s frontier. The final of the global frontier difference 
comparisons is relative to the OAPEC frontier. These scores are reported in Table 5.14. 
OAPEC’s frontier shows some interesting results. Whereas on average, OAPEC’s frontier 
seems better than the OPEC (0.7991) and FSU (0.6651) frontiers, it is not better than that 
of IEA (2.1236). 
Table 5.14: Global Frontier Differences against OAPEC Frontier 
YEAR OPEC IEA FSU OAPEC 
2000 0.7377 1.9235 0.6063 1.0000 
2001 0.7466 2.0046 0.6253 1.0000 
2002 0.7712 2.0777 0.6487 1.0000 
2003 0.8420 2.0478 0.6556 1.0000 
2004 0.8237 2.0694 0.6454 1.0000 
2005 0.8372 2.0946 0.6575 1.0000 
2006 0.8385 2.1303 0.6684 1.0000 
2007 0.7875 2.2160 0.6592 1.0000 
2008 0.7611 2.2360 0.6636 1.0000 
2009 0.8175 2.2279 0.6786 1.0000 
2010 0.8088 2.2170 0.6927 1.0000 
2011 0.7962 2.1572 0.6970 1.0000 
2012 0.8162 2.1844 0.7185 1.0000 
2013 0.8129 2.1715 0.7044 1.0000 
Pooled 0.7991 2.1236 0.6651 1.0000 
 
From all the GFDs reported in the four preceding tables, IEA’s frontier has consistently 
been better than that of all other IGOs. IEA’s frontier on average is 2.5145 times better than 
OPEC’s frontier, 3.1928 times better than FSU frontier and 2.1236 times better than 
OAPEC’s frontier. This seems to give credence to earlier observation that IEA is the best 
performing IGO in terms of oil and gas production efficiency in the international oil 
111 
 
University of Ghana  http://ugspace.ug.edu.gh
industry. Best performing countries in IEA are much better than even the best performing 
countries in all other IGOs under study. OAPEC seems to be the next best IGO based on the 
GFDs. Its frontier, although 62.37% ([10.3763 ]100)  worse than IEA’s frontier, is 
25.82% ([1.25821]100)  better than OPEC frontier and 50.35% ([1.50351]100)  better 
than FSU frontier. OPEC then follows in terms of rankings. Its frontier is seen to be better 
than that of FSU whiles trailing those of IEA and OAPEC. Finally, the frontier of FSU states 
is seen to be the worse relative to all other IGOs in the international oil industry. 
The final part of this section is aimed at achieving the fifth and final objective of this study. 
This is to examine whether a statistically significant difference exist in the technology 
indices of the various IGOs. Whereas the global frontier differences will inform on the 
difference between specific frontiers, it does not inform on whether the differences between 
two frontiers are statistically significant. Therefore, three tests of differences are used in this 
part. In achieving objective five of this study, a pairwise comparison of the scores of these 
IGOs are presented based on a dataset from 2000 to 2013. Independent t-test and Mann 
Whitney U tests are used to measure the pairwise comparison of the means and ranks of the 
various IGOs. Additionally, the SZAL test which uses kernel density estimators to compare 
the distribution of the scores are used here. Test statistics are for each statistical technique 
are presented together with the p-values for each pairwise comparison.   
  
112 
 
University of Ghana  http://ugspace.ug.edu.gh
Table 5.15: Pairwise Comparisons of Technology Indices 
  t-test   Mann Whitney   SZAL 
FSU - IEA -45.262 (0.000)***  105 (0.000)***  6.11 (0.000)*** 
      
FSU - OAPEC -21.034 (0.000)***  105 (0.000)***  7.72 (0.000)*** 
      
FSU – OPEC -9.019 (0.000)***  105 (0.000)***  7.58 (0.000)*** 
      
IEA – OAPEC 34.985 (0.000)***  105 (0.000)***  7.10 (0.000)*** 
      
IEA – OPEC 41.924 (0.000)***  105 (0.000)***  6.10 (0.000)*** 
      
OAPEC - OPEC 13.755 (0.000)***  105 (0.000)***  7.75 (0.000)*** 
      
***p <0.001.   **p<0.01.  *p<0.05  
values in parenthesis ( ) are the p-values 
Pairwise comparison begin with the scores of FSU and IEA. Based on results in Table 5.10, 
IEA was seen to have the greatest TI than all other IGOs. Statistically, IEA’s higher TIs are 
also statistically significant based on all three statistical techniques. IEA also has 
significantly larger TIs than OAPEC and OPEC as all of these comparisons are statistically 
significant, IEA is therefore the best IGO in this industry. OAPEC is also seen to have 
significantly larger TIs than OPEC and FSU whiles it has a statistically significantly lower 
TI than IEA. OAPEC is therefore the second best IGO in this industry. This is then followed 
by OPEC which has statistically larger TIs than FSU but statistically lower than OAPEC 
and IEA. Finally, just as previously observed, FSU is seen to statistically underperform all 
other IGOs in industry since their TIs are seen to be significantly lower than all three other 
IGOs. 
The technological index comparison shows the magnitude of differences in the frontier in 
relations to the four IGOs being evaluated. IEA is seen to have higher technological indices 
and higher production frontier relative to the three other IGOs. The difference may be due 
to fact that the IEA is made up of more industrialized western countries with much higher 
resource in terms of human capital, technology, infrastructure and capital for investment, 
political and economic stability, bargaining power and collaborations with many more 
113 
 
University of Ghana  http://ugspace.ug.edu.gh
intergovernmental organizations (Bamberger et al., 2004; Colgan et al., 2011; Duffield, 
2012; Jollands et al., 2010). This is theoretically supported by the resource base theory 
(RBT), which believes that the way an organization is organized combined with it resources, 
can better enhance the positive relationship between resources and the performance of the 
organization. This is justified by the empirical works which give credence to the importance 
of these resource characteristics for firm performance (Crook et al., 2008; Sirmon et al., 
2011). This means that the IEA is able to put their resources to a better use for high 
performance. With respect to game theory the IEA strategy of maintaining minimum oil and 
gas stocks in the face of production cuts and price hikes (Bamberger et al., 2004) as well as 
the strong inter and intra group collaboration is consistent with the theory. When there is 
high inter and intra group collaborations it results in high exchange of information within 
and between groups. It then leads to organizational learning, innovation and information 
asymmetries are addressed (Rigby et al., 2013). 
OAPEC’s frontier showed a better score as compared with OPEC and FSU. This is an all 
Arab organization and perhaps this is the reason for their good performance. The group is 
seen to share similar cultural and political values due to them being closely linked with each 
other and in the same geographical location. This is buttressed by social network theory 
which posits that, states are said to be socially networked when they tend to think and behave 
similarly because they are connected (Garton et al., 1997; Miles, 2012). OAPEC sponsored 
ventures also help them to keep pace with developments and succeeds in enhancing their 
performance in the industry (OAPEC, 2010). This is true as predicted by game theory that 
stronger and more strategic the group collaboration within and with other groups, the higher 
the levels of learning and innovation and this leads to improved performance. OAPEC’s oil 
reserves have been estimated at about 713 billion barrels about 43% of the proven world’s 
reserves in 2014 (OAPEC, 2015). The large oil and gas resource endowment may be another 
114 
 
University of Ghana  http://ugspace.ug.edu.gh
factor influencing their level of performance as predicted by the RBT. The theory indicates 
that, there is a link between organization resources and performance. The more resource 
endowed an organization is, the better its performance. Overall, during the entire period of 
the study, OAPEC countries displayed almost similar scores indicating that the level of 
influence by member countries are relatively the same. This is justified by the theory of 
social facilitation which postulates that members of an organization are expected to exhibit 
similar levels of performance  (Crawford, 1939; Miles, 2012).  
OPEC’s frontier follows OAPEC as the next best production frontier with scores better than 
the FSU. The composition of the members cuts across countries from Africa, South America 
and Asia. OPEC’s coordination between major producers and consumers and their 
participation in international energy forums show the extent of engagement within the group 
and among other groups in the international oil and gas industry (Goldthau & Witte, 2011). 
It stand to reason that their better performance as compared with FSU may be a result of 
this collaborations, solidarity and other factor favouring the group in the coalition among 
member countries (Mikdashi, 1974). This is supported by game theory (Von Neumann & 
Morgenstern, 1944), which explains that, where there is a strategic collaboration within a 
particular group and among other groups there is a group learning and innovation that is 
possible to enhance the group’s performance. Since the FSU frontier is significantly below 
the production frontiers of the other IGOs, the suggestion is that there is inadequate levels 
of coordination and collaboration among member countries and other groups. It will 
therefore be critical for them to come together and form a stronger IGO to create stronger 
and more sustainable ties among them. As suggested by social network theory. The establish 
ties will help improve their performance. There is probably information asymmetry among 
member states since the levels of cooperation among FSU states is not adequate. Hence, 
115 
 
University of Ghana  http://ugspace.ug.edu.gh
their coming together will to some extent help improve their efficiencies significantly and 
contribute to individual country’s stronger development and growth.   
 
5.5 Conclusions 
In this chapter, results of the metafrontier and global frontier difference analysis were 
presented. This was earlier preceded by an examination of the returns to scale properties of 
the international oil and gas industry. Discussions based on theory, empirical literature and 
practice are provided to answer the research questions of the study. The analysis compared 
the production and supply efficiencies of the 53 countries in the four IGOs for the period.  
  
116 
 
University of Ghana  http://ugspace.ug.edu.gh
CHAPTER SIX 
SUMMARY, CONCLUSIONS AND RECOMMENDATIONS 
6.0 Introduction 
This is the final chapter of this study. It first summarises the research process and findings. 
Based on the findings, conclusions are drawn which are used to make recommendations for 
practice, policy and further research. 
 
6.1 Summary of study 
The main aim of this study was to examine the production and supply efficiency of OPCs 
belonging to four IGOs in the international oil industry. Many oil and gas efficiency related 
studies have been conducted, but little knowledge on inter-country and group level 
efficiency exist. The contributions of this research is first to add to the literature in the 
international oil and gas industry by exploring knowledge on the performance of OPCs. In 
addition to this, the study does not only assesses the nexus between OPCs performance and 
their association with a particular IGOs but also does inter-IGO performance using two DEA 
estimating techniques- the Metafrontier Analysis (Battese et al., 2004) and the Global 
Frontier Differences (Asmild, Paradi, Reese, & Tam, 2007) approaches. The within group 
(intra IGO) and again between groups (inter IGO) comparison have seen little or no 
consideration by most authors in the oil and gas industry. The study employed the 
metafrontier approach which interestingly caters for the challenge of group heterogeneity, 
and the global frontier difference which compares the frontiers of the various groups in the 
international oil and gas industry for the first time. Also, the Simar-Zelenyuk adapted Li 
test, Simar and Wilson (2002) test of returns to scale and other innovative modelling 
approaches are used to ensure robust research conclusions. Output-orientation was used for 
all modelling processes. This model allowed management to distinguish results based on 
117 
 
University of Ghana  http://ugspace.ug.edu.gh
policy and strategy and industry knowledge regarding increasing production levels from a 
given level of resource.  
Data used was source from the U.S Energy Information Association’s web site and the 
World Bank’s World Development Indicators database. In all 53 OPCs from the four IGOs 
in the international oil and gas industry for the 14-year period from 2000 to 2013 were 
sampled. The scope of the database comprises country level data on oil and gas reserves and 
production outputs. The main findings identified in the study are: 
i. The production technology of oil and gas producing nations that were sampled, 
seems to exhibit a constant return to scale rather than variable return to scale. 
ii. From the intra group assessment; 
a. Although some members of the FSU have appreciable levels of efficiency, 
most members of this IGO have high levels of group and meta-
inefficiencies. There is high discrepancies between the group and meta-
efficiencies of the FSU states. 
 
b. IEA states are low inefficiencies and similar levels of group and meta-
efficiency scores. Additionally, assessment of their technology gap ratios 
(TGRs) show that most IEA states have high TGRs. 
 
c. The performance of members of the OAPEC bloc is seen to be moderate 
to high for almost all the member countries. Although some member states 
have quite high results, a few other states are low. This show some level of 
dispersion in the performance of member states. 
 
118 
 
University of Ghana  http://ugspace.ug.edu.gh
d. OPEC member states produce the highest levels of oil outputs. This is 
encouraging since they also are the highest endowed IGO in terms of oil 
reserves. This notwithstanding they seem to have quite high meta and 
group inefficiency levels. However, the levels of inefficiencies for the 
group and meta scores are quite similar for most OPEC states. 
 
iii. For the inter group assessment: 
a. By comparing the averages and distributions of the groups, IEA states on 
average were the best performers followed by OAPEC, OPEC and FSU in 
that order. The average levels of meta-efficiency and technology gap ratios 
of IEA were seen to be significantly larger than the averages of the other 
IGOs 
 
b. This was also confirmed from the Global Frontier analysis since the best 
performing countries in the IEA are seen to significantly outperform even 
the best performing countries in the other three IGOs. IEA and OAPEC 
production frontiers were seen to be consistently better than both the OPEC 
and FSU frontiers for all 14 years. 
 
6.2  Conclusions of the study 
The study through its findings has identified some interesting issues that need careful 
consideration in the performance assessment and benchmarking of oil producing countries 
and IGOs related to the international oil industry. 
First, the issue of scale of operation and its possible effects on productive capabilities of 
OPCs in the industry was examined using the test of returns to scale property of the 
119 
 
University of Ghana  http://ugspace.ug.edu.gh
production frontier in the international oil and gas industry. Results show that size does not 
matter in this industry since constant returns to scale was observed for the industry for each 
of the years examined. This means that all oil producing countries in the international oil 
industry examined are operating at optimal production scales. Countries need not bother 
about either increasing their productive capacities or reducing it in order to improve supply 
efficiencies since their current production levels are of the appropriate scale. This also 
means that rate of change in reserve capacities (inputs) is consistent with changes in their 
production levels (outputs) (Asmild et al., 2013). Oil producing countries are of similar 
production capacities irrespective of the size of the country’s oil and gas resource 
endowment. This is probably because of the capital intensive nature of oil and gas 
exploration and production activities (Szilas, 1985). 
Second issue of concern pertains to the intra group performance assessments. FSU states 
were seen to be underperformers with only few lower inefficiencies and higher TGRs. On 
the other hand, IEA states were mostly seen as consistent performers since most of them 
had high TGRs. For OAPEC and OPEC states, their performance were mostly mixed. 
Whereas some member states were high performers, a few others were not that good. 
However, for both OPEC and OAPEC, number states who were generally high performers 
outweighed those who were not that good. The low performance of the FSU states could be 
partly as a result of the level of infrastructural investment in the region and partly due to the 
low level of cooperation by member states in the production efforts. It can be argued that 
the low level of capital investment and technology, linked with the lack of cooperation could 
be the cause of the disparities in the performance within the IGO. This is evident in the very 
high meta-inefficiencies and group inefficiencies exhibited by the countries in the IGO. The 
IEA’s consistent performance and a high intra-group scores mean that member countries 
are all performing very close to the frontier in a very similar manner. This is an indication 
120 
 
University of Ghana  http://ugspace.ug.edu.gh
that countries in IEA may be producing using the best state of knowledge in the industry. 
There high performance may be due to intra group collaboration of members and the vibrant 
nature of this IGO especially in seeking international expertise to develop the IEA. 
OAPEC countries’ performances are moderately high, signifying that members of the IGO 
are not too far from the frontier in terms of the group efficiency. This is evident in the small 
differences between the meta-efficiencies and group efficiencies giving an indication that 
countries in this IGO are producing using close to the best technology in the industry to a 
large extent. This notwithstanding few members are a distant from the frontier and requires 
improvement. This could be that these states may not be following the laid down policies in 
the group and that group policies and guidelines may not actively drive members to the same 
direction. OPEC states show a lot of differences in their performance, indicating that some 
countries are using similar state of knowledge in the industry while others fail to take 
advantage of the best available state of knowledge in their production efforts. Even though 
member states produce the highest levels of oil outputs. Issues of internal collective action 
challenges and the rise of new producers can explain the differences in their performance 
(Goldthau & Witte, 2011). Group policies may not be as effective as it was previously with 
the rise of new and powerful non-member players like USA and China. 
Finally, for the inter-group performance assessment, whereas IEA was seen as the best 
performer in the industry, FSU was seen to be not as improved as all the other IGOs in the 
international oil industry that were assessed. OAPEC and OPEC came second and third 
respectively on the TGRs ranking. IEA frontier was the best performer in the study over the 
14 years period from 2000 to 2013 outperforming OAPEC, OPEC and the FSU in terms of 
meta-efficiency and group efficiency. It is also evident in the TGR. The same situation was 
observed when the production frontiers of these four groups were compares. The IEA 
frontier was seen to be the best in the industry that sets the pace for all other groups. Second 
121 
 
University of Ghana  http://ugspace.ug.edu.gh
was the OAPEC frontier. Whereas the OPEC production frontier comes closely third with 
that of the OAPEC frontier, both groups’ frontier are a distant away from the current state 
of knowledge employed by IEA members. FSU had the lowest production frontier in the 
international oil industry. IEA’s performance could be as a result of both the high intra-
group collaboration and even higher inter-group (external) collaborations. The result is 
therefore high exchange of information that results in better organizational learning and 
innovation by members and addressed all information asymmetries and bottlenecks that may 
hinder the progress of the production and supply capabilities of member states. 
 
6.3 Recommendations of the study 
Based on the findings and conclusions of this study some essential recommendations can be 
proffered on policy, practice and future research. This will assist oil producing countries, 
policy makers in the industry, investors, oil related IGOs in the international oil industry 
and academics to better appreciate the dynamics of the international oil industry. 
Recommendations on Policy 
a. Due to the fact that constant returns to scale was observed, OPCs are seen to be 
producing at optimal production sizes. This means that policy makers should not 
concentrate on developing policies that bother on capacity expansion or reduction 
as current levels are adequate.  
b. FSU states should put in more efforts to formalise their association with clearer 
policy guidelines that enshrine better collaboration among member states.  
c. IEA was seen to be the best IGO. Whiles this is good, the IGO should strive for more 
collaborations especially with the other IGOs in this industry. Other IGOs like 
122 
 
University of Ghana  http://ugspace.ug.edu.gh
OAPEC and OPEC can use IEA as a benchmark when developing operational 
targets. 
d. Countries that are yet to join an IGO can take a close look at the activities of IEA in 
ensuring both internal and external collaborations. Choice of an IGO to join should 
be guided by the level of collaboration among member states and the technical 
capabilities of external collaborators. 
Recommendations on Practice 
a. Management of the production activities in these countries should ensure that current 
optimal levels are maintained. This is because, whereas these countries may not be 
facing scale inefficiencies now, it is very possible that future production capacities 
may result in higher costs due to large operation. 
b. FSU states should be more open to international collaboration instead on focusing 
on internal capabilities as was observed in literature. Close association with higher 
performing states and organizations can better streamline their activities and reduce 
their inefficiencies. 
c. Whereas OAPEC and OPEC were among the best performers, individual members 
experienced different levels of performance. Organizational policies and guidelines 
should be better institutionalised. Efforts should be directed towards ensuring that 
member countries adhere to these policies that work. 
Recommendations on Further Research 
a. To develop greater understanding, further research in productive efficiencies or any 
frontier performance criteria in the industry should adopt a constant returns to scale 
model especially when assessing the performance of these OPCs under 
consideration. 
123 
 
University of Ghana  http://ugspace.ug.edu.gh
b. While Social network theory believes that common backgrounds among actors 
ensure close collaboration, for FSU states, this common ancestry does not seem to 
improve their performance. Future research should delve more into precise ties that 
can improve efficiencies in the international oil industry and how countries with 
adversarial histories can be joined towards a common productive goal. 
c. Further research is desirable to develop greater insight of how production quotas and 
other regulatory guidelines of some IGOs, such as OPEC, affects the performance 
of individual countries. 
d. OPCs were examined at the composite country level. It would be interesting for 
further research to shed more insight on how domestic dynamics and 
macroeconomic conditions translate to performance. 
e. Further research can explore how various IGOs in the industry handle price 
volatilities that affect the performances of different IGOs.   
 
 
 
 
 
 
  
124 
 
University of Ghana  http://ugspace.ug.edu.gh
REFERENCES 
Abdalla, K. L. (1995). The changing structure of the international oil industry. Energy 
Policy, 23(10), 871-877. doi: http://dx.doi.org/10.1016/0301-4215(95)98710-A 
Abu-Alkheil, A. M., Burghof, H.-P., & Khan, W. A. (2012). Islamic Commercial Banking 
In Europe: A Cross-Country And Inter-Bank Analysis Of Efficiency Performance. 
International Business & Economics Research Journal (IBER), 11(6), 647-676.  
Adetutu, M. O. (2014). Energy efficiency and capital-energy substitutability: Evidence from 
four OPEC countries. Applied Energy, 119, 363-370. doi: 
http://dx.doi.org/10.1016/j.apenergy.2014.01.015 
Adler-Milstein, J., Ronchi, E., Cohen, G. R., Winn, L. A. P., & Jha, A. K. (2014). 
Benchmarking health IT among OECD countries: better data for better policy. 
Journal of the American Medical Informatics Association, 21(1), 111-116.  
Afonso, A., & St Aubyn, M. (2005). Non-parametric approaches to education and health 
efficiency in OECD countries. Journal of Applied Economics, 8(2), 227-246.  
Afonso, A., & St. Aubyn, M. (2006). Cross-country efficiency of secondary education 
provision: A semi-parametric analysis with non-discretionary inputs. Economic 
Modelling, 23(3), 476-491. doi: http://dx.doi.org/10.1016/j.econmod.2006.02.003 
Aguilera, R. F. (2012). The economics of oil and gas supply in the Former Soviet Union. 
International Journal of Global Energy Issues, 35(6), 480-493.  
Ahmed, E. M., & Krishnasamy, G. (2013). Are Asian technology gaps due to human capital 
quality differences? Economic Modelling, 35, 51-58.  
Aiello, J. R., & Douthitt, E. A. (2001). Social facilitation from Triplett to electronic 
performance monitoring. Group Dynamics: Theory, Research, and Practice, 5(3), 
163-180.  
125 
 
University of Ghana  http://ugspace.ug.edu.gh
Aigner, D., Lovell, C. A. K., & Schmidt, P. (1977). Formulation and estimation of stochastic 
frontier production function models. Journal of Econometrics, 6(1), 21-37.  
Al-Essa, R. K., Al-Rubaie, M., Walker, S., & Salek, S. (2015). The Pharmaceutical 
Companies Assessment and Experience with the Centralised Procedure 
Pharmaceutical Regulatory Environment (pp. 171-188): Springer. 
Al-Obaidan, A. M., & Scully, G. W. (1995). The theory and measurement of the net benefits 
of multinationality: the case of the international petroleum industry. Applied 
Economics, 27(2), 231-238. doi: 10.1080/00036849500000029 
Al-Rashed, Y., & León, J. (2015). Energy efficiency in OPEC member countries: analysis 
of historical trends through the energy coefficient approach. OPEC Energy Review, 
39(1), 77-102. doi: 10.1111/opec.12041 
Al‐Rashed, Y., & León, J. (2015). Energy efficiency in OPEC member countries: analysis 
of historical trends through the energy coefficient approach. OPEC Energy Review, 
39(1), 77-102.  
Alcacer, J., & Ingram, P. (2013). Spanning the Institutional Abyss: The Intergovernmental 
Network and the Governance of Foreign Direct Investment1. American journal of 
sociology, 118(4), 1055-1098.  
Allport, F. H. (1920). The influence of the group upon association and thought. Journal of 
experimental psychology, 3(3), 159.  
Alston, L., Eggerston, T., & North, D. (1996). Empirical Studies of Organizational Change: 
Cambridge: Cambridge University Press. 
Amit, R., & Schoemaker, P. J. (2012). Z STRATEGIC ASSETS AND 
ORGANIZATIONAL RENT. Strategische Managementtheorie, 14, 325.  
126 
 
University of Ghana  http://ugspace.ug.edu.gh
Arena, M. (2008). Bank failures and bank fundamentals: A comparative analysis of Latin 
America and East Asia during the nineties using bank-level data. Journal of Banking 
& Finance, 32(2), 299-310. doi: DOI: 10.1016/j.jbankfin.2007.03.011 
Arestis, P., Chortareas, G., & Desli, E. (2006). FINANCIAL DEVELOPMENT AND 
PRODUCTIVE EFFICIENCY IN OECD COUNTRIES: AN EXPLORATORY 
ANALYSIS*. The Manchester School, 74(4), 417-440.  
Aristovnik, A. (2012). The impact of ICT on educational performance and its efficiency in 
selected EU and OECD countries: a non-parametric analysis. Available at SSRN 
2187482.  
Armaroli, N., & Balzani, V. (2007). The future of energy supply: challenges and 
opportunities. Angewandte Chemie International Edition, 46(1‐2), 52-66.  
Arnade, C. A. (1994). Using Data Envelopment Analysis To IVIeasure International 
Agricultural Efficiency and Productivity.  
Asif, M., & Muneer, T. (2007). Energy supply, its demand and security issues for developed 
and emerging economies. Renewable and Sustainable Energy Reviews, 11(7), 1388-
1413.  
Asmild, M., Hollingsworth, B., & Birch, S. (2013). The scale of hospital production in 
different settings: one size does not fit all. Journal of Productivity Analysis, 40(2), 
197-206. doi: 10.1007/s11123-012-0332-9 
Asmild, M., Paradi, J. C., Reese, D. N., & Tam, F. (2007). Measuring overall efficiency and 
effectiveness using DEA. European Journal of Operational Research, 178(1), 305-
321.  
Asmild, M., & Tam, F. (2007). Estimating global frontier shifts and global Malmquist 
indices. Journal of Productivity Analysis, 27(2), 137-148.  
127 
 
University of Ghana  http://ugspace.ug.edu.gh
Assaf, A., Barros, C. P., & Josiassen, A. (2010). Hotel efficiency: A bootstrapped 
metafrontier approach. [doi: 10.1016/j.ijhm.2009.10.020]. International Journal of 
Hospitality Management, 29(3), 468-475.  
Atkinson, S. E., & Cornwell, C. (1994). Estimation of Output and Input Technical 
Efficiency using a Flexible Functional Form and Panel Data. International 
Economic Review, 35(1), 245-255. doi: 10.2307/2527100 
Badunenko, O. (2008). Downsizing in the German chemical manufacturing industry during 
the 1990s. Why is small beautiful? [journal article]. Small Business Economics, 
34(4), 413-431. doi: 10.1007/s11187-008-9142-x 
Bamberger, C. S., Scott, R., Agency, I. E., & Development, O. E. C. (2004). IEA : the First 
30 Years: Organisation for Economic Co-operation and Development. 
Banker, R. D. (1984). Estimating Most Productive Scale Size Using Data Envelopment 
Analysis. European Journal of Operational Research, 17(1), 35-44.  
Banker, R. D. (1993). Maximum Likelihood, Consistency and Data Envelopment Analysis: 
A Statistical Foundation. Management Science, 39(10), 1265-1273. doi: 
10.1287/mnsc.39.10.1265 
Banker, R. D. (1996). Hypothesis tests using data envelopment analysis. Journal of 
Productivity Analysis, 7(2), 139-159.  
Banker, R. D., Charnes, A., & Cooper, W. W. (1984). Some Models for Estimating 
Technical and Scale Inefficiencies in Data Envelopment Analysis. Management 
Science, 30(9), 1078-1092.  
Banker, R. D., Cooper, W. W., Seiford, L. M., Thrall, R. M., & Zhu, J. (2004). Returns to 
scale in different DEA models. European Journal of Operational Research, 154(2), 
345-362. doi: Doi: 10.1016/s0377-2217(03)00174-7 
128 
 
University of Ghana  http://ugspace.ug.edu.gh
Barnes, A. P., & Revoredo-Giha, C. (2011). A metafrontier analysis of technical efficiency 
of selected European agricultures. Paper presented at the Paper provided by 
European Association of Agricultural Economists in its series 2011 International 
Congress. 
Barnes, J. A. (1954). Class and committees in a Norwegian island parish: Plenum. 
Barnett, M. N., & Finnemore, M. (1999). The politics, power, and pathologies of 
international organizations. International Organization, 53(04), 699-732.  
Barney, J. (1991). Firm Resources and Sustained Competitive Advantage. Journal of 
Management, 17(1), 99-120. doi: 10.1177/014920639101700108 
Barney, J., Wright, M., & Ketchen Jr, D. J. (2001). The resource-based view of the firm: 
Ten years after 1991. Journal of Management, 27(6), 625-641. doi: 
http://dx.doi.org/10.1016/S0149-2063(01)00114-3 
Barney, J. B. (1986). Strategic factor markets: Expectations, luck, and business strategy. 
Management science, 32(10), 1231-1241.  
Barney, J. B. (1995). Looking inside for competitive advantage. The Academy of 
Management Executive, 9(4), 49-61.  
Barney, J. B. (2001). Resource-based theories of competitive advantage: A ten-year 
retrospective on the resource-based view. Journal of Management, 27(6), 643-650.  
Barney, J. B., Ketchen, D. J., & Wright, M. (2011). The future of resource-based theory 
revitalization or decline? Journal of Management, 37(5), 1299-1315.  
Barros, C., & Antunes, O. (2014). Productivity change in the oil blocks of Angola. Energy 
Sources, Part B: Economics, Planning, and Policy, 9(4), 413-424.  
Barros, C. P., & Assaf, A. (2009). Bootstrapped efficiency measures of oil blocks in Angola. 
Energy Policy, 37(10), 4098-4103. doi: 
http://dx.doi.org/10.1016/j.enpol.2009.05.007 
129 
 
University of Ghana  http://ugspace.ug.edu.gh
Barros, C. P., & Managi, S. (2009a). Productivity assessment of Angola's oil blocks. Energy, 
34(11), 2009-2015. doi: http://dx.doi.org/10.1016/j.energy.2009.08.016 
Barros, C. P., & Managi, S. (2009b). Regulation, pollution and heterogeneity in Japanese 
steam power generation companies. Energy Policy, 37(8), 3109-3114. doi: DOI: 
10.1016/j.enpol.2009.04.003 
Battese, G. E., & Rao, D. S. P. (2002a). Technology gap, efficiency, and a stochastic 
metafrontier function. International Journal of Business and Economics, 1(2), 87-
93.  
Battese, G. E., Rao, D. S. P., & O'Donnell, C. J. (2004). A Metafrontier Production Function 
for Estimation of Technical Efficiencies and Technology Gaps for Firms Operating 
Under Different Technologies. Journal of Productivity Analysis, 21(1), 91-103. doi: 
10.1023/b:prod.0000012454.06094.29 
Battese, G. E., & Rao, P. D. S. (2002b). Technology Gap, Efficiency, and a Stochastic 
Metafrontier Function. International Journal of Business and Economics, 1(2), 87-
93.  
Behname, M. (2012). The Compare of concentration and Efficiency in Banking industry: 
An Evidence from the OPEC Countries. Eurasian Journal of Business and 
Economics, 5.  
Belyi, A. V., & Talus, K. (2015). States and markets in hydrocarbon sectors, from 
http://www.palgraveconnect.com/doifinder/10.1057/9781137434074 
Bennett, A. L., & Oliver, J. K. (2002). International Organizations: Principles and Issues 
(7th ed.). Upper Saddle River: Prentice Hall. 
Berger, P. L., & Luckmann, T. (1991). The social construction of reality: A treatise in the 
sociology of knowledge: Penguin UK. 
130 
 
University of Ghana  http://ugspace.ug.edu.gh
Bharadwaj, A. S. (2000). A resource-based perspective on information technology 
capability and firm performance: an empirical investigation. MIS quarterly, 169-
196.  
Bhattacherjee, A. (2012). Social science research: principles, methods, and practices.  
Biermann, F., & Bauer, S. (2004). Assessing the effectiveness of intergovernmental 
organisations in international environmental politics. Global Environmental 
Change, 14(2), 189-193.  
Blascovich, J., Mendes, W. B., Hunter, S. B., & Salomon, K. (1999). Social" facilitation" 
as challenge and threat. Journal of personality and social psychology, 77(1), 68.  
Bosseboeuf, D., Chateau, B., & Lapillonne, B. (1997). Cross-country comparison on energy 
efficiency indicators: the on-going European effort towards a common 
methodology. Energy Policy, 25(7), 673-682.  
Bott, E. (1957). Family and social network: Roles, norms, and external relationships in 
ordinary urban families: Tavistock Publications. 
BP. (2014). ANNUAL REPORT AND FORM 20-F  
Bradley, K., & Ramirez, F. O. (1996). World polity and gender parity: Women’s share of 
higher education, 1965–1985. Research in sociology of education and socialization, 
11(1), 63-91.  
Çakır, S., Perçin, S., Min, H., & Gunasekaran, A. (2015). Evaluating the comparative 
efficiency of the postal services in OECD countries using context-dependent and 
measure-specific data envelopment analysis. Benchmarking: An International 
Journal, 22(5).  
Camanho, A. S., & Dyson, R. G. (2005). Cost efficiency, production and value-added 
models in the analysis of bank branch performance. Journal of the Operational 
Research Society, 56(2), 483–494. doi: 10.1057/palgrave.jors.2601839 
131 
 
University of Ghana  http://ugspace.ug.edu.gh
Camisón, C., & Villar-López, A. (2014). Organizational innovation as an enabler of 
technological innovation capabilities and firm performance. Journal of Business 
Research, 67(1), 2891-2902.  
Canhoto, A., & Dermine, J. (2003). A note on banking efficiency in Portugal, New vs. Old 
banks. Journal of Banking & Finance, 27(11), 2087-2098. doi: Doi: 10.1016/s0378-
4266(02)00316-3 
Cao, X. (2009). Networks of Intergovernmental Organizations and Convergence in 
Domestic Economic Policies. International Studies Quarterly, 53(4), 1095-1130. 
doi: 10.1111/j.1468-2478.2009.00570.x 
Casu, B., & Girardone, C. (2004). Large banks' efficiency in the single European market. 
The Service Industries Journal, 24(6), 129 - 142.  
Casu, B., & Girardone, C. (2006). Bank competition, concentration and efficiency in the 
single European market*. The Manchester School, 74(4), 441-468. doi: 
10.1111/j.1467-9957.2006.00503.x 
Casu, B., & Molyneux, P. (2003). A comparative study of efficiency in European banking. 
Applied Economics, 35, 1865-1876. doi: 10.1080/0003684032000158109 
Charnes, A., Cooper, W. W., & Rhodes, E. (1978a). Measuring Efficiency of Decision-
Making Units. European Journal of Operational Research, 2(6), 429-444.  
Charnes, A., Cooper, W. W., & Rhodes, E. (1978b). Measuring the Efficiency of Decision-
Making Units. [10.1016/0377-2217(78)90138-8]. European Journal of Operations 
Research, 2, 429-444.  
Chase, J. (2012). Operations management: Tata McGraw-Hill. 
CIA. (2014). World Energy Outlook International Energy Agency. 
132 
 
University of Ghana  http://ugspace.ug.edu.gh
Cissokho, L., Haughton, J., Makpayo, K., & Seck, A. (2013). Why Is Agricultural Trade 
within ECOWAS So High? Journal of African Economies, 22(1), 22-51. doi: 
10.1093/jae/ejs015 
Claeys, S., & Vander Vennet, R. (2008). Determinants of bank interest margins in Central 
and Eastern Europe: A comparison with the West. Economic Systems, 32(2), 197-
216. doi: DOI: 10.1016/j.ecosys.2007.04.001 
Cleveland, C. J., Costanza, R., Hall, C. A., & Kaufmann, R. (1997). Energy and the US 
Economy: A Biophysical Perspective. INTERNATIONAL LIBRARY OF CRITICAL 
WRITINGS IN ECONOMICS, 75, 295-302.  
Colgan, J., Keohane, R., & Van de Graaf, T. (2011). Institutional Change in the Energy 
Regime Complex. Paper presented at the Political Economy of International 
Organizations Meeting, Zurich. 
Colgan, J. D., Keohane, R. O., & Van de Graaf, T. (2012). Punctuated equilibrium in the 
energy regime complex. The Review of International Organizations, 7(2), 117-143.  
Conner, K. R. (1991). A historical comparison of resource-based theory and five schools of 
thought within industrial organization economics: do we have a new theory of the 
firm? Journal of Management, 17(1), 121-154.  
Conner, K. R., & Prahalad, C. K. (1996). A resource-based theory of the firm: Knowledge 
versus opportunism. Organization Science, 7(5), 477-501.  
Cooper, W. W., Seiford, L. M., & Zhu, J. (2011). Data Envelopment Analysis: History, 
Models, and Interpretations, Handbook on Data Envelopment Analysis. In W. W. 
Cooper, L. M. Seiford & J. Zhu (Eds.), (Vol. 164, pp. 1-39): Springer US. 
Cottrell, N. B. (1968). Performance in the presence of other human beings: Mere presence, 
audience, and affiliation effects. Social facilitation and imitative behavior. Boston: 
Allyn & Bacon, 91-110.  
133 
 
University of Ghana  http://ugspace.ug.edu.gh
Crawford, M. P. (1939). The social psychology of the invertebrates. Psychological Bulletin, 
36, 407-446.  
Creswell, J. W. (2013). Research design: Qualitative, quantitative, and mixed methods 
approaches: Sage publications. 
Crook, T. R., Ketchen, D. J., Combs, J. G., & Todd, S. Y. (2008). Strategic resources and 
performance: a meta‐analysis. Strategic management journal, 29(11), 1141-1154.  
Cross, R. L., & Parker, A. (2004). The hidden power of social networks: Understanding 
how work really gets done in organizations: Harvard Business Press. 
Daly, A. J. (2012). Data, dyads, and dynamics: Exploring data use and social networks in 
educational improvement. Teachers College Record, 114(11), 1-38.  
De Witte, K., & Marques, R. C. (2009). Capturing the environment, a metafrontier approach 
to the drinking water sector. International Transactions in Operational Research, 
16(2), 257-271. doi: 10.1111/j.1475-3995.2009.00675.x 
Desta, M. G. (2003). The Organization of Petroleum Exporting Countries, the World Trade 
Organization, and Regional Trade Agreements. Journal of World Trade, 37(3), 523-
551.  
Dietsch, M., & Lozano-Vivas, A. (2000). How the environment determines banking 
efficiency: A comparison between French and Spanish industries. Journal of 
Banking & Finance, 24(6), 985-1004. doi: Doi: 10.1016/s0378-4266(99)00115-6 
Dike, J. C. (2013). Measuring the security of energy exports demand in OPEC economies. 
Energy Policy, 60, 594-600. doi: http://dx.doi.org/10.1016/j.enpol.2013.05.086 
DiMaggio, P., & Powell, W. W. (1983). The iron cage revisited: Institutional isomorphism 
and collective rationality in organizational fields. American Sociological Review, 48, 
147-160.  
134 
 
University of Ghana  http://ugspace.ug.edu.gh
Dobbin, F., & Baum, J. A. C. (2000). Introduction: Economics meets sociology in strategic 
management Economics Meets Sociology in Strategic Management (pp. 1-26). 
Donni, O., & Fecher, F. (1997). Efficiency and productivity of the insurance industry in the 
OECD countries. The Geneva Papers on Risk and Insurance, 22, 523-535.  
Dorussen, H., & Ward, H. (2008). Intergovernmental Organizations and the Kantian Peace: 
A Network Perspective. Journal of Conflict Resolution, 52(2), 189-212. doi: 
10.1177/0022002707313688 
Drakos, K. (2003). Assessing the success of reform in transition banking 10 years later: an 
interest margins analysis. Journal of Policy Modeling, 25(3), 309-317.  
Duffield, J. S. (2012). The Return of Energy Insecurity in the Developed Democracies. 
Contemporary Security Policy, 33(1), 1-26.  
Dyer, J. H., & Singh, H. (1998). The relational view: Cooperative strategy and sources of 
interorganizational competitive advantage. Academy of management review, 23(4), 
660-679.  
Dyson, R. G., Allen, R., Camanho, A. S., Podinovski, V. V., Sarrico, C. S., & Shale, E. A. 
(2001). Pitfalls and protocols in DEA. European Journal of Operational Research, 
132(2), 245-259. doi: Doi: 10.1016/s0377-2217(00)00149-1 
Easterly, W., & Fischer, S. (1994). What we can learn from the Soviet collapse. Finance 
and Development, 31(4), 2.  
Efron, B. (1979). Bootstrap Methods: Another Look at the Jackknife. The Annals of 
Statistics, 7(1), 1-26.  
EIA. (2013). International Energy Outlook  
EIA. (2014). Country Analysis Brief: Saudi Arabia: U.S Energy Information 
Administration. 
EIA. (2015). International Energy Outlook  
135 
 
University of Ghana  http://ugspace.ug.edu.gh
Eleftheriadou, E. M., Y. (2008). Game Theoretical Approach to Conflict Resolution in 
Transboundary Water Resources Management. Journal of Water Resources 
Planning and Management, 134(5), 466-473. doi: doi:10.1061/(ASCE)0733-
9496(2008)134:5(466) 
Eller, S., Hartley, P., & Medlock, K., III. (2011). Empirical evidence on the operational 
efficiency of National Oil Companies. Empirical Economics, 40(3), 623-643. doi: 
10.1007/s00181-010-0349-8 
Embirbayer, M., & Goodwin, J. (1994). Network analysis, culture, and the problem of 
agency. Administrative Science Quarterly, 99, 1411-1454.  
Emrouznejad, A., & De Witte, K. (2010). COOPER-framework: A unified process for non-
parametric projects. European Journal of Operational Research, 207(3), 1573-1586. 
doi: DOI: 10.1016/j.ejor.2010.07.025 
Epure, M., Kerstens, K., & Prior, D. (2011). Technology-based total factor productivity and 
benchmarking: New proposals and an application. Omega, 39(6), 608-619. doi: 
http://dx.doi.org/10.1016/j.omega.2011.01.001 
Escobar, O. R., & Le Chaffotec, A. (2015). The influence of OPEC membership on 
economic development: A transaction cost comparative approach. Research in 
International Business and Finance, 33, 304-318. doi: 
http://dx.doi.org/10.1016/j.ribaf.2014.04.005 
Färe, R., & Grosskopf, S. (1985). A nonparametric cost approach to scale efficiency. The 
Scandinavian Journal of Economics, 594-604.  
Fare, R., Grosskopf, S., Norris, M., & Zhang, Z. (1994). Productivity growth, technical 
progress, and efficiency change in industrialized countries. American Economic 
Review, 84(1), 66-83.  
136 
 
University of Ghana  http://ugspace.ug.edu.gh
Filippini, M., & Hunt, L. C. (2011). Energy demand and energy efficiency in the OECD 
countries: a stochastic demand frontier approach. Energy Journal, 32(2), 59-80.  
Finnemore, M. (1993). International organizations as teachers of norms: the United Nations 
Educational, Scientific, and Cutural Organization and science policy. International 
Organization, 47(04), 565-597.  
Foss, N. J. (1996). Knowledge-based approaches to the theory of the firm: Some critical 
comments. Organization Science, 7(5), 470-476.  
Francisco, C. A. C., de Almeida, M. R., & da Silva, D. R. (2012). Efficiency in Brazilian 
refineries under different DEA technologies. International Journal of Engineer‐ing 
Business Management, 4, 1-11.  
Frank, D. J. (1997). Science, Nature, and the Globalization of the Environment, 1870–1990. 
Social Forces, 76(2), 409-435.  
Frank, D. J. (1999). The social bases of environmental treaty ratification, 1900–1990. 
Sociological Inquiry, 69(4), 523-550.  
Frank, D. J., Hironaka, A., & Schofer, E. (2000). The nation-state and the natural 
environment over the twentieth century. American Sociological Review, 96-116.  
Frisvold, G. B., & Caswell, M. F. (2000). Transboundary water management Game-
theoretic lessons for projects on the US–Mexico border⋆. Agricultural Economics, 
24(1), 101-111. doi: 10.1111/j.1574-0862.2000.tb00096.x 
Garton, L., Haythornthwaite, C., & Wellman, B. (1997). Studying Online Social Networks. 
Journal of Computer-Mediated Communication, 3(1), 0-0. doi: 10.1111/j.1083-
6101.1997.tb00062.x 
Gaytaranov, J. (2013). Export performance of transition countries: uga. 
137 
 
University of Ghana  http://ugspace.ug.edu.gh
Ghedrovici, O., & Ostapenko, N. (2013). The Glaring Socioeconomic Meltdown in Post-
Soviet Ukraine, Moldova, and Belarus: A Distorted Mindset in Search of a Way Out. 
International Journal of Business and Social Research, 3(5), 202-211.  
Ghemawat, P. (1991). Commitment: Simon and Schuster. 
Gitto, S., & Mancuso, P. (2012). Bootstrapping the Malmquist indexes for Italian airports. 
International Journal of Production Economics, 135(1), 403-411. doi: 
http://dx.doi.org/10.1016/j.ijpe.2011.08.014 
Goldemberg, J. (2000). World Energy Assessment: Energy and the challenge of 
sustainability: United Nations Pubns. 
Goldthau, A., & Witte, J. M. (2011). Assessing OPEC’s performance in global energy. 
Global Policy, 2(s1), 31-39.  
Gómez-Calvet, R., Conesa, D., Gómez-Calvet, A. R., & Tortosa-Ausina, E. (2014). Energy 
efficiency in the European Union: What can be learned from the joint application of 
directional distance functions and slacks-based measures? Applied Energy, 132(0), 
137-154. doi: http://dx.doi.org/10.1016/j.apenergy.2014.06.053 
Gomez, C., & Parigi, P. (2013). The Embedded Duality of Structure and Harmonic Change 
in Intergovernmental Organization Networks. Paper presented at the APSA 2013 
Annual Meeting Paper. 
Gorton, M., & Davidova, S. (2004). Farm productivity and efficiency in the CEE applicant 
countries: a synthesis of results. Agricultural Economics, 30(1), 1-16. doi: 
http://dx.doi.org/10.1016/j.agecon.2002.09.002 
Granovetter, M. S. (1973). The strength of weak ties. American journal of sociology, 1360-
1380.  
Greenwood, R., Sage, P., & Sage, e. (2008). The SAGE handbook of organizational 
institutionalism, from http://site.ebrary.com/id/10501920 
138 
 
University of Ghana  http://ugspace.ug.edu.gh
Hawdon, D. (2003). Efficiency, performance and regulation of the international gas 
industry—a bootstrap DEA approach. Energy Policy, 31(11), 1167-1178. doi: 
http://dx.doi.org/10.1016/S0301-4215(02)00218-5 
Hayami, Y., & Ruttan, V. W. (1971). Agricultural development: an international 
perspective: Baltimore, Md/London: The Johns Hopkins Press. 
Hayward, R., & Cutler, P. (2006). Stakeholder involvement: a challenge for 
intergovernmental organisations. Journal of Public Mental Health, 5(1), 14-17.  
Henderson, R., & Cockburn, I. (1994). Measuring competence? Exploring firm effects in 
pharmaceutical research. Strategic management journal, 15, 63-63.  
Hoff, A. (2006). Bootstrapping Malmquist Indices for Danish Seiners in the North Sea and 
Skagerrak. Journal of Applied Statistics, 33(9), 891-907. doi: 
10.1080/02664760600742151 
Holdren, J. P. (2006). The energy innovation imperative: addressing oil dependence, climate 
change, and other 21st century energy challenges. innovations, 1(2), 3-23.  
Holland, G. L. (1998). The role of intergovernmental organizations in coastal zone 
management. Ocean & Coastal Management, 39(1–2), 25-31. doi: 
http://dx.doi.org/10.1016/S0964-5691(98)00010-6 
Honma, S., & Hu, J.-L. (2008). Total-factor energy efficiency of regions in Japan. Energy 
Policy, 36(2), 821-833. doi: http://dx.doi.org/10.1016/j.enpol.2007.10.026 
Hoopes, D. G., Madsen, T. L., & Walker, G. (2003). Guest editors' introduction to the 
special issue: why is there a resource‐based view? Toward a theory of competitive 
heterogeneity. Strategic management journal, 24(10), 889-902.  
Hori, S. (2012). Implications of energy efficiency and economic growth in developing 
countries. Journal of novel carbon resource sciences, 6, 9-14.  
139 
 
University of Ghana  http://ugspace.ug.edu.gh
Huang, Z., & Li, S. X. (2001). Co-op advertising models in manufacturer–retailer supply 
chains: A game theory approach. European Journal of Operational Research, 
135(3), 527-544. doi: http://dx.doi.org/10.1016/S0377-2217(00)00327-1 
Hunt, P. J., & Hillery, J. M. (1973). Social facilitation in a coaction setting: An examination 
of the effects over learning trials. Journal of Experimental Social Psychology, 9(6), 
563-571. doi: http://dx.doi.org/10.1016/0022-1031(73)90038-3 
IEA. (2014). World Energy Outlook International Energy Agency. 
Igos, E., Rugani, B., Rege, S., Benetto, E., Drouet, L., Zachary, D., & Hass, T. (2015). 
Integrated environmental assessment of future energy scenarios based on economic 
equilibrium models.  
Ike, C. B., & Lee, H. (2014). Measurement of the efficiency and productivity of national oil 
companies and its determinants. Geosystem Engineering, 17(1), 1-10. doi: 
10.1080/12269328.2014.887045 
Ingram, P., Robinson, J., & Busch, M. L. (2005). The Intergovernmental Network of World 
Trade: IGO Connectedness, Governance, and Embeddedness1. American journal of 
sociology, 111(3), 824-858.  
Ismail, Z., Tai, J. C., Kong, K. K., Law, K. H., Shirazi, S. M., & Karim, R. (2013). Using 
data envelopment analysis in comparing the environmental performance and 
technical efficiency of selected companies in their global petroleum operations. 
Measurement, 46(9), 3401-3413. doi: 
http://dx.doi.org/10.1016/j.measurement.2013.04.076 
Jackson, M. O. (2010). An overview of social networks and economic applications. The 
handbook of social economics, 1, 511-585.  
140 
 
University of Ghana  http://ugspace.ug.edu.gh
Jacobs, R. (2001). Alternative methods to examine hospital efficiency: data envelopment 
analysis and stochastic frontier analysis. Health Care Management Science, 4(2), 
103-115.  
Jollands, N., Waide, P., Ellis, M., Onoda, T., Laustsen, J., Tanaka, K., . . . Meier, A. (2010). 
The 25 IEA energy efficiency policy recommendations to the G8 Gleneagles Plan 
of Action. Energy Policy, 38(11), 6409-6418.  
Jones, R. B. (1995). Globalisation and interdependence in the international political 
economy: rhetoric and reality: Pinter Pub Ltd. 
Kadushin, C. (2004). Too much investment in social capital? Social Networks, 26(1), 75-
90.  
Kalb, A. (2010). The Impact of Intergovernmental Grants on Cost Efficiency: Theory and 
Evidence from German Municipalities. Economic Analysis and Policy, 40(1), 23-
48. doi: http://dx.doi.org/10.1016/S0313-5926(10)50002-X 
Kashani, H. A. (2005a). Regulation and efficiency: an empirical analysis of the United 
Kingdom continental shelf petroleum industry. Energy Policy, 33(7), 915-925. doi: 
http://dx.doi.org/10.1016/j.enpol.2003.10.014 
Kashani, H. A. (2005b). State intervention causing inefficiency: an empirical analysis of the 
Norwegian Continental Shelf. Energy Policy, 33(15), 1998-2009. doi: 
http://dx.doi.org/10.1016/j.enpol.2004.03.020 
Katz, N., Lazer, D., Arrow, H., & Contractor, N. (2004). Network Theory and Small Groups. 
Small Group Research, 35(3), 307-332. doi: 10.1177/1046496404264941 
Kelley, H. H., & Thibaut, J. W. (1954). Experimental studies of group problem solving and 
process. Handbook of social psychology, 2, 735-785.  
Kenjegalieva, K., Simper, R., Weyman-Jones, T., & Zelenyuk, V. (2009). Comparative 
analysis of banking production frameworks in eastern european financial markets. 
141 
 
University of Ghana  http://ugspace.ug.edu.gh
European Journal of Operational Research, 198(1), 326-340. doi: DOI: 
10.1016/j.ejor.2008.09.002 
Khalil, E. L. (1995). Organizations Versus Institutions. Journal of Institutional and 
Theoretical Economics (JITE) / Zeitschrift für die gesamte Staatswissenschaft, 
151(3), 445-466. doi: 10.2307/40751821 
Kilduff, M., & Brass, D. J. (2010). Organizational social network research: Core ideas and 
key debates. The Academy of Management Annals, 4(1), 317-357.  
Kim, T.-Y., Lee, J.-D., Park, Y. H., & Kim, B. (1999). International comparisons of 
productivity and its determinants in the natural gas industry. Energy Economics, 
21(3), 273-293. doi: http://dx.doi.org/10.1016/S0140-9883(99)00007-9 
Kneip, A., Simar, L., & Wilson, P. W. (2008). Asymptotics and Consistent Bootstraps for 
DEA Estimators in Non-parametric Frontier Models. Econometric Theory, 24(06), 
1663-1697. doi: doi:10.1017/S0266466608080651 
Košak, M., Zajc, P., & Zorić, J. (2009). Bank efficiency differences in the new EU member 
states. Baltic Journal of Economics, 9(2), 67-89.  
Kounetas, K., Mourtos, I., & Tsekouras, K. (2009). Efficiency decompositions for 
heterogeneous technologies. European Journal of Operational Research, 199(1), 
209-218. doi: DOI: 10.1016/j.ejor.2008.11.015 
Krishnasamy, G., & Ahmed, E. M. (2009). Productivity Growth Analysis in OECD 
Countries. Korea and the World Economy, 10(2), 225-244.  
Lauber, T. B., Decker, D., & Knuth, B. (2008). Social Networks and Community-Based 
Natural Resource Management. Environmental Management, 42(4), 677-687. doi: 
10.1007/s00267-008-9181-8 
142 
 
University of Ghana  http://ugspace.ug.edu.gh
Lee, C. H., Kim, J. S., & Park, K. H. (1996). Automatic human face location in a complex 
background using motion and color information. Pattern recognition, 29(11), 1877-
1889.  
Li, Q. (1996). Nonparametric testing of closeness between two unknown distribution 
functions. Econometric Reviews, 15(3), 261-274. doi: 10.1080/07474939608800355 
Lincoln, J. R. (1995). The New Institutionalism in Organizational Analysis. Edited by 
Walter W. Powell and Paul J. DiMaggio. University of Chicago Press, 1991. 478 pp. 
Cloth, $65.00; paper, $24.95. Social Forces, 73(3), 1147-1148. doi: 
10.1093/sf/73.3.1147 
Lothgren, M., & Tambour, M. (1999). Bootstrapping the data envelopment analysis 
Malmquist productivity index. Applied Economics, 31(4), 417-425. doi: 
10.1080/000368499324129 
Lovejoy, K., & Handy, S. (2011). Social networks as a source of private-vehicle 
transportation: The practice of getting rides and borrowing vehicles among Mexican 
immigrants in California. Transportation Research Part A: Policy and Practice, 
45(4), 248-257.  
Lovejoy, K., Sciara, G.-C., Salon, D., Handy, S. L., & Mokhtarian, P. (2013). Measuring 
the impacts of local land-use policies on vehicle miles of travel: The case of the first 
big-box store in Davis, California. Journal of Transport and Land Use, 6(1), 25-39.  
Lucas, K., & Mayne, R. (2013). Social network theory and analysis. Oxford: Oxford 
Brookes University. 
Madhani, P. M. (2010). Salesforce compensation: Game theory. SCMS Journal of Indian 
Management, 7(4), 72-82.  
143 
 
University of Ghana  http://ugspace.ug.edu.gh
Mahlberg, B., & Url, T. (2010). Single Market effects on productivity in the German 
insurance industry. Journal of Banking & Finance, 34(7), 1540-1548. doi: 
http://dx.doi.org/10.1016/j.jbankfin.2009.09.005 
Mamatzakis, E., Staikouras, C., & Koutsomanoli-Filippaki, A. (2008). Bank efficiency in 
the new European Union member states: Is there convergence? [doi: DOI: 
10.1016/j.irfa.2007.11.001]. International Review of Financial Analysis, 17(5), 
1156-1172.  
Managi, S., Opaluch, J. J., Jin, D., & Grigalunas, T. A. (2006). Stochastic frontier analysis 
of total factor productivity in the offshore oil and gas industry. Ecological 
Economics, 60(1), 204-215. doi: http://dx.doi.org/10.1016/j.ecolecon.2005.11.028 
March, J. G., & Olsen, J. P. (1983). The New Institutionalism: Organizational Factors in 
Political Life. American Political Science Review, 78(03), 734-749. doi: 
doi:10.2307/1961840 
Marius Andrieş, A., & Căpraru, B. (2012). Competition and efficiency in EU27 banking 
systems. Baltic Journal of Economics, 12(1), 41-60.  
Markus, H. (1978). The effect of mere presence on social facilitation: An unobtrusive test. 
Journal of Experimental Social Psychology, 14(4), 389-397.  
Martens, R. (1969). Effect on performance of learning a complex motor task in the presence 
of spectators. Research Quarterly. American Association for Health, Physical 
Education and Recreation, 40(2), 317-323.  
Marwell, G., Oliver, P. E., & Prahl, R. (1988). Social Networks and Collective Action: A 
Theory of the Critical Mass. III. American journal of sociology, 94(3), 502-534. doi: 
10.2307/2780252 
Meyer, J. W., Ramirez, F. O., & Soysal, Y. N. (1992). World expansion of mass education, 
1870-1980. Sociology of education, 128-149.  
144 
 
University of Ghana  http://ugspace.ug.edu.gh
Meyer, J. W., & Rowan, B. (1977). Institutionalized Organizations: Formal Structure as 
Myth and Ceremony. American journal of sociology, 83(2), 340-363. doi: 
10.2307/2778293 
Mikdashi, Z. (1974). Cooperation Among Oil Exporting Countries with Special Reference 
to Arab Countries: a Political Economy Analysis. International Organization, 
28(01), 1-30. doi: doi:10.1017/S0020818300004343 
Miles, J. A. (2012). Management and organization theory: A Jossey-Bass reader (Vol. 9): 
John Wiley & Sons. 
Miller, D., & Shamsie, J. (1996). The resource-based view of the firm in two environments: 
The Hollywood film studios from 1936 to 1965. Academy of management journal, 
39(3), 519-543.  
Minescu, A., Hagendoorn, L., & Poppe, E. (2008). Types of identification and intergroup 
differentiation in the Russian Federation. Journal of Social Issues, 64(2), 321-342.  
Mitchell, J. C. (1969). The concept and use of social networks: Bobbs-Merrill. 
Muldoon, K. A., Galway, L. P., Nakajima, M., Kanters, S., Hogg, R. S., Bendavid, E., & 
Mills, E. J. (2011). Health system determinants of infant, child and maternal 
mortality: A cross-sectional study of UN member countries. Global Health, 7(42).  
Murphy, D. J., & Hall, C. A. (2011a). Energy return on investment, peak oil, and the end of 
economic growth. Annals of the New York Academy of Sciences, 1219(1), 52-72.  
Murphy, D. J., & Hall, C. A. S. (2011b). Energy return on investment, peak oil, and the end 
of economic growth. Annals of the New York Academy of Sciences, 1219(1), 52-72. 
doi: 10.1111/j.1749-6632.2010.05940.x 
Nash, J. (1996). Equilibrium Points in N-Person Games', Proceedings of the National 
Academy of Sciences of the United States, 36, 48-9. INTERNATIONAL LIBRARY 
OF CRITICAL WRITINGS IN ECONOMICS, 67, 52-53.  
145 
 
University of Ghana  http://ugspace.ug.edu.gh
Newman, M., Barabasi, A.-L., & Watts, D. J. (2006). The structure and dynamics of 
networks: Princeton University Press. 
North, D. C. (1990). Institutions, Institutional Change and Economic Performance: 
Cambridge University Press. 
O'Toole, L. J. (2004). The Theory–Practice Issue in Policy Implementation Research. Public 
Administration, 82(2), 309-329. doi: 10.1111/j.0033-3298.2004.00396.x 
O’Donnell, C., Rao, D., & Battese, G. (2008a). Metafrontier frameworks for the study of 
firm-level efficiencies and technology ratios. Empirical Economics, 34(2), 231-255. 
doi: 10.1007/s00181-007-0119-4 
O’Donnell, C., Rao, D. S. P., & Battese, G. (2008b). Metafrontier frameworks for the study 
of firm-level efficiencies and technology ratios. Empirical Economics, 34(2), 231-
255. doi: 10.1007/s00181-007-0119-4 
OAPEC. (2010). ANNUAL REPORT (Vol. 37). Kuwait: OAPEC. 
OAPEC. (2015). Current Developments in the World and Arab Energy Industry (Vol. 41). 
Kuwait: OAPEC. 
Ocasio, W. (1997). TOWARDS AN ATTENTION-BASED VIEW OF THE FIRM 
WILLIAM OCASlO. Psychology, 1, 403-404.  
Oderkirk, J., Ronchi, E., & Klazinga, N. (2013). International comparisons of health system 
performance among OECD countries: Opportunities and data privacy protection 
challenges. Health Policy, 112(1–2), 9-18. doi: 
http://dx.doi.org/10.1016/j.healthpol.2013.06.006 
Oliver, C. (1997). Sustainable competitive advantage: Combining institutional and 
resource-based views. Strategic Management Journal, 18(9), 697-713.  
OPEC. (2002). Annual Statistical Bulletin. Vienna,Australia: OPEC. 
OPEC. (2012). Statute. Vienna,Australia: OPEC. 
146 
 
University of Ghana  http://ugspace.ug.edu.gh
OPEC. (2014). Annual Statistical Bulletin. Vienna,Australia: OPEC. 
OPEC. (2015). Annual Statistical Bulletin OPEC. 
Otsuki, T. (2013). Nonparametric measurement of the overall shift in the technology 
frontier: an application to multiple-output agricultural production data in the 
Brazilian Amazon. Empirical Economics, 44(3), 1455-1475. doi: 10.1007/s00181-
012-0582-4 
Oum, T. H., & Yu, C. (1994). Economic efficiency of railways and implications for public 
policy: A comparative study of the OECD countries' railways. Journal of transport 
economics and policy, 121-138.  
Park, C.-u. (2015). Institutional embeddedness of market integration: The formation of free 
trade agreements in 1957–2008. International Sociology, 30(1), 39-60.  
Penrose, E. T. (1959). The Theory of the Growth of the Firm (3rd ed.). Oxford, UK: Oxford 
University Press. 
Peteraf, M. A. (1993). The cornerstones of competitive advantage: A resource-based view. 
Strategic management journal, 14(3), 179-191. doi: 10.1002/smj.4250140303 
Peteraf, M. A., & Barney, J. B. (2003). Unraveling the resource-based tangle. Managerial 
and Decision Economics, 24(4), 309-323. doi: 10.1002/mde.1126 
Podkorytova, O., & Raskina, Y. (2014). Former Soviet Union countries and European 
Union: overcoming the energy efficiency gap. Robert Schuman Centre for Advanced 
Studies Research Paper(2014/03).  
Price, C. W., & Weyman-Jones, T. (1996). Malmquist indices of productivity change in the 
UK gas industry before and after privatization. [10.1080/000368496327633]. 
Applied Economics, 28, 29-39.  
Ramachandra, T., Loerincik, Y., & Shruthi, B. (2006). Intra-and Inter-Country Energy 
Intensity Trends. JOURNAL OF ENERGY AND DEVELOPMENT, 31(1), 43.  
147 
 
University of Ghana  http://ugspace.ug.edu.gh
Ramanathan, R. (2007). Performance of banks in countries of the Gulf Cooperation Council. 
[DOI: 10.1108/17410400710722635]. International Journal of Productivity and 
Performance Management, 56(2), 137-154.  
Ramcharran, H. (2002). Oil production responses to price changes: an empirical application 
of the competitive model to OPEC and non-OPEC countries. Energy Economics, 
24(2), 97-106.  
Rasmussen, E. (1989). Games and Information Oxford: Basil Blackwell. 
Retzlaff-Roberts, D., Chang, C. F., & Rubin, R. M. (2004). Technical efficiency in the use 
of health care resources: a comparison of OECD countries. Health Policy, 69(1), 55-
72. doi: http://dx.doi.org/10.1016/j.healthpol.2003.12.002 
Rigby, J., Dewick, P., Courtney, R., & Gee, S. (2013). Limits to the Implementation of 
Benchmarking Through KPIs in UK Construction Policy: Insights from game 
theory. Public Management Review, 16(6), 782-806. doi: 
10.1080/14719037.2012.757351 
Rogner, H., Aguilera, R. F., Archer, C., Bertani, R., Bhattacharya, S., Dusseault, M., . . . 
Johnson, A. (2012). Energy resources and potentials.  
Rumelt, R. P. (1974). Strategy, structure, and economic performance.  
Sari, R., & Soytas, U. (2009). Are global warming and economic growth compatible? 
Evidence from five OPEC countries? Applied Energy, 86(10), 1887-1893.  
Schafer, J. L. (1999). Multiple imputation: a primer. Statistical methods in medical 
research, 8(1), 3-15.  
Scott, W. R. (1995). Institutions and organizations. Thousand Oaks [etc.]: Sage. 
Selowsky, M., & Martin, R. (1997). Policy performance and output growth in the transition 
economies. The American Economic Review, 349-353.  
148 
 
University of Ghana  http://ugspace.ug.edu.gh
Shahabinejad, V., Mehrjerdi, M. R. Z., & Yaghoubi, M. (2013). Total Factor Productivity 
Growth, Technical Change and Technical Efficiency Change in Asian Economies: 
Decomposition Analysis. Iranian Journal of Economic Studies, 2(2), 47-69.  
Shanks, C., Jacobson, H. K., & Kaplan, J. H. (1996). Inertia and change in the constellation 
of international governmental organizations, 1981–1992. International 
Organization, 50(04), 593-627.  
Simar, L., & Wilson, P. W. (1998). Sensitivity Analysis of Efficiency Scores: How to 
Bootstrap in Nonparametric Frontier Models. Management Science, 44(1), 49-61. 
doi: 10.1287/mnsc.44.1.49 
Simar, L., & Wilson, P. W. (1999). Estimating and bootstrapping Malmquist indices. 
European Journal of Operational Research, 115(3), 459-471. doi: 
http://dx.doi.org/10.1016/S0377-2217(97)00450-5 
Simar, L., & Wilson, P. W. (2000). A general methodology for bootstrapping in non-
parametric frontier models. Journal of Applied Statistics, 27(6), 779 - 802.  
Simar, L., & Wilson, P. W. (2002). Non-parametric tests of returns to scale. [doi: DOI: 
10.1016/S0377-2217(01)00167-9]. European Journal of Operational Research, 
139(1), 115-132.  
Simar, L., & Wilson, P. W. (2011). Inference by the m out of n bootstrap in nonparametric 
frontier models. Journal of Productivity Analysis, 36(1), 33-53. doi: 
10.1007/s11123-010-0200-4 
Simar, L., & Wilson, P. W. (2015). Statistical Approaches for Non‐parametric Frontier 
Models: A Guided Tour. International Statistical Review, 83(1), 77-110.  
Simar, L., & Zelenyuk, V. (2006). On Testing Equality of Distributions of Technical 
Efficiency Scores. Econometric Reviews, 25(4), 497-522. doi: 
10.1080/07474930600972582 
149 
 
University of Ghana  http://ugspace.ug.edu.gh
Sirmon, D. G., Hitt, M. A., Ireland, R. D., & Gilbert, B. A. (2011). Resource orchestration 
to create competitive advantage breadth, depth, and life cycle effects. Journal of 
Management, 37(5), 1390-1412.  
Stevens, P. (2008). A methodology for assessing the performance of national oil companies. 
Washington, DC, World Bank.  
Strange, S. (1989). States and markets. London: Pinter. 
Sueyoshi, T., & Goto, M. (2012a). Data envelopment analysis for environmental 
assessment: Comparison between public and private ownership in petroleum 
industry. European Journal of Operational Research, 216(3), 668-678. doi: 
http://dx.doi.org/10.1016/j.ejor.2011.07.046 
Sueyoshi, T., & Goto, M. (2012b). Environmental assessment by DEA radial measurement: 
U.S. coal-fired power plants in ISO (Independent System Operator) and RTO 
(Regional Transmission Organization). Energy Economics, 34(3), 663-676. doi: 
http://dx.doi.org/10.1016/j.eneco.2011.08.016 
Sueyoshi, T., & Goto, M. (2014). Environmental assessment for corporate sustainability by 
resource utilization and technology innovation: DEA radial measurement on 
Japanese industrial sectors. Energy Economics, 46(0), 295-307. doi: 
http://dx.doi.org/10.1016/j.eneco.2014.09.021 
Szilas, A. P. (1985). Production and Transport of Oil and Gas: Gathering and 
transportation: Elsevier. 
Taylor, P. G., d’Ortigue, O. L., Francoeur, M., & Trudeau, N. (2010). Final energy use in 
IEA countries: The role of energy efficiency. Energy Policy, 38(11), 6463-6474. doi: 
http://dx.doi.org/10.1016/j.enpol.2009.05.009 
Teece, D., & Pisano, G. (1994). The dynamic capabilities of firms: an introduction. 
Industrial and corporate change, 3(3), 537-556.  
150 
 
University of Ghana  http://ugspace.ug.edu.gh
Teece, D. J., Pisano, G., & Shuen, A. (1997). Dynamic capabilities and strategic 
management. Strategic management journal, 18(7), 509-533.  
Thanassoulis, E. (2001). Introduction of the Theory and Application of Data Envelopment 
Analysis (1st ed.). New York: Springer. 
Thompson, R., Dharmapala, P. S., Rothenberg, L., & Thrall, R. (1994). DEA ARs and CRs 
applied to worldwide major oil companies. Journal of Productivity Analysis, 5(2), 
181-203. doi: 10.1007/bf01073854 
Thompson, R. G., Dharmapala, P. S., Humphrey, D. B., Taylor, W. M., & Thrall, R. M. 
(1996). Computing DEA/AR efficiency and profit ratio measures with an illustrative 
bank application. Annals of Operations Research, 68, 303-327.  
Tichy, N. M., Tushman, M. L., & Fombrun, C. (1979). Social network analysis for 
organizations. Academy of management review, 4(4), 507-519.  
Tindall, D. B., & Wellman, B. (2001). Canada as social structure: Social network analysis 
and Canadian sociology. Canadian Journal of Sociology, 265-308.  
Tordo, S., Tracy, B. S., & Arfaa, N. (2011). National Oil Companies and Value Creation: 
Washington: World Bank. 
Tortosa-Ausina, E., Grifell-Tatjé, E., Armero, C., & Conesa, D. (2008). Sensitivity analysis 
of efficiency and Malmquist productivity indices: An application to Spanish savings 
banks. European Journal of Operational Research, 184(3), 1062-1084. doi: DOI: 
10.1016/j.ejor.2006.11.035 
Trestian, R., Ormond, O., & Muntean, G.-M. (2012). Game Theory-Based Network 
Selection: Solutions and Challenges. Communications Surveys & Tutorials, IEEE, 
14(4), 1212-1231. doi: 10.1109/surv.2012.010912.00081 
151 
 
University of Ghana  http://ugspace.ug.edu.gh
Trinh, K. D., & Zelenyuk, V. (2015). Bootstrap-based testing for network DEA: Some 
Theory and Applications: School of Economics, University of Queensland, 
Australia. 
Triplett, N. (1898). The dynamogenic factors in pacemaking and competition. The American 
journal of psychology, 9(4), 507-533.  
Uziel, L. (2007). Individual differences in the social facilitation effect: A review and meta-
analysis. Journal of Research in Personality, 41(3), 579-601. doi: 
http://dx.doi.org/10.1016/j.jrp.2006.06.008 
Vlontzos, G., Niavis, S., & Manos, B. (2014). A DEA approach for estimating the 
agricultural energy and environmental efficiency of EU countries. Renewable and 
Sustainable Energy Reviews, 40, 91-96.  
Volgy, T. J., Fausett, E., Grant, K. A., & Rodgers, S. (2008). Identifying Formal 
Intergovernmental Organizations. Journal of Peace Research, 45(6), 837-850. doi: 
10.1177/0022343308096159 
Von Neumann, J., & Morgenstern, O. (1944). Theory of games and economic behavior. 
Princeton, NJ: Princeton University Press. 
von Neumann, J., & Morgenstern, O. (1947). Theory of games and economic behavior. 
Princeton: Princeton University Press. 
Wanke, P., Barros, C. P., & Faria, J. R. (2015). Financial distress drivers in Brazilian banks: 
A dynamic slacks approach. European Journal of Operational Research, 240(1), 
258-268. doi: http://dx.doi.org/10.1016/j.ejor.2014.06.044 
Wasserman, S., & Faust, K. (1994). Social network analysis: Methods and applications: 
Cambridge University Press. 
Weibull, J. W. (1997). Evolutionary Game Theory: MIT Press. 
152 
 
University of Ghana  http://ugspace.ug.edu.gh
Wernerfelt, B. (1984). A resource-based view of the firm. Strategic Management Journal, 
5(2), 171-180. doi: 10.1002/smj.4250050207 
Wiklund, J., & Shepherd, D. (2003). Research notes and commentaries: knowledge-based 
resources, entrepreneurial orientation, and the performance of small and medium-
sized businesses. Strategic Management Journal, 24(13), 1307-1314.  
Wilson, J. (2010). Essentials of Business Research Jonathan Wilson SAGE. SAGE, 2010, 
04-20.  
Wolf, C. (2009). Does ownership matter? The performance and efficiency of State Oil vs. 
Private Oil (1987–2006). Energy Policy, 37(7), 2642-2652. doi: 
http://dx.doi.org/10.1016/j.enpol.2009.02.041 
World Bank. (2011). The Changing Wealth of Nations: Measuring Sustainable 
Development in the New Millennium. Washington DC: The International Bank for 
Reconstruction and Development- The World Bank. 
Yergin, D. (1991). Oil: the strategic prize. The Gulf War Reader, New York, Times Books, 
21-26.  
Yergin, D. (2011). The prize: The epic quest for oil, money & power: Simon and Schuster. 
Yin, Z., Jiang, A. X., Tambe, M., Kiekintveld, C., Leyton-Brown, K., Sandholm, T., & 
Sullivan, J. P. (2012). TRUSTS: Scheduling randomized patrols for fare inspection 
in transit systems using game theory. AI Magazine, 33(4), 59.  
Zajonc, R. B. (1965). Social facilitation: Research Center for Group Dynamics, Institute for 
Social Research, University of Michigan. 
Zajonc, R. B. (1968). Social facilitation in cockroaches. In E. C. Simmel, R. A. Hoppe & 
G. A. Milton (Eds.), Social facilitation and imitative behaviour (pp. 73-90). Boston: 
Allyn & Bacon. 
153 
 
University of Ghana  http://ugspace.ug.edu.gh
Zajonc, R. B., & Sales, S. M. (1966). Social facilitation of dominant and subordinate 
responses. Journal of Experimental Social Psychology, 2(2), 160-168. doi: 
http://dx.doi.org/10.1016/0022-1031(66)90077-1 
Zelenyuk, V., & Zheka, V. (2006). Corporate Governance and Firm’s Efficiency: The Case 
of a Transitional Country, Ukraine. Journal of Productivity Analysis, 25(1-2), 143-
157. doi: 10.1007/s11123-006-7136-8 
Zhou, P., Ang, B. W., & Poh, K. L. (2008). A survey of data envelopment analysis in energy 
and environmental studies. European Journal of Operational Research, 189(1), 1-
18. doi: http://dx.doi.org/10.1016/j.ejor.2007.04.042 
Zucker, L. G. (1987). Institutional Theories of Organization. Annual Review of Sociology, 
13(ArticleType: research-article / Full publication date: 1987 / Copyright © 1987 
Annual Reviews), 443-464. doi: 10.2307/2083256 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
154 
 
University of Ghana  http://ugspace.ug.edu.gh
APPENDIX A 
Taxonomy of Papers on Efficiency of IGO states 
No. Author Issue Methods Sample Industry  IGO 
1.  Abu-Alkheil, Bank efficiency DEA 40 Banks in UK, Banking  OAPEC 
Burghof & Khan,  Turkey, GCC GCC  
(2012) state & Malaysia  
(2005-2008) 
2.  Adetutu (2014) Energy efficiency Modified 4 OPEC Countries Energy  OPEC 
Translog Cost (1972–2010) 
Function 
3.  Adler-Milstein J, Health efficiency Conceptual 20 OECD countries Health  OECD 
et al.(2014) 
4.  Afonso & Aubyn Education and FDH 24 OECD Education/ OECD 
Countries 
(2004) Health efficiency DEA Health 
 
5.  Al-Rashed & León Energy efficiency Energy OPEC countries Energy OPEC 
(2015) coefficient 
(2010) 
6.  Al-Rubaie, Salek Efficiency of Regression  30 responses were Health OAPEC 
& Walker (2015) pharmaceuticals from 17 
GCC 
international, 7 
GCC, 3 non-GCC  
Arab, and 3 Asian  
7.  Andrieş  and Competition and SFA 27 EU Countries Banking  EU 
Căpraru Bank efficiency 
Regression (2001- 2009) 
(2012)  
8.  Arestis, Chortareas Efficiency of DEA 26 OECD Economy OECD 
& Desli (2006) economy Countries 
(1963–1992) 
9.  Aristovnik (2012) Educational DEA 27 EU and OECD Education EU 
efficiency Countries 
OECD 
(1999–2007) 
10.  Arnade (1994) Agricultural DEA 77 countries EU, Agriculture  EU, etc. 
USA, Australia & 
 Efficiency and New Zealand  
Productivity 
 (1961 - 1987) 
 
11.  Asongu (2013) Financial efficiency Regression 11 AU Countries Economy AU 
 (1981 - 2009) 
 
155 
 
University of Ghana  http://ugspace.ug.edu.gh
12.  Behname (2012) Bank efficiency  DEA Regression 8  OPEC Countries Banking  OPEC 
(1995-2009) 
  
13.  Bosseboeuf et al Energy efficiency Energy intensity 9 EU member Energy EU 
(1997) ratio and countries (1970 - efficiency 
technico- 1993) 
economic 
approach 
14.  Çakır, Perçin & Efficiency of the DEA 25 OECD Postal OECD 
Min (2015) postal services Countries (2010) Service 
 
15.  Casu & Girardone Bank efficiency SFA 5 EU Countries Banking EU 
(2004) (1990s) 
DEA 
Malmquist 
16.  Casu & Girardone Bank efficiency Panzar and EU-15 Countries Banking  EU 
(2006) Rosse model (1997–2003) 
DEA  
17.  Casu & Molyneux Bank efficiency DEA 5 EU Countries Banking  EU 
(1993 - 1997) 
(2003) Tobit 
Regression 
18.  Claeys &Vennet Bank efficiency Translog Cost 1130 banks from Banking  EU 
(2008) Function 31 European 
countries (1994-
2001) 
19.  Dike  (2013) Measuring the REED index 12 OPEC Countries Oil  OPEC 
security of energy 
exports demand in (2009) 
OPEC economies 
20.  Donni & Fecher Efficiency and DEA and 15 OECD Insurance  OECD 
(1997) productivity of the Malmquist Countries 
insurance industry (1983-1991) 
21.  Drakos (2003) Bank efficiency Regression 10 CEE and FSU Banking  FSU 
Countries 
CEE 
(1993–1999) 
22.  Fare, Grosskopf, Productivity DEA 17 OECD Economy OECD 
Norris and Zhang  Growth, Countries 
(1994) Malmquist 
Technical Progress, (1979-1988) 
and Efficiency 
Change 
23.  Filippini  &  Hunt Energy efficiency SFA 29 OECD Energy  OECD 
Countries 
(2009) 
156 
 
University of Ghana  http://ugspace.ug.edu.gh
(1978 - 2006) 
24.  Fredriksson, Energy efficiency Regression 12 OECD Energy OECD 
Vollebergh and Countries 
Dijkgraaf (2004) (1982–1996) 
25.  Fu et al. (2015) Economy Regression China and G20 Economy G20  
member countries 
26.  Geller, Harrington, Energy efficiency Energy intensity 10 OECD Energy OECD 
Rosenfeld, Countries (1973-
Tanishima, 1998) 
Unander (2006) 
27.  Goldthau & Witte Performance in Review OPEC Oil OPEC 
(2011) energy 
(1960-2009) 
 
28.  Gorton & Farm productivity DEA 6 CEE countries Agriculture CEE 
Davidova (2004) and efficiency  
SFA 
 
29.  Gupta & Efficiency of FDH 37 Countries in Education/ AU 
Verhoeven (1999) government Africa Union 
expenditure Health 
(1984–1995) 
30.  Hori (2012) Energy efficiency Energy intensity OECD and Non- Energy OECD 
OECD countries 
 
31.  Jollands et al Energy efficiency Review/Policy G8 (2008) Energy G8 
(2010) 
  
 
32.  Košak, Zajc and Bank efficiency  SFA 5 New EU Member Banking  EU 
Zorić States 
 
(2009) (1996 – 2006) 
33.  Krishnasamy &  Efficiency and Metafrontier 26 OECD Economy OECD 
Ahmed (2009) Productivity of Countries 
economy MPI 
(1980-2008) 
34.  Mamatzakis, Bank efficiency  SFA 10 E U Countries Banking  EU 
Staikouras and 
Koutsomanoli-  ( 1998–2003) 
Filippaki (2008) 
35.  Meier et al.(2013) Fuel energy Review 6 IEA members Bioenergy IEA 
 
 
157 
 
University of Ghana  http://ugspace.ug.edu.gh
36.  Muldoon et al Health efficiency Regression 192 UN Health  UN 
(2011) 
Countries 
(2001-2008) 
37.  Oderkirk, Ronchi Health efficiency Regression 20 OECD countries Health OECD 
& Klazzinga 
(2013) (2011-2012) 
38.  Oum & Yu (1994) Efficiency of DEA  19 OECD Railways  OECD 
Railways Countries 
TFP 
(1978-89) 
Tobit regression 
39.  Ramcharran  Efficiency and Oil Griffin’s Model 12 OPEC Countries Oil  OPEC 
(2002) production intensity ratios (1973-1997) 
responses to price and 
 changes 
 
40.  Reiche (2010) Energy efficiency Review/Policy 6 GCC countries Energy OAPEC 
 GCC 
41.  Retzlaff-Roberts, Technical efficiency DEA 29 OECD Health  OECD 
Chang & Rubin in health care Countries (1998) 
(2003) 
 
42.  Sari & Soytas Efficiency Auto Regressive 5 OPEC Countries Economy  OPEC 
(2009) Distributed Lag (1971–2002.) 
(ARDL) 
Approach, 
43.  Selowsky & Policy Performance  Regression 25 CEE & FSU Economy CEE 
Martin 
 (1990-1995) FSU 
(1997) 
44.  Shahabinejad, Productivity SFA 44 Asian Countries Economy ASEAN 
Mehrjerdi & (2000-2010) 
Yaghoubi (2013)   
45.  Sharma & Thomas R&D efficiency  DEA 22 UN member R&D UN 
Countries (2004) 
(2008)  
46.  Staníčková & Efficiency DEA 27  EU Countries Economy EU 
Skokan (2012) (2000 -2010) 
 
47.  Taylor  (2010) Energy efficiency Review 16 IEA countries Energy IEA 
48.  Vlontzos et al Agricultural energy DEA 25 EU Countries Agriculture  EU 
(2014). (2001–2008) 
and environmental 
 efficiency 
IEA – OAPEC – FSU EU AU ASEAN UN GCC CEE 
 
158 
 
University of Ghana  http://ugspace.ug.edu.gh
APPENDIX B 
CHRONOLOGICAL TAXONOMY OF EFFICIENCY STUDIES IN OIL AND GAS INDUSTRY 
NO. AUTHOR (YEAR) ISSUES METHOD INPUTS OUTPUTS SAMPLE/STUDY Product SECTOR 
RESEARCH PERIOD 
1. A l-Obaidan and Scully Ownership and SFA  • Total assets  • Total revenue  44 oil companies  Oil and Upstream  
(1991)  efficiency  (1979-1983)  Gas 
• Barrels of crude 
 oil produced + 
 barrels of crude 
 oil refined 
2. A l-Obaidan and Scully Backward Vertical Aigner-Chu • Total assets • Total revenue 55 oil companies Oil and Upstream  
(1993)  Integration deterministic frontier (1979-1982)  Gas 
3. T hompson DEA/ AR and DEA • Total Cost • Additions to 14 integrated oil Oil and Upstream  
Dharmapala,  Profitability    reserves companies in US Gas 
Rothenberg and Thrall • Proved (combined)  (1980-1987)  
(1994)   reserves  
(combined) • Sales of 
 production from 
reserves 
 
4. A l-Obaidan and  Scully, Multinationality  SFA • Total assets • Total revenue  44 oil companies Oil and Upstream  
(1995)    (1976-1982) Gas 
. Barrels crude oil 
produced + 
barrels crude oil 
refined 
5. P rice and Weyman- Privatization  DEA • Number of • Domestic gas 12 distribution Gas Downstream 
Jones (1996)  SFA employees  sales regions in UK 
• Length of gas  • Industrial gas (1977-78 to 1990-
Mains sales 91)  
Transmission • Commercial gas 
sales  
159 
 
University of Ghana  http://ugspace.ug.edu.gh
 and distribution • Number of 
system Customers  
• Gas using 
appliances sold 
6. L ee, Park and Kim. International Edgeworth Index  • Capital • Gas deliveries  28 natural gas Gas Downstream  
(1996)  Comparison   • Labour              transportation 
•Administration utilities in 8 
 countries (1987-
 1995)  
 
7. T hompson, Conceptual Paper  DEA  • Total • Oil Production• 30 oil companies Oil and Upstream  
Dharmapala, Diaz, production costs Gas Production (1983-1985)  Gas 
Gonzalez-Lima and • Total proven  
Thrall(1996)  reserves of 
crude 
• Total 
exploratory and 
development 
wells drilled  
• Total proven 
reserves of 
natural gas 
8. T hompson, Application od DEA  • Expenditure in • Crude oil 14 integrated oil Oil and Upstream  
Dharmapala, DEA AR and CR   exploration discovered companies in US Gas 
Rothenberg and Thrall  • Crude oil proved reserves      (1980-1991)  
(1996)  reserves • Natural gas 
• Natural gas discovered 
reserves proved reserves 
  
9. K im, Lee, Park and International Multilateral Tornqvist • Labour • Total volume of 28 Natural Gas Gas Downstream  
Kim, (1999)  Comparison, Managerial Index • Capital gas supplied transmission and 
Determinants of System Analysis  (Assets)  distribution 
Productivity  Non-parametric • Administration • Revenue from companies 
Efficiency Analysis  gas operating 8 
  transportation countries (1987-
  1995)  
10. H awdon (2003)  Regulation DEA  • Employment • Gas Country-level  Gas Downstream 
• Length of  Consumption Dataof 33 
160 
 
University of Ghana  http://ugspace.ug.edu.gh
Bootstrapping pipelines • Number of countries 
Customers (1998, 1999)  
Gas Industry  
11. I smail, Tai, Kong,  Environmental DEA  • Assets  • Revenue 17 Oil Companies Oil and Upstream 
Law, Shirazi and Efficiency   • Employee  (2008) Gas 
Karim (2003) Numbers 
12. K ashani (2005)  Regulation DEA •Construction • Oil Production 66 oil and gas fields, Oil and Upstream 
SFA Regression Cost                      • • Gas Production  67 oil fields. United  Gas 
 Variable Cost  Kingdom. (1974-
 • Water depth  1991)  
• Revenue depth 
• Number of 
partners 
13. K ashani (2005)  State Intervention DEA • Construction • Oil Production 37 Gas Fields in Oil and Upstream  
SFA Cost                     • • Gas Production Norway. (1972- Gas 
Regression  Variable Cost•  2000)  
Water depth 
• Revenue depth 
• Number of 
partners 
14. M anagi, Opaluch, Jin Technology Regression  • Drilling • Quantity of oil  370 Drilling wells in Oil and Upstream 
and Grigalunas, (2005) Change  distance per and gas reserves gulf of Mexico-US Gas 
exploratory well  discovered in (1947-1998)  
• Drilling  barrels of oil  
distance per equivalent  
development 
well  
• Total number 
of exploratory 
and 
development 
wells  
• Price of oil & 
gas  
• Water depth 
 
161 
 
University of Ghana  http://ugspace.ug.edu.gh
15. M anagi, Opaluch, Jin Technology SFA • Oil reserves• • Porosity  370 Drilling wells in Oil and Upstream 
and Grigalunas (2006)  Change  Water depth  gulf of Mexico-US Gas 
• 5 yrs. ex drill (1947-1998)  
mills ratio 
• Gas reserves 
 
16. B arros and Assaf Bootstrapping  DEA •Operational • Gross production 9 Angolan oil  Oil Upstream 
(2009)  Bootstrapping cost Blocks 
Bootstrapped  • Investment   (2002-2007)  
truncated regression Premium 
• Taxes 
17. B arros and Managi Growth DEA •Operational • Gross 9 Angolan oil blocks Oil Upstream 
(2009) Accounting vs cost production  (2002-2007)  
Productivity • Investment   
Method  premium• Taxes  
18. W olf (2009)  Ownership  and Regression • Oil and gas • Annual oil and 87 oil firms       Oil and Upstream  
efficiency reserves gas production            (  1  9  87-2006)  Gas 
• OPEC • Revenues 
membership • Net income 
(Binary)  
• Total assets   
• State 
ownership 
Percentage  
• Ratio of oil and 
gas reserves  
• Number of 
employees  
19. E ller, Hartley and Ownership  DEA • Oil reserves  • Revenues 78 oil firms (2006)  Oil and Upstream 
Medlock (2011). SFA • Number of Gas 
 employees• 
Natural gas 
reserves 
20. F rancisco ,de Almeida Environmental DEA • Amount of • Processed oil  10 Brazilian Oil and Downstream 
and de Silva (2012)  Efficiency                 water consumed • Effluents  Refineries  Gas 
 • Percentage of (Undesirable) (2004) 
Idleness 
162 
 
University of Ghana  http://ugspace.ug.edu.gh
 • Age of Refinery 
(Uncontrollable)                     
21. S ueyoshi and Goto Conceptual Paper: DEA  • Amount of Oil • Oil Production       19 oil firms. (2005- Oil and Upstream  
(2012)  Environmental  Reserves                     •   Gas Production• 2009)  Gas 
Efficiency and • Amount of Gas CO2 emission 
Ownership  Reserves                   (Undesirable) 
• Total operating  
cost   
• Number of  
employees 
22. S ueyoshi and Goto Conceptual Paper: DEA             • Amount of Oil • Oil Production• 19 oil firms. (2005- Oil and Upstream 
(2012)  Environmental  Reserves Gas Production 2009)  Gas 
Efficiency and • Amount of Gas • CO2 emission 
Ownership  Reserves (Undesirable) 
• Total operating 
cost  
• Number of 
employees 
23. B arros and Antunes Productivity Luenberger •Operational • Production of 9 Angolan oil  Oil Upstream 
(2014)  Change. Malmquist Productivity Indicator cost  oil  Blocks 
vs Luenberger  • Taxes • Investment  (2002-2008)  
premium 
 
24. I ke and Lee, H. (2014) Ownership DEA • Oil reserves • Oil production 38 oil companies Oil and Upstream 
Slack Based • Gas reserves • Gas Production  (2003-2010)  Gas 
MPI • Number of  
Regression employees 
 
 
 
 
 
 
 
 
163 
 
University of Ghana  http://ugspace.ug.edu.gh
APPENDIX C 
DESCRIPTIVE STATISTICS BY TIME 
ANOVA 
  N Mean Std. Deviation Minimum Maximum F Sig. 
Oil Reserves 2000 52 27.55 58.04 0.01 263.50 0.216 0.998 
2001 52 27.91 57.71 0.01 261.70    
2002 52 28.05 57.70 0.01 261.75    
2003 52 31.91 61.29 0.01 261.80    
2004 52 33.11 62.32 0.01 261.90    
2005 52 33.44 62.44 0.01 261.90    
2006 52 33.88 63.51 0.01 266.81    
2007 51 34.85 63.26 0.01 262.30    
2008 51 35.29 64.16 0.01 266.75    
2009 52 34.91 63.82 0.01 266.71    
2010 52 35.26 62.99 0.01 262.40    
2011 52 37.53 67.02 0.01 262.60    
2012 49 39.56 71.33 0.01 267.02    
2013 52 41.33 74.70 0.01 297.57    
Total 723 33.87 63.28 0.01 297.57    
Gas 2000 52 106.04 260.35 0.04 1700.00 0.137 1.000 
Reserves 2001 52 110.27 263.90 0.02 1700.00    
2002 52 115.42 267.30 0.02 1680.00    
2003 52 116.63 267.06 0.02 1680.00    
2004 52 135.44 305.83 0.04 1680.00    
2005 52 135.01 306.20 0.04 1680.00    
2006 52 136.56 308.04 0.04 1680.00    
2007 51 140.24 310.52 0.04 1680.00    
2008 51 140.68 308.99 0.07 1680.00    
2009 52 138.72 307.69 0.07 1680.00    
2010 52 145.75 311.72 0.04 1680.00    
2011 52 147.21 311.89 0.04 1680.00    
2012 49 155.20 328.04 0.04 1680.00    
2013 52 149.12 321.10 0.04 1688.00    
Total 723 133.63 296.81 0.02 1700.00    
Labour 2000 52 14236015.56 24214013.86 303604.00 147134193.00 0.070 1.000 
Force 2001 52 14369218.42 24276345.89 319802.00 148216979.00    
2002 52 14555641.44 24403122.85 335376.00 149007489.00    
2003 52 14757966.10 24552424.66 355406.00 149705300.00    
2004 52 14968418.56 24714179.17 382853.00 150729170.00    
2005 52 15214033.67 24985272.40 424303.00 152676462.00    
2006 52 15417363.88 25232314.85 477508.00 154694540.00    
2007 51 15813970.04 25671774.47 540749.00 155976570.00    
2008 51 16029800.27 25890710.46 605930.00 157724796.00    
2009 52 16021090.44 25686856.43 664212.00 157889958.00    
2010 52 16182553.33 25653982.28 707016.00 157464257.00    
2011 52 16351310.33 25683846.47 731901.00 157635584.00    
2012 49 17361678.67 26375113.98 741723.00 158786582.00    
2013 52 16706223.83 25903640.65 738890.00 159144632.00    
Total 723 15561972.32 25022196.20 303604.00 159144632.00    
164 
 
University of Ghana  http://ugspace.ug.edu.gh
Oil 2000 52 1463.53 2376.25 0.39 9475.75 0.120 1.000 
Production 2001 52 1450.19 2344.96 0.33 9156.64    
2002 52 1420.02 2308.82 0.31 8998.43    
2003 52 1488.37 2512.56 0.35 10076.81    
2004 52 1596.69 2658.80 0.25 10796.24    
2005 52 1646.90 2751.14 0.28 11496.31    
2006 52 1646.07 2705.06 0.30 11098.44    
2007 51 1669.42 2695.40 0.28 10748.62    
2008 51 1721.25 2793.66 0.24 11428.60    
2009 52 1655.06 2656.51 0.22 10314.71    
2010 52 1703.70 2779.79 0.22 10908.35    
2011 52 1714.87 2907.60 0.22 11469.90    
2012 49 1846.31 3101.30 0.22 11840.68    
2013 52 1792.97 3095.59 0.21 12342.77    
Total 723 1628.59 2679.15 0.21 12342.77    
Gas 2000 52 1945.14 4385.98 0.46 24174.00 0.155 1.000 
Production 2001 52 1975.79 4411.85 0.57 24501.00    
2002 52 2001.99 4416.89 0.35 23941.00    
2003 52 2053.45 4509.73 0.71 24119.00    
2004 52 2137.78 4572.80 0.71 23970.00    
2005 52 2216.74 4562.44 0.53 23457.00    
2006 52 2272.76 4631.64 0.35 23535.00    
2007 51 2403.20 4771.08 0.35 24664.00    
2008 51 2485.87 4867.01 0.35 25636.00    
2009 52 2380.94 4713.91 0.35 26057.00    
2010 52 2519.17 4988.60 0.25 26836.00    
2011 52 2599.57 5276.69 0.21 28479.00    
2012 49 2734.64 5412.11 0.00 29542.00    
2013 52 2629.20 5359.24 0.00 30005.00    
Total 723 2309.03 4749.99 0.00 30005.00    
 
 
 
 
 
 
 
 
 
 
  
 
 
 
 
 
 
 
165 
 
University of Ghana  http://ugspace.ug.edu.gh
APPENDIX D 
SUMMARY OF TUKEY HSD RESULTS 
Summary of Tukey HSD Results for FSU  
Country N Subset for alpha = 0.05 
1 2 3 4 5 6 
Meta Efficiency Scores       
Turkmenistan 14 1.336127      
Belarus 14 2.008064      
Uzbekistan 14 2.718953      
Russia 14 2.900439      
Azerbaijan 14 3.842545      
Kazakhstan 14 4.585289      
Ukraine 14 5.001232      
Kyrgyzstan 14  26.211315     
Georgia 14  26.412623     
Tajikistan 14   73.502855    
        
Group Efficiency Scores       
Belarus 14 1.124731      
Uzbekistan 14 1.142826      
Turkmenistan 14 1.226900      
Kazakhstan 14 1.305782      
Russia 14 1.326069      
Azerbaijan 14 1.569428      
Ukraine 14 1.645614      
Georgia 14  7.668168     
Kyrgyzstan 14  9.191870 9.191870    
Tajikistan 14   10.217240    
Technological Gap Ratios       
Tajikistan 14 .142701      
Georgia 14  .291215     
Kazakhstan 14  .310537     
Ukraine 14  .328497     
Kyrgyzstan 14  .354955 .354955    
Uzbekistan 14   .425406 .425406   
Russia 14    .456646   
Azerbaijan 14    .485949 .485949  
Belarus 14     .562084  
Turkmenistan 14      .926923 
  
166 
 
University of Ghana  http://ugspace.ug.edu.gh
Summary of Tukey HSD Results for IEA States 
Country N Subset for alpha = 0.05 
1 2 3 4 5 6 7 8 9 
        
Meta Efficiency Scores  
Norway 14 1.26         
France 14 1.29 1.29        
Netherlands 14 1.32 1.32        
Denmark 14 1.33 1.33        
Japan 14 1.39 1.39 1.39       
US 14 1.47 1.47 1.47       
Canada 14 1.47 1.47 1.47       
UK 14 1.48 1.48 1.48       
NZ 14 1.82 1.82 1.82 1.82      
Greece 14 1.90 1.90 1.90 1.90      
Hungary 14  1.95 1.95 1.95 1.95     
Czech Republic 14   2.07 2.07 2.07     
Austria 14    2.38 2.38 2.38    
Slovakia 14    2.46 2.46 2.46 2.46   
Spain 14    2.48 2.48 2.48 2.48   
Turkey 14     2.62 2.62 2.62   
Australia 14     2.63 2.63 2.63   
Germany 14      3.01 3.01   
Italy 14       3.07   
  
Group Efficiency Scores        
UK 14 1.15         
Netherlands 14 1.16         
Canada 14 1.16         
Norway 14 1.16         
France 14 1.17         
Denmark 14 1.19 1.19        
Japan 14 1.28 1.28 1.28       
US 14 1.30 1.30 1.30       
NZ 14 1.33 1.33 1.33       
Spain 14 1.39 1.39 1.39 1.39      
Turkey 14 1.54 1.54 1.54 1.54      
Greece 14  1.77 1.77 1.77 1.77     
Hungary 14   1.79 1.79 1.79 1.79    
Czech Republic 14    1.97 1.97 1.97 1.97   
Austria 14     2.26 2.26 2.26 2.26  
Slovakia 14      2.38 2.38 2.38  
Germany 14       2.40 2.40  
Australia 14       2.47 2.47  
Italy 14        2.77  
Poland 14         3.59 
  
Technological Gap Ratios        
Spain 14 0.58         
Turkey 14 0.59         
NZ 14  0.76        
167 
 
University of Ghana  http://ugspace.ug.edu.gh
UK 14  0.78        
Canada 14  0.80 0.80       
Germany 14  0.80 0.80 0.80      
US 14   0.88 0.88 0.88     
Netherlands 14    0.88 0.88     
France 14     0.90 0.90    
Poland 14     0.90 0.90    
Italy 14     0.91 0.91    
Denmark 14     0.91 0.91    
Hungary 14     0.92 0.92    
Japan 14     0.92 0.92    
Norway 14     0.93 0.93    
Greece 14     0.93 0.93    
Australia 14     0.95 0.95    
Austria 14     0.95 0.95    
Czech Republic 14     0.95 0.95    
Slovakia 14      0.97    
 
Summary of Tukey HSD Results for OAPEC  
Country N Subset for alpha = 0.05 
1 2 3 4 5 6 
Meta Efficiency Scores       
Kuwait 14 1.200795      
Qatar 14 1.281469      
Bahrain 14 1.308595      
Saudi Arabia 14 1.419018 1.419018     
Algeria 14 2.497571 2.497571 2.497571    
Tunisia 13  2.689048 2.689048    
Libya 13  2.708792 2.708792    
Syria 14   3.020070 3.020070   
Egypt 14   3.624552 3.624552   
Iraq 14    4.319599  
 
Group Efficiency Scores       
Saudi Arabia 14 1.095871      
Kuwait 14 1.132239      
Bahrain 14 1.142174      
Algeria 14 1.163587      
Qatar 14 1.172159      
Tunisia 13 1.250483      
Libya 13 1.441912 1.441912     
Syria 14 1.616586 1.616586     
Egypt 14  1.929254 1.929254    
Iraq 14   2.468775    
Technological Gap Ratios       
Tunisia 13 .464449      
Algeria 14 .466356      
Egypt 14  .526895     
168 
 
University of Ghana  http://ugspace.ug.edu.gh
Libya 13  .527096     
Syria 14  .567047 .567047    
Iraq 14   .572819    
Saudi Arabia 14    .777449   
Bahrain 14     .873982  
Qatar 14      .916989 
Kuwait 14      .944147 
 
Summary of Tukey HSD Results for OPEC   
Country N Subset for alpha = 0.05 
1 2 3 4 5 6 7 
Meta Efficiency Scores       
Kuwait 14 1.2008       
Qatar 14 1.2815 1.2815      
Saudi Arabia 14 1.4190 1.4190      
Ecuador 12 1.4951 1.4951      
Angola 14 1.5313 1.5313 1.5313     
UAE 14  2.2333 2.2333 2.2333    
Algeria 14   2.4976 2.4976    
Libya 13    2.7088    
Iraq 14     4.3196   
Venezuela 14      5.4929  
Nigeria 14      5.6513  
Iran 14       8.4099 
Group Efficiency Scores        
Algeria 14 1.0824       
Saudi Arabia 14 1.0943       
Kuwait 14 1.1304       
Qatar 14 1.1929       
Angola 14 1.1953       
UAE 14 1.2031       
Ecuador 12 1.2042       
Libya 13 1.3733       
Venezuela 14  2.0420      
Iran 14   2.5250     
Iraq 14   2.7189     
Nigeria 14    3.3197    
Technological Gap Ratios        
Iran 14 0.3043       
Venezuela 14  0.3740      
Algeria 14  0.4351      
Libya 13   0.5189     
UAE 14   0.5835 0.5835    
Nigeria 14    0.5895    
Iraq 14    0.6331    
Saudi Arabia 14     0.7762   
Angola 14     0.7811   
169 
 
University of Ghana  http://ugspace.ug.edu.gh
Ecuador 12     0.8092   
Qatar 14      0.9292  
Kuwait 14      0.9427  
         
 
  
170 
 
University of Ghana  http://ugspace.ug.edu.gh
APPENDIX E 
QUARTILE ANALYSIS OF METAFRONTIER SCORES 
Meta Group 
Country IGO Efficiencies Efficiencies TGR Rank Quartile 
Slovakia IEA 2.2430 2.1672 0.9662 1 1st 
Czech Republic IEA 1.8292 1.9230 0.9512 2 1st 
Austria IEA 2.3571 2.2408 0.9506 3 1st 
Australia IEA 2.5025 2.3667 0.9457 4 1st 
Kuwait OAPEC 1.1972 1.1299 0.9438 5 1st 
Kuwait OPEC 1.1972 1.1282 0.9424 6 1st 
Greece IEA 1.8586 1.7283 0.9299 7 1st 
Qatar OPEC 1.2729 1.1826 0.9290 8 1st 
Norway IEA 1.2524 1.1629 0.9285 9 1st 
Turkmenistan FSU 1.3249 1.2247 0.9244 10 1st 
Japan IEA 1.3818 1.2702 0.9192 11 1st 
Qatar OAPEC 1.2729 1.1669 0.9167 12 1st 
Hungary IEA 1.8614 1.7059 0.9165 13 1st 
Italy IEA 3.0119 2.7236 0.9043 14 2nd 
France IEA 1.2901 1.1651 0.9031 15 2nd 
Denmark IEA 1.3178 1.1886 0.9020 16 2nd 
Poland IEA 3.9728 3.5796 0.9010 17 2nd 
Netherlands IEA 1.3123 1.1573 0.8819 18 2nd 
US IEA 1.4600 1.2827 0.8786 19 2nd 
Bahrain OAPEC 1.3046 1.1399 0.8738 20 2nd 
Ecuador OPEC 1.4769 1.1941 0.8085 21 2nd 
Germany IEA 2.9925 2.3907 0.7989 22 2nd 
Canada IEA 1.4636 1.1617 0.7938 23 2nd 
Angola OPEC 1.5269 1.1925 0.7810 24 2nd 
Saudi Arabia OAPEC 1.4135 1.0949 0.7746 25 2nd 
Saudi Arabia OPEC 1.4135 1.0934 0.7735 26 2nd 
UK IEA 1.4797 1.1413 0.7713 27 3rd 
NZ IEA 1.7235 1.2943 0.7510 28 3rd 
Iraq OPEC 4.2290 2.6755 0.6326 29 3rd 
Nigeria OPEC 5.5772 3.2812 0.5883 30 3rd 
Turkey IEA 2.5924 1.5230 0.5875 31 3rd 
Iraq OAPEC 4.2290 2.4220 0.5727 32 3rd 
Spain IEA 2.4080 1.3765 0.5716 33 3rd 
UAE OPEC 2.1083 1.1957 0.5671 34 3rd 
Syria OAPEC 2.5099 1.4184 0.5651 35 3rd 
Belarus FSU 2.0023 1.1233 0.5610 36 3rd 
Libya OAPEC 2.4402 1.2840 0.5262 37 3rd 
Egypt OAPEC 3.5822 1.8823 0.5255 38 3rd 
Libya OPEC 2.4402 1.2646 0.5182 39 3rd 
Algeria OAPEC 2.4920 1.1606 0.4657 40 4th 
Tunisia OAPEC 2.6699 1.2394 0.4642 41 4th 
171 
 
University of Ghana  http://ugspace.ug.edu.gh
Russia FSU 2.8920 1.3199 0.4564 42 4th 
Azerbaijan FSU 3.2583 1.4763 0.4531 43 4th 
Algeria OPEC 2.4920 1.0820 0.4342 44 4th 
Uzbekistan FSU 2.7007 1.1421 0.4229 45 4th 
Venezuela OPEC 5.4166 2.0236 0.3736 46 4th 
Kyrgyzstan FSU 24.6788 8.7513 0.3546 47 4th 
Ukraine FSU 4.9945 1.6374 0.3278 48 4th 
Iran OPEC 8.3536 2.5220 0.3019 49 4th 
Kazakhstan FSU 4.3545 1.3011 0.2988 50 4th 
Georgia FSU 24.7591 7.2097 0.2912 51 4th 
Tajikistan FSU 68.0373 9.6844 0.1423 52 4th 
 
 
172