Final Thesis Sobia Khurram

CORRUPTION IN THE DEVELOPING WORLD: THE CASE OF SEMI-
AUTONOMOUS REVENUE AUTHORITIES WITH SPECIAL

REFERENCE TO PAKISTAN
SOBIA KHURRAM
A Thesis Submitted in Partial Fulfilment of the Requirements for the Degree

of
DOCTOR OF PHILOSOPHY
Stirling Management School

University of Stirling
September 2016
ii
DEDICATION
This thesis is dedicated to the memories of my late father Ch. Dilshad Ahmed Khatana and my
late supervisor Professor. Rob Ball

iii
ACKNOWLEDGEMENTS
Writing a PhD thesis is a long and challenging journey. This journey would not have been
possible without the help of my family, friends and supervisors at Stirling University.
I am particularly grateful for the opportunity to work with late Professor Rob Ball, as my
supervisor, in his last years. I deeply miss him and still cannot believe that he is not around to
see me finish. I will never forget his unstinting support and guidance towards my thesis till the
very last week of his life. This has not only deeply moved me, but also kept on motivating me to
keep on going despite many hurdles, particularly in the final stages of the write up. Part of my
confidence in my work comes from the confidence and belief he had in my work.
I am also very grateful to Professor William Webster for immense support and contribution
towards my work during the last years of my PhD. After the shock of sad demise of Prof. Rob
Ball, I was immensely relieved when he took over my principal supervision for the rest of my PhD
years. I believe that his comments and detailed feedback on my work enabled me to see the
thesis from another perspective and helped me to improve it immensely. I believe that without his
support and guidance in the critical stage of PhD, this thesis would not have been completed.
I am also grateful to the staff (academic and non-academic) of the Division of Management, Work
and Organization, Stirling Management School at University of Stirling for their administrative
support. The financial support provided by the School for funding my attendance in various
doctoral conferences is gratefully acknowledged. My gratitude also goes to my Alma Mater
University of the Punjab for approving my study leave to do the PhD and for the financial support.
The administrative and financial support by University of the Punjab made my life comfortable
during this long and tiring journey of PhD. I would also like to extend my gratitude to the various
institutions visited for document collection, and the anonymous interviewees, in addition to the
facilitators who made my access to the interviewees and institutions possible during my fieldwork
in Pakistan.
I cannot thank enough the support of every kind extended by my family during my PhD years. I
would not have made it without them. Thanks are due to my family including my mother and my
siblings for their understanding, love, moral support and continuous prayers. Their belief in my
abilities always encouraged and motivated me during the tough times. How can I thank my
iv
children, Anoushey, Ushney and Abdullah for bearing with me during the PhD experience with so
much support and patience? They persevered with me all along the PhD years, even during the
testing times when our family was split because of work commitments. I tried my best to make up
for the absence of their Dad when he had to be away for work commitments. I hope the whole
experience has been worth the effort for all of us. They summed it up for me when despite all
challenges, both of my girls at tender ages of 5 and 9 expressed wishes of doing a PhD when
grown up. In the end, I simply cannot thank enough my husband, Dr. Khurram Shahzad, for
everything he has done for me in my PhD years and beyond. This one is for you and was not
possible without you.
v
ABSTRACT
SARAs have been established to achieve dual objectives of improving efficiency and controlling
corruption in tax administrations. Key question for this research include: why some SARAs have
failed to effectively control opportunities for corruption and what should be done about it? To
explore these questions, this thesis has set out an Anti-Corruption SARA Framework. The novel
contribution of this thesis lies in developing the new lens (analytical framework) which causes us
to see the topic of SARAs against corruption quite differently (due to differentiating between
motivations and opportunities for corruption). In order to analyze the validity of the framework, a
two-staged analysis of SARAs, labelled as the macro and micro analysis, was conducted for
SARAs and Federal Board of Revenue (FBR) in Pakistan. In the macro level of analysis (through
secondary literature analysis), it was found that SARAs made partial progress to control
corruption by focusing more on controlling motivations for corruption (through personnel
autonomy) and lesser focus on controlling opportunities for corruption (through effective
accountability). In the micro analysis (through semi-structured interviews and secondary
literature), it was found that FBR remained ineffective in controlling both motivations and
opportunities for corruption despite focusing more on controlling motivations for corruption
(through personnel autonomy) and lesser focus on controlling opportunities for corruption
(through effective accountability). In both macro and micro analyses, continued interference from
Ministries of Finance (MoF) was found to undermine not only effective accountability for SARAs
and FBR, but also undermining control of opportunities for corruption. Findings of both macro and
macro level of analysis resulted in three main recommendations. These findings pointed towards
a recommendation of reforming SARAs into an organizational form which is far more
disaggregated from the parent ministry, such that SARAs have no accountability link with MoFs in
the presence of other effective oversight bodies. It is recommended that SARA countries should
develop this understanding about importance of balancing both autonomy and accountability
mechanisms to be fully effective against corruption for controlling both motivations and
opportunities for corruption. Thirdly, research findings point towards a case of converting FBR
into a SARA in the form of Pakistan Revenue Authority, such that it is free from any direct
oversight by the MoF, and in contrast should be subjected to effective oversight by other
oversight bodies, such as already existing Cabinet Committee for Federal Revenue.
vi
CONTENTS
Title Page i
Dedication ii
Acknowledgements iii
Abstract v
Contents vi
List of Tables x
List of Figures xii
List of Appendices xiii
List of Acronyms xiv
CHAPTER ONE: INTRODUCTION TO THE THESIS 1
1.0 INTRODUCTION 1
1.1 AIMS AND OBJECTIVES 4
1.2 SIGNIFICANCE OF THE RESEARCH 5
1.3 STRUCTURE OF THE THESIS 6
CHAPTER TWO: LITERATURE REVIEW: SARAs AND CORRUPTION 10
2.0 INTRODUCTION 10
2.1 SEMI-AUTONOMOUS REVENUE AUTHORITIES IN DEVELOPING COUNTRIES 12
2.1.1 What is Semi- Autonomous Revenue Authority Model? 12
2.1.2 Origin of Semi-Autonomous Revenue Authorities in Developing Countries 14
2.1.3 Motivation for Semi-Autonomous Revenue Authorities in Developing Countries 17
2.1.3.1 Pursuit of Efficiency 17
2.1.3.2 Combating Corruption 18
2.1.3.3 Improvement in Taxpayer Services 19
2.2 WHAT IS CORRUPTION? 20
2.2.1 Types of Corruption 21
2.2.2 Costs of Corruption in Revenue Administration 23
2.2.3 Measuring Corruption 24
2.2.4 Approaches to Corruption Measurement 26
2.2.4.1 Non-Survey-based Measures of Corruption 26
2.2.4.2 Survey-based Measures of Corruption 27
2.3 MOTIVATIONS AND OPPORTUNITIES FOR FISCAL CORRUPTION IN REVENUE 28
ADMINISTRATION
2.3.1 Factors affecting Motivations and Opportunities for Corruption in Revenue Administration 30
2.3.1.1 Approaches to Combating Fiscal Corruption in Revenue Administration 31
vii
2.4 THEORETICAL UNDERPINNINGS OF SEMI-AUTONOMOUS REVENUE AUTHORITY MODEL 32

2.4.1 Which theory: Agencification or Central bank Model? 32
2.4.2 Interdisciplinary or Lack of Research? 35
2.4.3 Arguments for SARAs as Executive Agencies 36
2.4.4 Arguments for SARAs as More Autonomous Bodies (MABs) 38
2.4.5 Arguments for SARAs as Executive Agencies not MABs 42
2.4.6 Lack of Research on Theoretical Underpinnings 45
2.5 THEORY-PRACTICE PARADOX IN SARAs 45
2.5.1 Semi-Autonomous or Autonomous Revenue Authorities? 47
2.5.2 How much Disaggregation for SARAs? 52
2.5.2.1 SARAs for Credible Commitments 52
2.5.2.2 Delegation of Authority to Third Parties: Autonomous Bureaucrats 54
2.5.2.3 Formation of Corporate Bodies: Accountability by Taxpayers in Revenue Boards 58
2.5.3 Balancing Autonomy with Accountability for Controlling Corruption 60
2.5.3.1 Corruption equals Monopoly plus Discretion minus Accountability 61
2.5.3.2 Corruption equals Autonomy minus Accountability for SARAs 65
2.6 SUMMARY AND CONCLUSION 71
CHAPTER THREE: DEVELOPMENT OF ANALYTICAL FRAMEWORK FOR CONTROLLING 72
CORRUPTION IN SARAs
3.0 INTRODUCTION 72
3.1 AUTONOMY VS. ACCOUNTABILITY IN SARAs 75
3.1.2 Preference of Autonomy for Controlling Motivations and Accountability for Controlling Opportunities 78
for Corruption in SARAs
3.1.3 Preference of Personnel Autonomy for Controlling Motivations for Corruption in SARAs 80
3.1.4 Preference of Effective Accountability for Controlling Opportunities for Corruption in SARAs 83
3.1.5 Preference of Personnel Autonomy over Effective Accountability for Controlling Corruption in 85
SARAs
3.1.6 Why Ineffective Accountability for Controlling Corruption in SARAs? 88
CHAPTER FOUR: RESEARCH METHODS 95
4.0 INTRODUCTION 95
4.1 RESEARCH METHODS 97
4.2 METHODS FOR MACRO ANALYSIS OF SARAS 100
4.3 METHODS FOR MICRO ANALYSIS OF PAKISTAN 108
4.3.1 Interviews 109
viii
4.3.2 Document Analysis 118

CHAPTER FIVE: MACRO LEVEL OF ANALYSIS: SYSTEMATIC REVIEW OF SARAs AGAINST 123
CONTROLLING CORRUPTION
5.0 INTRODUCTION 123
5.1 PERSONNEL AUTONOMY CONTROLLING MOTIVATIONS FOR CORRUPTION IN SARAs 129
5.1.1 Personnel Autonomy Controlling Motivations for Corruption by Increases in Wages and Rewards 129
5.1.2 Personnel Autonomy Controlling Motivations for Corruption by Increases in the Probability of 136
Detection
5.1.3 Personnel Autonomy Controlling Motivations for Corruption Increases in and Stricter Enforcement 142
of Penalties for Corruption
5.1.4 Personnel Autonomy Controlling Motivations for Corruption by Instilling Ethics in Tax Officials 148
5.1.5 Synthesis of Analysis for Research Hypothesis 1 152
5.2 EFFECTIVE ACCOUNTABILITY CONTROLLING OPPORTUNITIES FOR CORRUPTION IN SARAs 154
5.2.1 Effective Accountability Controlling Opportunities for Corruption by Introducing Effective Oversight 155
Mechanisms
5.2.2 Effective Accountability Controlling Opportunities for Corruption by De-politicizing Tax Officials 160
5.2.3 Effective Accountability Controlling Opportunities for Corruption by Reducing Discretionary Powers 165
of Tax Officials
5.2.4 Synthesis of Analysis for Research Hypothesis 2 170
CHAPTER SIX: MICRO LEVEL OF ANALYSIS: CASE STUDY OF TAX ADMINISTRATION REFORMS 180
IN PAKISTAN
6.1 HISTORY AND STATUS OF TAX ADMINISTRATION REFORMS IN PAKISTAN 185
6.2 PERSONNEL AUTONOMY CONTROLLING MOTIVATIONS FOR CORRUPTION IN FBR 197
6.2.1 Personnel Autonomy Controlling Motivations for Corruption in FBR by Increasing Wages and 198
Rewards
6.2.2 Personnel Autonomy Controlling Motivations for Corruption in FBR by Increasing the Probability of 202
Detection
6.2.3 Personnel Autonomy Controlling Motivations for Corruption in FBR by Increases in and Stricter 209
Enforcement of Penalties for Corruption
6.2.4 Personnel Autonomy Controlling Motivations for Corruption in FBR by Instilling Ethics in Tax 216
Officials
6.2.5 Integration of Sub-Analyses for Research Hypothesis 1 221
6.3 INEFFECTIVENESS OF ACCOUNTABILITY TO CONTROL OPPORTUNITIES FOR CORRUPTION 223
ix
IN FBR
6.3.1 Ineffectiveness of Accountability to Control Opportunities for Corruption in FBR by Failing to 224
Introduce Effective Oversight Mechanisms
6.3.2 Ineffectiveness of Accountability to Control Opportunities for Corruption in FBR by Failing to De- 231
politicize Tax Officials
6.3.3 Ineffectiveness of Accountability to Control Opportunities for Corruption in FBR by Failing to 235
Reduce Discretionary Powers of Tax Officials
6.3.4 Integration of Sub-Analyses for Research Hypothesis 2 240
CHAPTER SEVEN: DISCUSSION AND CONCLUSION 247
7.1 SUMMARY OF THE THESIS 250
7.2 RECOMMENDATIONS FOR POSSIBLE REFORMS 252
7.2.1 Culprit Ministries of Finance? 252
7.2.2 A Case for Pakistan Revenue Authority? 255
7.2.3 Adoption of SARAs: Balancing Autonomy with Accountability 257
7.3 LIMITATIONS OF THE STUDY 258
7.4 CONTRIBUTIONS OF THE THESIS AND AVENUES FOR FURTHER RESEARCH 261
7.5 CONCLUSION 263
APPENDICES 266
BIBLIOGRAPHY 283
x
LIST OF TABLES
Table 4.1 Overview of Sub-Hypotheses along with Indicators and Interview Schedule 111
Questions for Micro Level of Analysis
Table 4.2 List of Tax Officials Interviewed 113
Table 4.3 List of Selected Secondary literature for Macro Analysis of Pakistan 121
Table 5.1: List of SARAs in Developing Countries (arranged in alphabetical order) 127
Table 5.2: Overview of Secondary Literature concerning Personnel Autonomy and Increases in 130
Wages and Rewards for SARAs
Table 5.3: Overview of Selected SARA Countries concerning Personnel Autonomy resulting in 135
Increases in Wages and Rewards
the Probability of Detection for SARAs
Increases in the Probability of Detection
and Stricter Enforcement of Penalties for Corruption in SARAs
Increases in and Stricter Enforcement of Penalties for Corruption
Table 5.8: Overview of Secondary Literature concerning Personnel Autonomy and Instilling 148
Ethics in Tax Officials in SARAs
Instilling Ethics in Tax Officials
Table 5.10: Overview of Selected SARA Countries concerning Personnel Autonomy and all 153
four preventive strategies to control Motivations for Corruption
Table 5.11: Overview of Secondary Literature concerning Effective Accountability and 155
Introduction of Oversight Mechanisms for SARAs
Table 5.12: Overview of Selected SARA Countries concerning Effective Accountability resulting 159
in Introduction of Oversight Mechanisms
Table 5.13: Overview of Secondary Literature concerning Effective Accountability and De- 161
politicization of Tax Officials for SARAs
in De-politicization of Tax Officials
Table 5.15: Overview of Secondary Literature concerning Effective Accountability and 166
Reduction of Discretionary Powers of Tax Officials for SARAs
xi
in Reduction of Discretionary Powers of Tax Officials

Table 5.17: Overview of Selected SARA Countries concerning Effective Accountability and all 171
three preventive strategies to control opportunities for Corruption
Table 5.18: Overview of Selected SARA Countries accumulating sub-analyses of the macro 176
level of SARA analysis for research hypotheses 1 and 2
Table 6.1: Overview of Pakistan’s Progress towards Revenue Improvement and Corruption 193
Control from 2001-2012
Control in Comparison to Regional Counterparts
Control in Comparison to SARA Countries in Macro level of Analysis
Table 6.4: Overview of FBR concerning Personnel Autonomy and all 4 preventive strategies to 221
control Motivations for Corruption
Table 6.5: Overview of FBR concerning Effective Accountability and all three preventive 240
strategies to control Opportunities for Corruption
Table 6.6: Overview of FBR accumulating sub-analyses of the micro level of analysis for 243
research hypotheses 1 and 2
xii
LIST OF FIGURES
Figure 1.1: Structure of the Thesis 6

Figure 2.1 Structure of Chapter 2 11
Figure 2.2 Nomenclature 51
Figure 3.2 Sign Graph Diagram for Hypothesis 1 87
Figure 3.3 Sign Graph Diagram for Hypothesis 2 91
Figure 3.4 Research Hypothesis 1 along with sub-hypotheses 93
Figure 3.5 Research Hypothesis 2 along with sub-hypotheses 93
Figure 4.1: Structure of Chapter 4 96
Figure 4.2: The Onion Model of Academic Discourse 99
Figure 4.3: Systematic Evaluation Process for Macro Level of SARA Analysis 102
Figure 4.4: Mapping the Field for Macro Level of SARA Analysis 104
Figure 5.2 Graphical Representation of Findings of Macro Level of SARA Analysis 174
Figure 6.1: Graphical Representation of Pakistan’s Progress towards Revenue Improvement 194
and Corruption Control from 2001-2012
and Corruption Control in Comparison to Regional Counterparts
and Corruption Control in Comparison to SARA Countries in Macro level of Analysis
Figure 6.4: Overview of Directorate General Intelligence & Investigation at FBR Headquarters 204
in Islamabad, along with regional Directorates
Figure 6.5: Organogram of FBR 226
Figure 7.2 Nomenclature for Solving SARAs Ineffectiveness against Corruption 253
xiii
LIST OF APPENDICES
Appendix 1 Semi-Structured Interview Schedule…………………………………………………………... 266

Appendix 2 Reward Rules of FBR Pakistan………………………………………………………………… 269
Appendix 3 Asset Declaration Form of FBR Pakistan……………………………………………………... 274
Appendix 4 Federal Board of Revenue Act 2007…………………………………………………………... 279
Appendix 5 List of Chairman of FBR from 1947-2013……………………………………………………... 281
xiv
LIST OF ACRONYMS
CBR Central Board of Revenue
DFID Department for International Development
DMG District Management Group
FBR Federal Board of Revenue
FPSC Federal Public Service Commission
FTO Federal Tax Ombudsman
GoP Government of Pakistan
HRM Human Resource Management
IIMU Internal Investigation and Monitoring Unit
IMF International Monetary Fund
IRS Inland Revenue Service
KRA Kenyan Revenue Authority
MABs More Autonomous Bodies
MoFs Ministries of Finance
NAB National Accountability Bureau
NADRA National Database and Registration Authority
NPM New Public Management
NRS National Revenue Secretariat Ghana
NTN National Tax Numbers
PAC Public Accounts Committee
QUANGOS Quasi-Autonomous Non-Governmental Organizations
Rs. Rupees (Pakistan Currency).
SARAs Semi-Autonomous Revenue Authorities
SARS South African Revenue Service

xv
SAT Mexican Tax Administration Service
SAT Superintendencia de Administración Tributaria Guatemala
SENIAT National Integrated Tax Administration Service
SIN National Intelligence Service Peru
SRO Statutory Rules and Orders
SUNAT Superintendencia Nacional de Administración Tributaria Peru
TARP Tax Administration Reform Project
TRA Tanzania Revenue Authority
URA Uganda Revenue Authority
WJP World Justice Project

1
CHAPTER ONE
INTRODUCTION TO THE THESIS
1.0 INTRODUCTION
In recent years, there has been an increasing adoption of Semi-Autonomous Revenue Authority
(SARA) reform model in developing countries to achieve dual objectives of improving efficiency
and controlling corruption in tax administrations (McCourt, 2002; Fjeldstad, 2003, 2005a, 2005b;
Kidd & Crandall, 2006; Mann, 2004; Martinez-Vazquez et al., 2004; McCarten, 2006; Taliercio,
2004; Zuleta, 2007). The focus of this research is in analysing how well SARA reform has fared in
achieving one of its objectives of corruption control in tax administration. Research showed how
SARA reform idea, conceptually designed to curb corruption, has not fared well when empirically
dealing with corruption in tax administrations in developing countries, where improved levels of
remuneration and increased managerial autonomy co-existed with simultaneous high levels of
corruption (Fjeldstad, 2003). Literature has highlighted failure of SARA reform to successfully
reform tax administrations in some developing countries with a strong custom of political
patronage, where in practice the uptake of SARA reform made problem of political patronage
worse than in the pre-reform state (Fjeldstad, 2005a). The literature highlighted many SARA
countries engaged in ‘tug-of-war’ with parent Ministries of Finance (MoFs) over control and
autonomy. This pointed that SARA reform translated into loss of considerable autonomy and
diminished patronage benefits for some parent MoFs, which in turn encouraged strong resistance
against successful implementation of SARA reform. Several other researches highlighted
deficiencies of the SARA reform model to achieve corruption control, and have argued against
consideration of SARA reform as a universal remedy for reforming tax administrations in
developing countries (Martinez-Vazquez et al., 2004; Fjeldstad, 2003, 2005a; Kidd & Crandall,
2006; Mann, 2004; McCarten, 2006).

2
Although the review of these studies succeeded in improving our understanding of the reform
model at a conceptual level, they also point towards a theory-practice paradox. These studies
highlight that experience of SARAs against corruption has not always been as expected or
explained by conceptual models’ consequences. This observation pointed to a need to closely
analyse the theory behind SARA reform, and try to find out why the literature is suggesting that
practice deviates from theory. In order to achieve this objective, this thesis will analyze prominent
SARA conceptual models and strategies to control corruption in tax administrations, not only to
throw some light on theories behind SARA reform idea, but to look for answers to the theory-
practice paradox. These SARA frameworks will be analysed to explain why experiences of SARA
adoption in developing countries deviates from conceptual prescriptions, and to identify potential
weaknesses and paradoxes in existing literature. The over-arching objective of this exercise is to
further discussion on how to better control opportunities for corruption in SARAs in developing
countries by analyzing prominent conceptual frameworks in the field.
In addition to aiming to analyse SARAs’ progress in developing countries against corruption, this
thesis also aims to analyse in detail the current state of tax administration reforms and potential
and suitability of SARA reform for Pakistan. Pakistan, which gained its independence and
emerged on world’s map in 1947, is the sixth most populous country in the world, with an
approximate population of 180 million (World Bank, 2013). Presently, less than 1% of the 180
million people pay income tax (Financial Times, 2011; Rana, 2013a). The tax machinery in
Pakistan has been characterized as inefficient and corrupt for a long time. As a consequence,
Pakistan has one of lowest tax-to-GDP ratios in the world of just above 10% in 2012, which is
lower than many poorer African countries. Recent estimates of Pakistan’s informal economy
stands at 91.4% of GDP (Kemal, 2013).The tax gap, which is the difference between sum of tax
due minus amount of tax paid on voluntary basis and in time, stood at 79% of the actual tax
3
collections in Pakistan, as quoted by one of former chairman of FBR in front of the Supreme
Court of Pakistan (Jamal, 2011). This is in comparison to the tax gap of about 9% in UK and 22%
in USA. As per World Bank (2009) the total tax evaded in Pakistan for the year 2007/08 was
around Rs. 796 billion against a total collection of Rs. 1.1 trillion.
A country of more than 180 million people only has 1.2 million registered users who file income
tax returns. In 2011, almost 3 million Pakistanis held National Tax Numbers (NTN), but less than
half (1.4 million) filed income tax returns (Khan, 2012a; Kemal, 2013). Only 118,000 entities were
registered with the sales tax system, and out of this meagre number, only about 15,000 actually
paid any tax (IMF, 2013). The FBR could identify only 768,000 people who paid income tax in
year 2011. Out of this number, only 270,000 consistently paid tax in the last three years (The
Economist, 2012). As per Federal Tax Ombudsman (FTO), Pakistan should have an average of
at least 4 million income tax returns, in comparison to Indian population, where almost 25 million
returns were filed (The News, 2013a). As per Pakistan’s National Database and Registration
Authority (NADRA) there were 2.38 millionaires in 2011 and 3.2 millionaires in 2012 in the
country who owned palatial houses in affluent areas, held luxury cars, frequently travelled
abroad, possessed arms, held multiple bank accounts, paid large utility bills, and sent children
abroad to study; but were not registered taxpayers, did not had NTNs, and did not pay any
income tax (Khan, 2012a). These dismal figures reflect the long-lasting failure of Pakistan’s
central tax collection agency, the FBR, to efficiently collect taxes, and the incapability of prior
reform efforts to enhance tax collection.

4
1.1 AIMS AND OBJECTIVES
The discussions highlighted above helped to identify two main questions of this research:
 Why some SARAs failed to effectively control opportunities for corruption?
 Why tax administration reform failed to effectively control opportunities for corruption in
FBR Pakistan?
This study is aimed at analysing corruption in tax administrations in developing countries
especially Pakistan. It maps existing SARAs to illustrate why tax administration reform failed to
effectively control opportunities for corruption. In addition, it conducts a case study to illustrate
why tax administration reform failed to control opportunities for corruption in Pakistan.
To explore these questions, this thesis will set out an Anti-Corruption SARA framework by
analyzing individual SARA design components towards controlling motivations and opportunities
for corruption in tax administrations in developing countries. The analytical framework will be
developed by answering questions such as: which SARA design components contribute towards
reducing corruption? what are the processes by which this outcome is achieved? and are certain
SARA design components better suited to curb corruption than others? This analytical framework
will be developed by utilizing two SARA frameworks including SARA design components
proposed by Taliercio (2004) and control of corruption framework for tax administrations by
Martinez-Vazquez et al. (2004). The development of this framework is aimed to contribute
towards explaining how existing SARA model can be improved not only to effectively control
corruption in tax administrations in developing countries, but also to improve revenue
performance as a consequence. In order to analyze the validity of the framework through

5
analysis of SARA cases, a two-staged analysis of SARA cases will be conducted, where the
analytical framework will be tested and revised for SARA country cases. The development of
analytical framework resulted in two research hypotheses:
Research hypothesis 1: By preferring personnel autonomy over effective accountability, SARAs
have been effective against motivations but not opportunities for corruption.
Research hypothesis 2: SARAs have been ineffective in controlling opportunities for corruption
due to ineffective accountability by MoF
1.2 SIGNIFICANCE OF THE RESEARCH
In spite of spread of SARAs in developing countries, there has been limited comparative
analytical work on their implementation against corruption. As a result there is dearth of
agreement and evidence on best practice in organizational design of SARAs. Also, the
researcher has not come across any significant research on the practical side of designing and
implementing SARA reform against controlling corruption. It is this gap that this research aims to
contribute towards filling by doing a practical work of designing an Anti-Corruption SARA
framework (Chapter 3) and applying it to SARAs (chapter 5) and Pakistan (Chapter 6). This
analysis aims to explain why certain SARA design components are more effective against
corruption than others, and will offer specific recommendations to improve SARAs’ capability
against corruption. The significance of this thesis lies in developing the new lens (analytical
framework) which causes us to see the topic of SARAs against corruption quite differently (due to
differentiating between motivations and opportunities for corruption).

6
1.3 STRUCTURE OF THE THESIS
The thesis is divided into seven main chapters including introductory chapter of the thesis,
literature reviews, analytical framework, research methods, analyses, and discussion and
conclusion. The structure of the thesis is illustrated in figure 1.1 below.
Figure 1.1: Structure of the Thesis
INTRODUCTION
2
LITERATURE REVIEW
SARAs & CORRUPTION
DEVELOPMENT OF ANALYTICAL
FRAMEWORK
4
RESEARCH METHODS
5 6
MACRO LEVEL OF ANALYSIS MICRO LEVEL OF ANALYSIS
DISCUSSION & CONCLUSION

7
The purpose of chapter 2 is to locate the thesis within existing knowledge about SARAs’ adoption
in developing countries for specifically controlling corruption in tax administrations. It is further
divided into five major sections. Section one aims to answer questions such as what are SARAs,
how they originated, what were the motivations behind their adoption. Section two aims to
provide an overview of what is corruption? This section comprises of literature on types of
corruption, costs of corruption in revenue administration and measurement of corruption, where
several survey as well as non-survey based approaches to measure corruption are discussed. In
section three, several factors affecting motivations and opportunities for corruption in revenue
administration are identified and then approaches to combat motivations and opportunities for
corruption in revenue administration are discussed. Section four elaborates in detail discussion
about theoretical underpinnings of SARAs. It highlights how present literature depicts contrasting
views about underlying theoretical assumptions for SARAs. Section five highlights a theory-
practice paradox in SARAs and discusses how to better control opportunities for corruption in
SARAs in developing countries by analysing prominent conceptual frameworks in this field.
Section six concludes the chapter with a general summary and discussion.
Drawing onto discussions in chapter two lead to development of analytical framework and two
research hypotheses in chapter 3. Both research hypotheses are formulated to lend answer to
question why some SARAs have failed to effectively control opportunities for corruption.
Research hypothesis 1 proposes in section 3.1.5 that by preferring personnel autonomy over
effective accountability, SARAs have been effective against motivations but not opportunities for
corruption. Also research hypothesis 2 proposes in section 3.1.6 that SARAs have been
ineffective in controlling opportunities for corruption due to ineffective accountability by MoF.
Section 3.2 concludes the chapter with a general summary and discussion.
8
Chapter 4 discusses research methods by explaining why multiple methods are employed to
undertake macro and micro level of analysis and sequence and timing of data collection. Section
4.2 details how systematic review is conducting for macro level of SARA analysis. Section 4.3
explains how both interview data and documentary sources are combined for conducting micro
level of analysis for Pakistan. Section 4.4 concludes the chapter.
To explore why some SARAs have failed to effectively control opportunities for corruption, a two-
staged analysis of SARAs is proposed to be conducted in chapters five and six of the thesis.
To inquire this question, a detailed case study of tax administration reforms in Pakistan will be
conducted as per analytical framework developed in this study. Two main sources of data will be
employed for the case study at the micro level of analysis: document analysis (secondary
literature) and semi-structured interviews. A detailed mapping of SARA cases is undertaken in
chapter 5 for conceptually testing the Anti-Corruption SARA framework. Chapter 5 contains
macro level of SARA analysis, where individual sub-hypotheses will be analyzed for selected
SARA countries as per analytical framework. Section 5.3 will summarize the key findings arising
out of macro level of SARA analysis and concludes. This section will also integrate the findings of
all sub-hypotheses for the two research hypotheses of this study.
Building onto the discussions raised in chapter five, chapter six aims to conceptually test the Anti-
Corruption SARA framework for FBR, by conducting a detailed case study of tax administration
reforms in Pakistan. In spite of increasing proliferation of SARAs in developing countries for
controlling corruption in tax administrations, the researcher has not come across any significant
research on the practical side of designing and implementing SARA reform against controlling
corruption. It is this gap that this research aims to contribute towards filling by doing a practical
9
work of designing an Anti-Corruption SARA framework (chapter 3) and applying it to Pakistan in
chapter six.
The final chapter seven of the thesis concludes the thesis by reviewing and discussing the
evidence provided in previous chapters. It also discusses the theoretical and empirical
contributions made by the research, in addition to putting forward three main recommendations.
This chapter was aimed at providing the introduction to the thesis. The next chapter two aims to
review literature in the field of SARAs’ adoption in developing countries for specifically controlling
corruption in tax administrations.

10
CHAPTER TWO
LITERATURE REVIEW: SARAs AND CORRUPTION
2.0 INTRODUCTION
The purpose of this chapter is to locate the thesis within existing knowledge about SARAs’
adoption in developing countries for specifically controlling corruption in tax administrations. It is
further divided into five major sections. Section one aims to answer questions such as what are
SARAs, how they originated, what were the motivations behind their adoption. Section two aims
to provide an overview of what is corruption? This section comprises of literature on types of
corruption, costs of corruption in revenue administration and measurement of corruption, where
several survey as well as non-survey based approaches to measure corruption are discussed. In
section three, several factors affecting motivations and opportunities for corruption in revenue
administration are identified and then approaches to combat motivations and opportunities for
corruption in revenue administration are discussed. Section four elaborates in detail discussion
about theoretical underpinnings of SARAs. It highlights how present literature depicts contrasting
views about underlying theoretical assumptions for SARAs. Section five highlights a theory-
practice paradox in SARAs and discusses how to better control opportunities for corruption in
SARAs in developing countries by analysing prominent conceptual frameworks in this field.
Section six concludes the chapter with a general summary and discussion.
The structure of Chapter 2 is as set out in Figure 2.1 below:

11
Figure 2.1 Structure of Chapter 2
2.0
INTRODUCTION
2.1
SEMI-AUTONOMOUS REVENUE
AUTHORITIES IN DEVELOPING
COUNTRIES
2.2
WHAT IS CORRUPTION?
2.3
MOTIVATIONS AND
OPPORTUNITIES FOR FISCAL
CORRUPTION IN REVENUE
ADMINISTRATION
2.4
THEORETICAL UNDERPINNINGS
OF SEMI-AUTONOMOUS REVENUE
AUTHORITY MODEL
2.5
THEORY-PRACTICE PARADOX IN
SARAs
2.6
SUMMARY AND CONCLUSION
12
2.1 SEMI-AUTONOMOUS REVENUE AUTHORITIES IN DEVELOPING COUNTRIES
The focus of this study is on adoption of revenue authorities experimented in several developing
countries. At least 25 developing countries can be highlighted that have embarked upon tax
administration reform in which the traditional tax departments are granted autonomy and
converted into (semi) autonomous revenue authorities (Taliercio, 2004a). The revenue authority
reform model, starting off in developing countries of Bolivia and Ghana in the late 1980s, proved
to be a success due to improvements in revenue collections, service delivery and tax policy
reforms. This initial success of revenue authority reform in some developing countries
encouraged worldwide replication of same autonomous organizational design to reform existing
revenue departments in many other developing countries. In Africa, Latin America, South and
Southeast Asia, and Europe, many developing countries converted their traditional customs and
income tax departments into unified and corporatized revenue authorities at a national scale.
Following the initial success and in part due to support by international donor agencies, the
countries joining the Semi-Autonomous Revenue Authority Reform bandwagon, rather speedily,
included (In Africa) Ghana in 1985, Uganda in 1991, Zambia in 1993, Kenya in1995, Tanzania in
1996, South Africa in 1997, Rwanda in 1998, and Malawi in 2000. In Latin America, Bolivia in
1987, Argentina in 1988, Peru in 1988, Colombia in 1991, Venezuela in 1994, Mexico in 1997,
Guatemala in 1999, and Guyana in 1999. In Southeast Asia, Singapore in 1992, and Malaysia in
1994. In Europe, Spain in 1991. (Jenkins, 1994, 1995; Jenkins & Khadka 2000; Polidano, 2001;
McCourt, 2002; Silvani & Baer 1997; Taliercio, 2004; Martinez-Vazquez et al. 2004).
2.1.1 What is Semi- Autonomous Revenue Authority Model?
In general, the Semi-Autonomous Revenue Authority (SARA) Model, as suggested by the term
‘Autonomous’ in its title, has been primarily adopted to provide more autonomy to pre-reform
revenue departments under control of Ministry of Finance (MoF), by employing autonomy-

13
boosting elements of self-funding arrangements, adoption of board of directors model with
representation of public as well as private sector, and supportive human resource management
systems. Interestingly, although many developing countries are involved, nonetheless, one can
highlight a similar pattern in reform of tax administration, since many countries have adopted
similar organizational forms. Specifically, in majority of reforming developing countries the long-
established tax departments were taken out of the ambit of Ministry of Finance, granted financial,
managerial and personnel autonomy and hence converted into an autonomous revenue
authority. ‘Semi-autonomous revenue authorities have been defined as tax administrations that
have greater than usual autonomy along several organizational design dimensions, including:
legal character, corporate governance, financing and budgeting, personnel policy, procurement
policy, and accountability relationships’ (Taliercio, 2004). The reformed revenue authorities in
developing countries has generally been referred to as semi-autonomous revenue authorities
(such as in Taliercio, 2004, 2004a) with the term ‘semi’ often attached to the Autonomous
Revenue Authority Model, since they are not created with the objective of being as autonomous
as public sector organizations such as central banks, but also not as un-autonomous and reliant
as are line departments on parent ministries, hence nicknamed ‘semi-autonomous’ (World Bank,
2002). Nevertheless, revenue authorities are distinguishable from pre-reform tax departments on
accounts of greater managerial, personnel and financial autonomy granted by the central
government. Regardless of the abundance of many SARA reform cases in many developing
countries, literature does not point out towards agreement on one best organizational design
suitable for revenue authority reform. World Bank (2002) has defined the semi-autonomous
revenue authority model in terms of its key design features including 1) legal character, 2)
governance structure, 3) financing mechanism, 4) personnel system, and 5) accountability
relationships. Taliercio (2004) has also identified certain suitable and practical organizational
design features for future autonomous revenue authority reforms in developing countries, dealing
14
with different dimensions of autonomy including legal foundations, corporate governance
mechanisms, funding mechanisms, personnel management systems, procurement and
expenditure management systems, and accountability arrangements.
2.1.2 Origin of Semi-Autonomous Revenue Authorities in Developing Countries
Literature points out that past three decades have seen many developing countries adopting
reform options to improve the functioning of their tax departments (Taliercio, 2004a). These
reform ideas have generally been based on two objectives of (1) reforming tax systems towards
collection of more efficient taxes and (2) adoption of those tax types that not only improve
revenue performance but are also possible administratively (Jenkins, 1994). In general, majority
of these developing countries have followed either of the two reform paths, where one reform
idea, majorly adopted by those countries where tax administrations were not performing too
poorly pre-reform (such as Singapore, Chile), was characterized by adoption of gradualist or
incremental approach to civil service reform, which entails adoption of a tax reform strategy
based upon a number of modest modernization techniques and generally carried out over long
term (Taliercio, 2004).These tax modernization techniques, especially suggested for developing
countries, included unified functional organization, human resource development,
computerization, taxpayer education program, self-contained tax law, effective penal system, and
efficient systems of appeal. (Jenkins & Khadka, 2000)
On the other hand, many countries with highly inefficient tax administrations preferred to uptake
more sweeping reform option of SARA model, based on the belief that it was far more easier and
faster to completely overhaul the preceding inefficient tax department, and replace it with a
reformed autonomous tax organization, such that the tax administration was taken out of the
control of the general civil service and reformed into an autonomous revenue authority, vested
15
with administrative autonomy but accountable to ministry of finance (Jenkins & Khadka, 2000).
The inception/or evolution of SARA reform model specifically in developing countries triggered
due to researchers (such as Jenkins, 1994) looking for answers to questions such as how
governments can convert corrupt tax administration into reformed professional organizations
staffed with employees working to earn competitive wages only. It has been argued that, if
properly designed and implemented, the SARA reform model can serve the purpose of rapidly
and comprehensively reforming tax administrations in developing countries (Taliercio, 2004). In
case of developing countries, drastically reforming tax administrations entrenched with
inefficiency and corruption was shown to be more plausible and easier if the organization was
made more autonomous (Silvani & Baer, 1997) and authors encouraged uptake of SARA reform
in case of developing countries (Jenkins, 1994). It was believed that the creation of a tax
administration that was perceived to be more receptive to the needs of the private sector could
be reformed faster by adopting SARA model, rather than adopting gradualistic or incremental
approach to reform traditional civil services (Jenkins & Khadka, 2000).
Major reform of tax departments into semi-autonomous revenue authorities in developing
countries was triggered due to presence of large tax gaps (defined as the difference between
potential tax that should be collected as per tax statutes and the actual tax that is being collected)
in such countries, such that Silvani and Baer (1997) suggested that the larger the tax gap in a
given country, the more sweeping reform options were needed to reform tax administration in that
country. In this study, the researchers divided the countries into four major groups based upon
the level of effectiveness (in other words the level of taxpayers’ non-compliance) at tax
administration, where effectiveness in tax administration is often measured by the level of tax
gap. It order to reform a developing country’s tax administration with a tax gap of more than 40%
(notably majority of developing countries contemplating SARA reform fall in this category), the
16
appropriate tax strategy entailed completely revamping the current inefficient tax administration to
improve tax compliance by taxpayers. This argument rested on the logic that a tax gap of more
than 40% was sufficient enough to indicate the entrenchment of inefficiencies system-wide and
any efforts to only partially reform some areas of tax organization would be insufficient to
successfully reform tax administration. Silvani and Baer (1997) also pointed out that the problems
faced by tax administrations in developing countries with a tax gap of more than 40% included
lack of financial and material resources, poorly qualified and trained staff, extremely ineffective
procedures, failure to implement measures that will reduce non-compliance, the absence of
effective taxpayer services, a high turnover of technical staff and management and corrupt
practices. Owing to these problems, the solution in the form of reform lied in completely
overhauling the old system of tax administration, by replacing outdated tax administration
procedures and practices with new ones, based on existing best practices. By analyzing the
experience of several developing countries, Silvani and Baer (1997) identified numerous guiding
principles to highlight the basis for successful tax administration reforms, such as political
commitment to and the sustainability of the reform; simplification of the tax system to facilitate
administration and reducing compliance costs; encouraging voluntary compliance; formulation of
a clear strategy; identifying the tax and accounting laws that require change; taking an integrated
approach to the tax collection process; differentiating the treatment of taxpayers by size; ensuring
the effective management of the reform process; setting priorities and establishing a timetable
and beginning fundamental reform with pilot projects. Reformers have also been cautioned that
detection of bottlenecks in following areas, which deter smooth functioning of tax administration,
is very important when designing an appropriate tax reform strategy. These areas include
taxpayer registration, returns and payment processing, computer operations, detection of stop-
filers, delinquent taxpayers, audit, sanctions and the penalty system, taxpayer services and
publicity, management and organization and personnel (Silvani and Baer, 1997).
17
2.1.3 Motivation for Semi-Autonomous Revenue Authorities in Developing Countries
Literature points out that SARA countries exhibit differences in motivations behind take up of
revenue authority reform. The major underlying objective behind revenue authority reform in
developing countries has been the obvious one i.e., improvement in revenue performance. In
addition, tax administration reform has also been taken up to achieve other objectives such as
combating corruption. Some countries (Kenya, South Africa, Uganda, Bolivia, Mexico, Peru and
Venezuela) have primarily adopted these reforms to improve bureaucratic performance while the
motivation behind revenue authority reform in other developing countries (Ghana,) was to fight
prevalent endemic corruption in tax departments.
2.1.3.1 Pursuit of Efficiency
In majority of developing countries, pre-reform tax administrations remained inefficient on
account of service-wide applicable budgetary and personnel rules and regulations. The typical
profile of developing countries undertaking SARA reforms has generally been characterized by
pervasiveness of rigid civil service rules and regulations encompassing recruitment, performance
appraisal and compensation etc. As a result of personnel and budgetary constraints, tax
departments were unsuccessful in providing market-based competitive compensation to their
employees, thus, failing to retain the competitive and skillful ones. In addition, tax departments
found it difficult to get rid of corrupt employees; and faced immense inflexibility in deciding how to
spend the allocated budget. The justification behind creation of semi-autonomous revenue
authorities was that corporatizing the existing tax administration was meant to free the traditional
tax department from the clutches of inefficient public sector rules and regulations, particularly
targeting funding and personnel issues. Majority developing countries (Kenya, South Africa,
Uganda, Bolivia, Mexico, Peru and Venezuela) have adopted these reforms to improve
bureaucratic performance in less efficient public sector by using autonomy as a tool to achieve
18
this objective under public management perspective, arguing that the cost benefit analysis
suggests that adopting autonomy is more beneficial in terms of improved organizational
performance (Taliercio, 2004).
2.1.3.2 Combating Corruption
In addition, tax administration reform has also been taken up to achieve another important
objective of controlling corruption and as a reason the motivation behind revenue authority reform
in many developing countries was to fight prevalent endemic corruption in tax departments. In
case of creation of revenue authorities in many developing countries, the objective behind
corporatization or agencification of civil service departments was not only attaining greater
efficiency, declining costs, and improvement in quality of service delivery (Minogue, 1998) but
also controlling corruption. Independent Revenue Authority Model aims to not only increase tax
revenue collections, but also reducing corruption (McCourt, 2002). The perception of prevalent
corruption in tax departments of developing countries undertaking revenue authority reform
remained very high, resulting in widespread perception of norm of negotiation of tax liability
between taxpayer and tax collector (Jenkins and Khadka, 2000). Before revenue authority
reform, these tax departments faced very low voluntary tax compliance arising out of perceived
unfairness in enforcement of tax laws (Jenkins, 1994). System-wide inefficiency and entrenched
corruption in pre-reform tax departments in developing countries served as major stimulus for
supporters of revenue authority reform to go ahead with the implementation of SARA reform
agenda. Developing countries have been suggested by Transparency International to rigorously
adopt anti-corruption reforms and starting point should be in priority area of revenue collection
and law enforcement organizations, since they are more prone to corruption by staff. These
organizations are to be converted into autonomous independent agencies with the objective of
increasing integrity and accountability in these important service delivery areas. (Pope, 1995).
19
For example, in case of Ghana, one of the main objectives behind creation of revenue service
was fight against corruption and that is the very reason why corrupt employees in the old revenue
service were not taken up after corporatization. This also enabled higher wages for remaining few
staff members to motivate them not to indulge in corrupt behaviors. (Chand & Moene 1999; de
Merode & Thomas 1994).
The reform of revenue authorities in developing countries has affected taxpayers’ perception
about prevalent corruption. In Peru, Kenya and South Africa, the reform resulted in perception of
significant reduction in corruption in the revenue authority by taxpayers, in comparison to pre-
reform corruption levels. On the other hand, taxpayers perceived still high levels of corruption in
Venezuela, Mexico and Uganda even after revenue authority reform (Taliercio, 2004). In
Venezuela, the perceived high levels of corruption by taxpayers even after reform pointed to the
relapse of the reformed revenue authority to the old ways; whereas in case of Mexico, the high
perception of existing corruption resulted due to insignificant and inconsequential effects of the
reform process on revenue authority. Although taxpayers also perceived high levels of corruption
in Uganda’s revenue authority after reform, but, nonetheless, some notable progress against
corruption was observed due to higher salaries for employees and strict punishments.
2.1.3.3 Improvement in Taxpayer Services
The motivation behind revenue authority reform in some developing countries went beyond
straight forward efficiency gains and reduction in corruption. These countries also do not exhibit
typical characteristics of countries undertaking SARA reform such as presence of large tax gap
and rigid civil service rules and regulations. The motivation behind revenue authority reform in
these countries is characterized by a need to focus on improving the quality of the taxpayers’
services and general business environment in the revenue authority. The argument is that these
20
cases initially focused on attainment of basic results of increased efficiency and reduction in
corruption and succeeded in achieving these objectives. Afterwards the focus shifted from
emphasis on quantity to emphasis on quality. For example, Singapore undertook SARA reform in
1992, although the pre-reform tax administration was considered as efficient and its tax gap was
less than 10%. Similarly, Malaysia embarked on revenue authority reform in 1994, although its
civil service was considered highly efficient even pre-reform.
2.2 WHAT IS CORRUPTION?
Although literature points out towards several definitions of ‘corruption’, both from academic or
research and policy- related stand points, but the most widely used definition of corruption
specifically pertaining to the public sector is “The abuse of public office for private gain”.
Comprehensively, according to World Bank (1997) “Public office is abused for private gain when
an official accepts, solicits, or exhorts a bribe. It is also abused when private agents actively offer
bribes to circumvent public policies and processes for competitive advantage and profit. Public
office can also be abused for personal benefit even if no bribery occurs, through patronage and
nepotism, the theft of state assets, or the diversion of state revenues”. As the focus of this
research is on corruption as pertaining to revenue authority reform, so specifically this study will
be dealing with fiscal dimension of corruption, or in other words fiscal corruption. As the term
fiscal is related to all topics encompassing taxation, public resources and spending policies, so
fiscal corruption includes all forms of corruption as related to tax administration including evasion
of taxes, corruption in tax administration and fraud in customs etc. Literature related to tax
administration differentiates between tax evasion and fiscal corruption on the basis of agent
involvement, where although tax evasion (Alm, 1998 and Andreoni, Erard and Feinstein, 1998)
might result in payment of bribes to tax officials but principally tax evasion is committed by private
21
sector agents or taxpayers; while public officials are directly involved in and responsible for fiscal
corruption.
2.2.1 Types of Corruption
There is general agreement in literature regarding detrimental effects of all types and forms of
corruption, although to varying degrees, on the governmental efforts to providing secure and
established economic structure, advancement in economic growth and improvement in general
wellbeing of citizens. Corruption exhibits itself in the shape of many types or forms and such
differentiation in corruption could be based upon several distinctions, such as unilateral or
multiparty corruption. In its most basic form, unilateral corruption in tax administration is carried
out by a single tax official by stealing revenue collections belonging to the treasury. Multiparty
corruption, on the other hand, includes formation of colluding alliances either between tax officials
and taxpayers, or between groups of tax officials partaking in corrupt practices. In case of
multiparty corruption in tax administration, Tanzi (1998) presented a very valuable list of
multiparty corruption including “(1) provision of certificates of exemption from tax to persons who
would not otherwise qualify; (2) deletion or removal of a tax payer’s records from the tax
administration’s registration, filling and accounting systems; (3) provision of confidential tax return
information to a taxpayer’s business competitors; (4) creation of multiple false taxpayers
identifications to facilitate tax fraud; (5) write-off of a tax debt without justifications; (6) closure of a
tax audit without any adjustment being made or penalties being imposed for an evaded liability;
(7)and manipulation of audit selection.”
Other differentiations in different forms of corruption could be based upon involvement of different
agents such as high or low level agents or private agents; the magnitude of the corruption such
as grand corruption or petty corruption; and the type of motivator behind corruption such as
22
insufficient incentives or institutional windows of opportunity to indulge in corruption. In addition,
government officials engage in corrupt acts for not only corruption involving direct monetary
payments, but also for other forms of corruption including understated benefits of political support
etc. In case of tax and customs administrations, corruption can take many forms and could come
up from variety of processes, including corruption entailing policy decisions such as provision of
tax incentives or the use of foreign trade taxes (Tanzi, 1998). In case of corruption, tax
administrators might not only ask for bribes to do something which they are not supposed to do,
like ignoring misrepresentation of tax liability by taxpayer, in exchange for bribe; but also
demanding “Speed or grease” money to do what they are supposed to do, anyways (Bardhan,
1997). Regardless of the fact that “speed money” type of corruption is not only morally wrong,
unlawful and pervasive in developing countries, literature points out towards perception of its
being less harmful to an economy, in comparison to other forms of corruption which involve
payment of bribes for agreeing to undertake illegal activities by tax administrators. Literature
points out towards disagreement among researchers about the negative effects exerted by speed
money type of corruption; with some researches (Leff, 1964; Hungtington, 1968; Mauro, 1995,
Mookherjee, 1997; Bardhan, 1997; and Fjeldstad and Tungodden, 2003) even pointing out
towards positive effects of this type of corruption such as “greasing the wheels of economy” and
creating incentives for tax officials to work harder in order to get more and more of speed money.
On the other hand, such an argument about the probable positive effect of “speed money” type of
corruption has been criticized by several researchers (Kaufmann, 1997; Doig and Theobald,
2000) with existing agreement that any form of corruption is generally distorting and
unproductive.
23
2.2.2 Costs of Corruption in Revenue Administration
Corruption, in all of its types and forms, imposes numerous significant costs particularly on
developing countries, and these various costs do not easily lend themselves to measurement.
The World Bank (2004b) has estimated that the total burden of bribes every year is equal to 1
trillion US$ worldwide. In majority of developing countries revenue departments are typically
ranked amongst the most corrupt public sector departments and costs of corruption in relation to
corruption in tax administration are numerous. The costs of corruption in revenue and customs
administration are not only relating to efficiency but also equity. Corruption represents one of the
main factors responsible for revenue leakage (Martinez-Vazquez et al., 2004). Corruption is also
responsible for presence and expansion of shadow or informal economy in countries where
corruption is high, which is representative of weak and depleted tax base, resulting in severe
revenues losses (Schneider and Enste 2000; Schneider, 2003; Johnson, Kaufmann, and Zoido-
Lobaton, 1998; Friedman et al., 2000; World Bank, 1997; Johnson et al., 1999; and Giles and
Caragata, 1999). Moreover, corruption in tax administration aggravates inequitable treatment of
honest taxpayers by corrupt tax officials and damages their competitiveness, and in doing so
corruption pushes businesses out of the formal sector and excluded from formal economy (Minh
Le, 2007). Literature also points out towards relative magnitude of the effect of corruption on
different types of taxes, with corruption exerting stronger negative impact on direct taxes in
comparison to indirect taxes (Tanzi and Davoodi, 2000). Corruption results in imbalances in
revenue collections from direct as well as indirect taxes, as it results in decline in revenue
collections from a group of taxpayers willing to indulge in corrupt practices and able to bribe tax
officials, and consequently increasing the relative tax liability of honest and less well-to-do
taxpayers. In case of customs administration, corruption can also result in very dangerous
outcome of threat to national security by becoming a channel for passage of drugs and arms
(Minh Le, 2007).

24
Considering the impact of corruption on tax administration in economic terms, literature points out
that increase in corruption levels affect revenues negatively with each one unit increase in level
of corruption having an economic impact of decrease in tax-to-GDP ratio by -1 to -2.9% (Ghura
and Katz , 1998) and -0.1 to -4.5% (Tanzi and Davoodi, 2000) with study encompassing 15
different types of government taxes. One area which is definitely paying very high cost for
presence of fiscal corruption is delivery of public services. One direct consequence of fiscal
corruption is reduction in availability of funds to finance public goods and services, due to
embezzlement of resources by high level politicians (Transparency International, 2004). Secondly
fiscal corruption as relating to public procurement can elevate the price of procured items and
thus increasing the cost of the public service, by 30 to 50 % (Wade, 1982; Manzetti and Drake,
1996; Langseth, Stapenhurst and Pope, 1997). Another cost of fiscal corruption can take the form
of additional user fee paid by the user in the form of bribe to actually receive the public service, in
addition to the taxes paid to sustain provision of those public services (Transparency
International, 1997). Additionally, fiscal corruption can result in reduction in quality of public
services available to users (Gould and Amaro–Reyes, 1983). Fiscal corruption can also result in
alteration in provision of public resources for health and education, with corruption resulting in
decreased ratio of health and education spending to GDP (Mauro, 1998), while increasing ratio of
military spending to GDP (Gupta, de Mello and Sharan, 2001).
2.2.3 Measuring Corruption
Given the recent emphasis on importance for developing countries (particularly corrupt ones) to
employ resources to measure existing levels of corruption, the obvious question is why corruption
should be measured? The answer lies in the fact that one needs to be aware of the problem
before trying to put in efforts to solve it. If governments are unclear about the extent of corruption
25
in the country and particularly corrupt sectors, then how can it devise an effective and targeted
anti-corruption strategy? Regardless of the fact that it is difficult to measure corruption particularly
in developing countries, governments need to invest resources to gauge actual extent of
corruption in different sectors of economy, since such measures of corruption will not only help to
gauge the effectiveness of anti-corruption policies and reforms, but will also serve to highlight key
corruption-struck areas, in order to devise anti-corruption strategies particularly targeted to those
areas.
Despite the fact that countries are increasingly realizing the importance of investing in efforts to
measure corruption, it is still very challenging to answer the question of How to measure
corruption? The biggest hurdle in the way of accurate measurement of corruption is imposed by
the unlawful or illegitimate nature of the corruption act itself. All parties including perpetrator of
fiscal corruption (tax official), partner in crime (evading taxpayer) and prey of corruption (taxpayer
facing extortion by tax official) have incentives to conceal the occurrence of corrupt act, either for
obvious reasons of not getting caught for misappropriating tax liability for bribes and tax evasion
by taxpayers or to avoid vengeance of powerful tax officials by victims of corruption. Fiscal
corruption is particularly difficult to measure because of involvement of tax officials having varying
degree of state power.
Fiscal corruption is particularly challenging and difficult to measure given the existence of many
units of measurement to gauge fiscal corruption alone in tax administration. Since fiscal
corruption is defined as “misuse of public power for private benefit”, literature points out to variety
of units of measurement gauging different dimensions of fiscal corruption including the number of
corrupt acts by different agents, the monetary cost of bribes received by the taxpayer, the worth
of resources misappropriated, the decrease in economic growth due to corruption, or increase in

26
inefficiency caused by rising level of corruption (Martinez-Vazquez et al., 2004). In order to
measure corruption comprehensively, efforts should be made to gauge all of these dimensions of
fiscal corruption, as all these units of measuring fiscal corruption are appropriate and significant
and when measured advance the present information about corruption in any given country.
2.2.4 Approaches to Corruption Measurement
Literature points out towards several measures devised to gauge corruption levels in countries
and broadly all of these measurement methods could be divided into survey-based or non-survey
based techniques to measure corruption.
2.2.4.1 Non-Survey-based Measures of Corruption
Non-survey-based measurement techniques to measure corruption include (1) prosecution
records, (2) economic indicators depicting corruption, and (3) evaluation of informal economy
(Martinez-Vazquez et al., 2004). In case of utilization of prosecution records to measure level of
corruption, data on number of court trials of corrupt cases and resulting convictions is analyzed
over a specific period of time. Literature also points out towards usage of certain economic
variables to measure levels of corruption with assumption that pattern of relationship between
certain noticeable and measurable economic variables can be utilized to estimate the extent of
out of sight and unknown levels of corruption. Some researchers have quantitatively measured
fiscal corruption by analyzing the ratio of potential tax yield to actual tax collections across
number of countries, with the assumption of evident relationship exhibited between tax revenues,
tax evasion, and corruption (Chand & Moene, 1997). Similarly, Huang and Wei (2003) have
conceptualized corruption as fraction measuring the percentage of governmental tax revenue
which “leaks” to corruption. Another economic alternative to measure corruption includes
assessment of the non-compliant economic behavior by estimating the size of informal or shadow
27
economy in any country (including unofficial household economy, unofficial productive activities,
and crime etc.) (Schneider, 2003).
2.2.4.2 Survey-based Measures of Corruption
Given the present worldwide focus of research community and international donors on
governance indicators including control of corruption, many researchers and institutions (such as
Transparency international) across the globe have developed diverse indicators of corruption
which has resulted in rapid growth of survey data sources within few years. Some researchers
have employed survey methodology to measure incidence of corruption directly by using
anonymous surveys to bring out self-disclosure of corruption by perpetrators, having obvious
disadvantage of considerable underreporting of corruption done. Another survey technique,
proving to be very popular and used extensively, does not try to directly measure occurrence of
corruption but rather surveys the perception of prevalent corruption within the country. Two most
extensive perception-based survey data sources for majority of countries across globe include
Corruption Perception Index (CPI) by Transparency International and Kaufmann’s Corruption
Index. The CPI by Transparency International represents the most widely used and referenced
index of corruption worldwide with perception surveys from businessmen, academics and risk
analysts are conducted to measure corruption levels in different countries on yearly basis. The
other widely used corruption perception index is Kaufmann’s “Control of Corruption” index with
survey data available for more than 200 hundred countries for several years (Kaufmann et al.,
2003).
28
2.3 MOTIVATIONS AND OPPORTUNITIES FOR FISCAL CORRUPTION IN REVENUE
ADMINISTRATION
All countries are faced with, though to varying extent, the problem of fiscal corruption and more
and more literature points out that controlling fiscal corruption is possible and it is not a problem
which cannot be solved. In order to devise an effective and targeted anti-corruption strategy, it is
essential to recognize the precise nature and distinct and individual patterns of corruption in
different countries and regions. Any successful anti-corruption strategy, as part of broader
institutional reform, includes a comprehensive approach to fighting corruption encompassing all
relevant areas, sustained effort and highest level of political support (Martinez-Vazquez et al.,
2004). Relevant literature on the incidence, extent and mechanisms of corruption in tax
administrations points out towards differences between two types of contributing factors to
corruption. The first set of factors influences the motivation or incentives for agents to indulge in
corruption, while the second set of factors contributes towards creation of ‘windows of
opportunity’ for agents to indulge in corruption (World Bank, 1999; Martinez-Vazquez et al., 2004;
Klitgaard, 1988, 1997; World Bank, 1999). The motivating factors to corruption influences the
motivation of tax officials to indulge in corruption given that an opportunity for corruption arises
and answer the question of what factors affect the decision by some tax officials to engage in
corruption when facing an opportunity for corruption? In addition to inclination or motivations of
tax officials to engage in corruption, the factors affecting the ‘ability’ of tax officials to indulge in
corruption represents the ‘windows of opportunity’ for corruption. A tax official might be willing
and perceive incentives and motivations to engage in corruption, but in order to engage in
corruption it is vital that an opportunity for corruption also arises. Although the clear-cut distinction
between factors creating either motivation or opportunities for corruption might not appear very
compelling owing to the significant interactions between dimensions affecting motivation and
29
opportunities in corruption decision. Such distinction is aimed at simplification of the intricate
system of interactions between two sets of factors affecting corruption.
In tax administrations, tax officials are faced with strong motivations and abundant opportunities
for corruption. The tax official makes an assessment of the prospective gains from corruption in
comparison to the estimated cost of getting caught and subsequent penalties, and only engages
in corruption if the expected gains are greater than the cost (Huther and Shah, 2000; Zuleta et
al., 2007). In turn, the potential gains and costs of corruption are determined by the governance
environment in which the tax administration functions. In order to deal with the menace of
corruption, conceptually an anti-corruption strategy could be devised focusing either on
opportunities or motivations for corruption (Zuleta et al., 2007). If a corruption control strategy is
largely focused towards controlling only opportunities for corruption, with weak control
mechanisms for motivations to corruption, then the success of this anti- corruption strategy will be
threatened by a large number of tax officials motivated to shatter the system of control on
corruption. On the other hand, an anti-corruption strategy with strong control mechanisms for
motivations or incentives for corruption, but weak focus on controlling opportunities of corruption
might prove inadequate since the plentiful and trouble-free opportunities for corruption might
even tempt those tax officials, who otherwise had no motivation or incentives to engage in
corruption. So although anti-corruption strategies focusing insistently on controlling either
motivations or opportunities might reduce existing levels of corruption, but only in the short run. A
comprehensive anti-corruption strategy for tax administrations targets not only motivations but
also opportunities paving the way for corruption, thus creating an institutional framework within
which tax officials as well as politicians make decisions about corruption (Martinez-Vazquez et
al., 2004; Pope, 1995; Rose-Ackerman, 1999). Institutionalizing anti-corruption strategy is
important as although individuals might be responsible for perpetrating corruption, but it occurs in
30
an institutional framework. Nonetheless, individuals not institutions indulge in corruption
(Kpundeh, 1997).
2.3.1 Factors affecting Motivations and Opportunities for Corruption in Revenue
Administration
The motivating factors to corruption influences the motivation of tax officials to indulge in
corruption given that an opportunity for corruption arises and answer the question of what factors
affect the decision by some tax officials to engage in corruption when facing an opportunity for
corruption? Probably, the biggest motivator for tax officials to engage in corruption is represented
by personal financial gain, but Martinez-Vazquez et al. (2004) have also identified following
motivating factors (both fiscal and non-fiscal) to engage in corruption:
i. Lack of moral and ethical behavior by tax officials;
ii. Low probabilities of detection;
iii. Weak penalization and prosecution;
iv. Inadequate wages and incentive compatible compensation;
v. The size of the window of opportunity (for instance, through pressure from taxpayers
seeking to evade taxes).
In addition to inclination or motivations of tax officials to engage in corruption, the factors affecting
the ‘ability’ of tax officials to indulge in corruption represents the ‘windows of opportunity’ for
corruption. A tax official might be willing and perceive incentives and motivations to engage in
corruption, but in order to engage in corruption it is vital that an opportunity for corruption also
arises. Martinez-Vazquez et al. (2004) theorized that as the number of opportunities to engage in
corruption increase in an institutional context of tax administration, the level of corruption also
31
increases in direct proportion and identified following factors creating opportunities for fiscal
corruption in revenue administration:
i. Absence of basic oversight and control of tax administration;
ii. Complexity of the tax and custom systems;
iii. Discretionary power of tax and customs officials;
iv. Politicization of civil servants.
2.3.1.1 Approaches to Combating Fiscal Corruption in Revenue Administration
A successful, sustainable, targeted and comprehensive anti-corruption strategy targets not only
the motivations for corruption, but also the opportunities paving the way for corruption, thus
creating an institutional framework within which tax officials as well as politicians make decisions
about corruption (Martinez-Vazquez et al., 2004). Given below are different preventive strategies
and enforcement mechanisms specifically aimed at controlling motivations and opportunities for
fiscal corruption.
Reducing Motivations for Corruption in Revenue Administration
Governments in developing countries faced with corruption problem can adopt different
preventive strategies aimed at influencing factors that motivate revenue officials to engage in
corrupt practices in tax administration. Given below are preventive strategies to control
motivations for corruption in tax administrations (identified by Martinez-Vazquez et al., 2004):
i. Instilling ethics in tax officials;
ii. Increases in the probability of detection;
iii. Increases in and stricter enforcement of penalties for corruption;
iv. Increases in wages in the public sector and the establishment of incentive compatible
compensation mechanisms;
32
v. Decreases in the overall tax burden on taxpayers
Reducing Opportunities for Corruption in Revenue Administration
In addition to the adoption of preventive strategies targeted to control the motivation for
corruption in tax administration, a sustainable and comprehensive control over fiscal corruption
requires creation of some enforcement mechanisms as well aimed at limiting the opportunities for
tax officials to indulge in corruption. These enforcement mechanisms include following strategies:
i. Introduction of oversight mechanisms;
ii. Simplification of the tax system;
iii. Reduction of discretionary powers of revenue officials;
iv. De-politicization of civil servants.
2.4 THEORETICAL UNDERPINNINGS OF SEMI-AUTONOMOUS REVENUE AUTHORITY
MODEL
In academic literature, differing statements concerning the underlying theoretical foundations of
SARA reform model can be identified. On one hand, public management researchers continue to
place SARA reform cases under discussions of agencification; and on the other hand public
policy researchers present the argument that this reform model was originally based on the
central banks model. Therefore, it seems appropriate that the present debate about the
theoretical background of SARAs should be subjected to more analysis.
2.4.1 Which theory: Agencification or Central bank Model?
Based upon discussions raised in chapter above, it became clear that reform of revenue
authorities in many developing countries was primarily aimed at improving efficiency and
reducing prevalent levels of corruption (Fjeldstad 2005; Kidd and Crandall, 2006; McCourt, 2002).
33
But review of relevant literature also pointed out towards continuing debate within organizational
theory and evident stark disagreement in current literature, whereby differing statements
concerning the underlying theoretical foundations of SARA reform model can be identified. On
one hand, public management researchers continue to place SARA reform cases under
discussions of agencification, where one can find prominent coverage of revenue authority reform
cases in literature encompassing implementation of element of agencification in the backdrop of
developing country context, with many researchers highlighting revenue authority reform cases
as success stories of agencification in developing countries (Minogue, 2001; McCourt, 2001a;
Polidano et al., 1998). While on the other hand, contrastingly, the analysis of another stream of
literature (mostly practitioner ‘s literature) relating to revenue authority reform in developing
countries, focused primarily on studying autonomy and its effect on performance and corruption,
and did not considered revenue authority reform model being based on agencification. In these
studies, researchers primarily focusing on public policy, considered SARAs to be based originally
and theoretically on central banks model (Jenkins, 1994; Jenkins and Khadka, 2000) and not
agencification.
Moreover, in addition to disagreements regarding theoretical foundation of SARA reform model,
present literature also seems unsettled on the topic of when this reform idea originated in
developing countries. According to the argument which considers SARAs as executive agencies,
the executive agency model originated in mid-1980s, when public sector reforms in developed
countries, such as ‘Next Steps’ initiative undertaken in UK in 1988 and the enclave SARA reform
model undertaken in developing countries of Bolivia and Ghana in the late 1980s, was utilized
particularly for efficiency gains (McCourt, 2002). On the other hand, Taliercio (2004) did not
considered SARA to be same as executive agencies, and also differentiated between the origins
of SARA reform model which he postulated originated in developing countries, while executive
34
agency reforms originated in developed countries such as UK and US. According to the
researcher the initiation of SARA reform and reform of executive agencies overlapped in 1980s,
with reform of revenue authorities in developing countries predating launch of executive agencies
reform in developed countries by few years. In addition to the above mentioned differentiation in
present literature on the timing of origins of both reform ideas of executive agencies and Semi-
Autonomous Revenue Authority model, and despite that reform of revenue authorities in
developing countries has been covered prominently in research literature encompassing
agencification, some researchers (Taliercio, 2004; Martinez-Vazquez et al., 2004; Jenkins, 1994;
Jenkins and Khadka, 2000) disagree with the idea of considering enclave Semi-Autonomous
Revenue Authority Reform Model as part of agencification reforms adopted in developing
countries.
In the wake of these highlighted issues and disagreement in present literature regarding
theoretical foundation of SARA reform model, it seems appropriate that the present debate about
the theoretical background of SARAs should be subjected to more discussion. This could be
done by analyzing the agency concept in detail, in order to arrive at the conclusion, if possible, of
placing SARA reform idea under true organization type. So discussion in this section is
undertaken with an objective of analyzing the theoretical foundations underlining SARA model to
identify if and to what extent this model is based upon agencification or Central Bank ideas? So
one important contribution of this research is to make effort towards contributing in discussion
towards resolving current disagreement in research literature about these two important reform
ideas. In order to achieve this objective, this section aims to analyze existing literature regarding
theoretical foundations of SARAs to identify the nature of theoretical linkages between SARA
reform model, agencification and Central Banks. Specifically in the current section of this chapter
35
generic organizational theory is contributed to by proposing that the current debates about
SARAs, Agencification and Central Banks seems to be in struggle in present literature.
2.4.2 Interdisciplinary or Lack of Research?
As highlighted above, one does find out reform of semi-autonomous revenue authorities covered
under the topic of agencification (Minogue, 2001; McCourt, 2001a; Polidano et al., 1998), but
some literature also points out towards contrasting differentiation such that the enclave SARA
Model and Executive Agency Model, whereby both models considered to be part of
agencification, have been differentiated on grounds of policy origins, where though both models
were described to be structurally similar, but having different policy origins (McCourt, 2008).
According to this argument Executive Agencies originated as a result of policy prescription for
efficiency only; whereas SARA reforms originated as a result of policy prescription for efficiency
as well as corruption control in tax administrations in developing countries. So taking into account
literature on SARAs in developing countries where one can find reform cases discussed
prominently in literatures in public management and public policy alike with theoretical linkages to
various theories of agencification and Central Banks. Does occurrence of research studies on
SARAs in both subject fields points out to the interdisciplinary nature of this reform idea? or is
literature in the area of theoretical linkages of SARAs with contending theories still unsettled on
the issue of placing these organizational forms definitively in research fields which has enabled
one group of researchers to specifically place SARAs in literature on agencification, while still
another group keep highlighting these cases strictly under public policy and influenced by central
banks. It can be argued that since literature may still be unsettled and divided on the topic of
theoretical linkage of SARAs with contending theories, this has enabled researchers working in
different subject areas to analyze SARA cases solely through their respective subject lenses, as
literature seems unclear on the issue of attributing the organizational form of SARAs to a
36
particular management field. Conversely more discussion in this area might as well highlight
SARAs to be interdisciplinary by taking theoretical inspirations for the reform model from various
theories under different subject areas. So this highlighted observation of occurrence of SARAs in
varied subject areas can be attributed to either lack of research on theoretical background of
SARAs or their interdisciplinary nature, or maybe both, and in turn highlights the need for and
significance of more research on the theoretical underpinnings/background of SARAs with an aim
to identifying and placing SARAs under theoretical boundaries. In other words, does occurrence
of SARAs across various subject areas points to its cross-cutting, inter-disciplinary nature or is it
simply lack of theoretical research in the area of placing these organizational forms definitively in
research fields enabling occurrence of SARAs in varies subject fields?
In order to answer the research questions highlighted above, following sections aim to analyze
the agency concept in detail with separate sections on executive agencies and central bank type
agencies.
2.4.3 Arguments for SARAs as Executive Agencies
The developments in the field of public management mainly in developed countries gave way to
the attractiveness of the autonomy concept in current literature. These developments have been
attributed to particular theoretical ideas which expect some effects when some tasks are put
further out from government control. NPM with fundamental elements of agencification,
autonomization and corporatisation of mostly executive tasks has been described quite
extensively by researchers (Bouckaert and Verhoest, 1999; Greve et al., 1999; OECD, 2002;
Pollitt et al., 2001; van Thiel, 2000). NPM finds its theoretical underpinnings in economic neo-
institutionalism, like agency theory (Jensen and Meckling, 1976; Moe, 1984; Pratt and
Zeckhauser, 1991). These theories conceptualize the autonomy concept as a sort of

37
specialization capable of resulting in benefits of better performance, with improvements in
economy, efficiency and effectiveness, with a condition of presence of incentives motivating
agencies to continue to improve performance.
According to the literature which considers SARAs as executive agencies, executive agencies
have been defined as organizational forms which ‘act within the executive’, are designed to work
at arms-length from parent ministries, are entrusted with the task of policy implementation only,
and are granted greater levels of managerial, financial and organizational autonomy to achieve
efficiency in service delivery (World Bank, 2000). A practitioner’s ideal type of agency is a tripod
model of functionally disaggregated, managerially autonomized, and performance oriented
organization (Talbot, 2004). As defined by Pollitt et al., (2004) agencies are supposed to be only
semi-autonomous and hence not fully independent of ministerial control which means that parent
ministry remains in charge of agencies and, if needed, can change operational objectives and/or
budgets of agencies without the need of introducing new legislation. So agencies do enjoy more
autonomy (along varying dimensions) but remain under ministerial control. Literature has pointed
out towards several prominent cases of agencification including UK Next Steps program, Latin
American Social Fund bodies, and Revenue Authorities in developing countries etc.
Both SARAs and executive agencies, as defined in existing literature, display many design
similarities. Similar to the agency concept, ‘SARAs have been defined as tax administrations that
have greater than usual autonomy along several organizational design dimensions, including
legal character, corporate governance, financing and budgeting, personnel policy, procurement
policy, and accountability relationships’ (Taliercio, 2004). This definition of SARA depicts the
adoption of characteristic feature of autonomization for SARAs, similar to agencies highlighted
above. The reformed revenue authorities in developing countries has generally been referred to
38
as semi-autonomous revenue authorities (such as in Taliercio, 2004, 2004a) with the term ‘semi’
often attached to the autonomous revenue authority model, since they are not created with the
objective of being as autonomous as public sector organizations such as central banks, but also
not as un-autonomous and reliant as are line departments on parent ministries, hence nicknamed
‘semi-autonomous’ (World Bank, 2002). Nevertheless, revenue authorities are distinguishable
from pre-reform tax departments on accounts of greater managerial, personnel and financial
autonomy granted by the central government. So based upon above definitions SARAs are
defined be similar to executives agencies. Due to the semi-autonomous status SARAs, same as
agencies, are assigned with autonomy features which are not available to departments still
working under parent ministries but are not fully independent of/and are subjected to ministerial
control.
2.4.4 Arguments for SARAs as More Autonomous Bodies (MABs)
There is a major chunk in reviewed literature according to which first of the series of SARAs in
developing countries have long considered to be originally based upon central banks model and
not agencification. According to Jenkins (1994) revenue authorities have been established in
many countries by plucking them out of the ambit of Ministry of Finance, and bestowing them with
similar type of autonomy as given to central banks, such as exemption from government-wide
civil service regulations and compensation etc. It has been argued that by design SARAs can
work just like central banks and Jenkins (1994) even proposed the reform of restructuring of a
revenue authority, taken up in any given developing country, by centering the model on the
organizational form of a central bank, such that the reformed revenue authority may be granted
financial autonomy (with features of budget formulation autonomy and automatic retention of
already approved percentage of funds), administrative autonomy (with independence to devise its
own administrative policies), and resulting out of these two dimensions of autonomy personnel
39
autonomy (in areas of recruitment, compensation, training, and code of conduct). Taking the form
of an autonomous organization, central banks not only implement monetary policy but also help
governments devise monetary policy in the first place. In the same vein, Jenkins and Khadka
(2000) argued that taking cues from the functioning of central banks, officials at SARAs have
been entrusted not only with a task of implementing tax policy by tax administration, but also aid
in development of tax policy by developing tax policy proposals. Such emphasis on taking
inspiration from central banks model even led to recruitment of managers of central banks in
newly reformed revenue authorities in many developing countries to manage and implement tax
administration reform (Silvani & Baer, 1997). In case of Peru, the main reform team responsible
for successfully taking off the SARA reform did not included any experts on tax administration but
was fully backed by the Central Bank (World Bank, 2001).
Based upon above mentioned discussion which considered SARAs to be originally based upon
central banks model, these studies theoretically placed SARAs as part of autonomization under
the category of ‘More Autonomous Bodies’ or as put by Pollitt et al., (2004) MABs or ‘type 2
agencies’ or even ‘non-agencies’. As depicted by the variety of titles or names given to this
organizational form, there is no explicit agreement in reviewed literature on the consensus of
uniformly naming this organizational form and the main difference between agencies and MABs
(such as central banks) is the level of autonomy which is granted by parent ministries to both
organizational forms with agencies granted with far lesser autonomy and operating under more
ministerial control then MABs. Contrary to agency concept, non-agency category of MABs enjoy
statutory independence and hence fully independent of ministerial control which means that
parent ministries are no longer in charge of MABs and cannot change operational objectives
and/or budgets of MABs without the need of introducing new legislation The basic difference
between agencies and MABs is the high level of independence which is accorded to MABs,
40
leaving very small or no scope for ministries to change their operational objectives and/or
budgets. In other words independence accorded to non-agency category of MABs is too great to
consider them same as agencies. Also as perceived by World Bank (2000) MABs or as they call
them ‘Type 2 Agencies’ ‘act to restrain the executive’ and are designed mainly to protect
organizational functions which are prone to short term orientations of the politicians (such as long
term policy making, quasi-judicial/regulatory functions etc.) and resulting capture by
governments, by bestowing these organizational forms with high levels of autonomy resulting in
statutory independence. Prominent types of MABs include statutory commissions or autonomous
regulators such as independent central banks (in-charge of monetary policy) and supreme audit
institutions such as anti-corruption commissions etc.
Theoretically SARAs in developing countries have been explained to be based upon the
insulation argument in public administration as for central banks. As per insulation argument
economic development could be aided by insulating public administration from politics (Evans,
1995; Grindle, 1996) and this argument finds its very popular practical applicability in creation of
independent central banks (Cukierman, Webb, & Neyapti, 1992). Taking cues from arguments for
central banks, SARAs have been designed to insulate public administration from politics to
varying extent, by making use of different dimensions of autonomy delegated to public managers
to give them choice and independence to manage internal organizational arrangements without
tight political control. MABs or ‘type 2 agencies’ such as central banks follow the idea of handing-
over of regulatory capabilities to organizational forms which are shielded from political control and
are not democratically accountable, and this central institutional feature has been described as a
prominent one for the developing regulatory state (Gilardi, 2002), and can be found prominently
in domain of monetary policy for central banks. At the theoretical level the basic premise behind
the insulation or separation argument for organizations such as central banks is described by the
41
problem faced by policy makers of temporal inconsistency of policies and resulting need of
governments to extend credible commitments to actors affected by these policies to gain trust for
policy implementation. In case of central banks, by sufficiently insulating them from political
control due to independence accorded, the governments try to signal out credible commitments
to central bank reform and the resulting diminished, if not eliminated, influence of politicians in the
working of the central banks. Shifting decision-making competencies towards institutions which
are not democratically accountable is expected to improve the credibility of government policies
(Tatcher and Sweet, 2002).
Literature points out to studies which have utilized similar arguments including insulation from
political control, temporal inconsistency of policies and credible commitments to reform, utilized
above to theoretically explain central bank reform, to theoretically describe reform of SARAs in
developing countries as well. Taliercio (2004a) studied why politicians in diverse developing
countries have been lending support to creation of SARAs by granting them autonomy? The
study showed that SARAs are instruments employed by politicians in developing countries to
endorse credible political commitment to reforms. The motive for politicians to support SARA
creation lies in the possible increased tax compliance by taxpayers due to positive perceptions
about tax reform, resulting in increased tax revenues available to politicians. Researcher argued
that politicians support creation of SARAs to indicate credible commitment to taxpayers from their
side, maintaining that such newly created SARAs will result in more competent, effective and fair
tax administration. The level of autonomy granted to SARAs enables the politicians to make
his/her commitment credible in the perception of taxpayers, with the prospect of resulting greater
tax compliance by taxpayers and thus increased level of tax revenues available to politicians.
In summary, all literature on SARAs explaining these revenue authorities to be based upon
central banks (such as Jenkins, 1994, Jenkins and Khadka, 2000) and considering them as
42
instruments used by politicians to show credible commitments towards SARA reform (such as
Taliercio, 2004, 2004a) consider them to fall under the category of MABs under autonomization
and not executive agencies.
2.4.5 Arguments for SARAs as Executive Agencies not MABs
As highlighted by the discussion in previous section and as per Pollitt et al., (2004) the basic
difference between agencies and (further out from ministries) MABs is the level of autonomy
granted by parent ministries to both organizational forms with agencies enjoying more autonomy
but still under ministerial control, while MABs enjoying far more autonomy from ministerial control
and having far more independence then agencies. Since SARAs designed as being semi-
autonomous revenue authorities are indeed granted with more levels of autonomy along several
dimensions in comparison to autonomy accorded to departments still working under ministerial
control, but nonetheless SARAs are supposed to be only semi-autonomous and hence not fully
independent of ministerial control which means that parent Ministries of Finance remain in charge
of SARAs. In spite of formal disaggregation from government hierarchy, parent Ministries of
Finance remain capable of changing operational objectives and/or budgets of SARAs without the
need of introducing new legislation. So as per this argument independence or autonomy
accorded to non-agency category of MABs is too great, resulting in very small or no scope for
parent ministries to change their operational objectives and/or budgets, and to consider them
same as SARAs and hence as per argument of intensity of ministerial control SARAs seems
more as agencies then MABs.
Also agencification model is about granting semi-autonomous status to agencies created under it,
since these executive agencies are not created to be as autonomous as MABs such as central
banks etc., and do not perform policy making functions, but are only policy implementation
43
bodies. So agencies created are supposed to be only semi-autonomous and not fully
autonomous and executive agencies, working as arms-length instruments of the parent
ministries, have increased level of autonomy but not as much as for MABs. Also in literature
SARAs in developing countries has generally been referred to as semi-autonomous revenue
authorities since they were not created with the objective of being fully autonomous as some
other public sector organizations. Taliercio (2004) also highlighted that the extent and nature of
autonomy granted to SARAs is much different from the kind of the autonomy given to Central
Banks. So in this case, in terms of level of autonomy granted, it is seen that SARAs are more
related to executive agencies due to semi-autonomous status then MABs.
Also as highlighted by the discussion World Bank (2000) postulated that the basic objective
behind creation of executive agencies is to ‘act within the executive’ with greater levels of
managerial, financial and organizational autonomy to achieve efficiency in service delivery, while
the objective behind creation of MABs is to ‘act to restrain the executive’ with protection from
short term orientations of the politicians resulting due to statutory independence. As clearly
evident from the already highlighted discussion, SARAs clearly fall into the former organizational
category of executive agencies since these have been created in developing countries
particularly to remain acting within the executive with greater levels of managerial, financial and
organizational autonomy to achieve efficiency in service delivery. So again as per this argument
of ‘to continue to act within executive’ SARAs seems more as agencies then MABs.
Also a practitioner’s ideal type of agency is a tripod model of functionally disaggregated,
managerially autonomized, performance oriented agencies with three component dimensions
including disaggregation, autonomization and contractualization (Talbot, 2004). According to this
model, the ideal type of agency is an organization which has certainly been detached from the
44
parent ministry (disaggregation); where it has some level of independence over internal
organizational arrangements, relating to managerial, personnel, financial domains etc.
(autonomization); and where it is obligated to enter into some sort of contractual or quasi-
contractual association, also with requirement of performance reporting (contractualization).
When SARAs are analyzed through the lens of this model of agencification, then they seem to
fulfill all three defining doctrinal components of agencification. SARAs as adopted in developing
countries are clearly detached from the parent Ministries of Finance (fulfilling disaggregation
dimension); where SARAs (as per Taliercio, 2004) have been granted with greater levels of
autonomy along several organizational design dimensions of legal character, corporate
governance, financing and budgeting, personnel policy, procurement policy, and accountability
relationships (fulfilling autonomization dimension); and SARAs have also been designed to enter
into contractual arrangements with parent Ministries of Finance with agreed output of a certain
annual revenue target to be achieved, along with requirement of performance reporting (fulfilling
contractualization dimension). As per this application of SARAs to the tripod model of
agencification with fulfillment of all three defining criteria of this model, SARAs seem more related
to executive agencies due to disaggregation, autonomization and contractualization then MABs.
All of the above mentioned discussion served to highlight that as per analysis of prominent
definitions and classifications of the agencification concept in current literature (World Bank,
2000; Pollitt et al., 2004; Talbot, 2004) and subsequent application to the SARA concept, the
balance is more tilted toward SARAs being more like executive agencies then their resemblance
to MABs. In summary, since SARAs are entrusted with only semi-autonomous status, since they
‘act within executive’ by operating under ministerial control and display all three dimensions of
disaggregation, autonomization and contractualization; and since all these characteristics are the
45
hallmark of the agencification concept, so theoretically SARAs have been conceived to work like
agencies and are unlike MABs.
2.4.6 Lack of Research on Theoretical Underpinnings
Also returning to the question inquiring does occurrence of research studies on SARAs across
various subjects fields points out to its cross-cutting, inter-disciplinary nature or is it simply lack of
theoretical research in the area of placing these organizational forms definitively in research
fields which has enabled occurrence of SARAs across varies subject fields? Continuing with the
discussion highlighted above, it follows that reviewed literature on SARAs highlighted lack of
research on the theoretical underpinnings of SARAs. This in turn enabled researchers to either
focus more on SARAs as MABs and hence contributed specifically to literature on public policy;
while other researchers focused more on examining SARAs as executive agencies and hence
their contribution to literature was more in the field of public management. It can be argued that
since literature still seems unsettled and divided on the topic of linkage of SARAs with contending
theories, this has enabled researchers working in different subject areas to analyze SARAs solely
through their respective subject lenses, and thus resulting in literature on SARAs across varying
subjects. One of the contributions of this discussion is to advance theory in the field of SARAs
and to highlight the agency nature of SARA model with a prospect of finding their more definitive
place in existing literature.
2.5 THEORY-PRACTICE PARADOX IN SARAs
SARAs have been established to achieve dual objectives of improving efficiency and controlling
corruption in tax administrations (McCourt, 2002; Fjeldstad, 2003, 2005a, 2005b; Kidd &
Crandall, 2006; Mann, 2004; Martinez-Vazquez et al., 2004; McCarten, 2006; Taliercio, 2004;
Zuleta, 2007). The focus of this research is in analysing how well SARA reform has fared in
46
achieving one of its objectives of corruption control in tax administration. So, the literature review
highlights numerous studies which sought to understand this aspect of the reform model. For
example, specifically in relation to dealing with corruption, Fjeldstad (2003) showed how this
reform idea, conceptually designed to curb corruption, has not fared well when empirically
dealing with corruption in SARA in Tanzania. He showed that even after SARA adoption,
improved levels of remuneration and increased managerial autonomy co-existed with
simultaneous high levels of corruption. Also the effect of SARA reform on controlling corruption in
developing countries including Peru, Guatemala and Tanzania highlighted how this reform model
has been unable to sustainability control corruption in tax administrations (Mann, 2004). Although
SARA reform initially resulted in reducing some corruption, it got worse over time and continued
to rise in some cases. In the same vein, several other researches have also highlighted
deficiencies of the SARA reform model to achieve corruption control, and have argued against
developing countries (Martinez-Vazquez et al., 2004; Fjeldstad, 2005a; Kidd & Crandall, 2006;
McCarten, 2006).
Although the review of these studies succeeded in improving our understanding of the reform
model at a conceptual level, they also point towards a theory-practice paradox. These studies
highlight that experience of SARAs against corruption has not always been as expected or
explained by conceptual models’ consequences. This observation points to a need to closely
analyse the theory behind SARA reform and try to find out why the literature is suggesting that
practice deviates from theory. In order to achieve this objective, the following sections of this
chapter aim to analyse some prominent SARA conceptual models developed in literature, not
only to throw some light on theories behind SARA reform idea, but to look for answers to the
theory-practice paradox. These SARA frameworks will be analysed to explain why experiences of
47
SARA adoption in developing countries deviates from conceptual prescriptions, and to identify
potential weaknesses and paradoxes in existing literature.
2.5.1 Semi-Autonomous or Autonomous Revenue Authorities?
Literature has highlighted the failure of SARA reform to successfully reform tax administrations in
some developing countries with a strong custom of political patronage, where in practice the
uptake of SARA reform has made the problem of political patronage worse than in the pre-reform
state (Fjeldstad, 2005a). The literature highlighted many SARA countries engaged in ‘tug-of-war’
with parent MoFs over control and autonomy. This pointed out that SARA reform translated into
loss of considerable autonomy and diminished patronage benefits for some parent MoFs, which
in turn encouraged strong resistance against successful implementation of SARA reform. In
many instances centralized MoFs were even able to over-turn the reform process and win back
autonomy and patronage benefits, as has happened in case of Tanzania and Uganda (Clark &
Wood, 2001). The above mentioned observation highlights a probable weakness of SARA reform
in achieving one of its intended objectives of controlling corruption. This weakness is proposed to
arise due to widespread political patronage pressures in tax administrations in developing
countries, when the revenue authority is disaggregated and placed further away from direct span
of political control. This has even urged some researchers to argue against take-up of SARA
reform to control widespread political patronage problems (Jenkins & Khadka, 2000). These
studies have argued that adoption of SARA reform in the wake of existing political patronage
traditions might make the situation worse and the inability of SARA reform to effectively fend off
political patronage pressures lies in its semi-autonomous status. The proposed solution of
granting even more autonomy to revenue authorities to resist political patronage is similar to the
one granted to central banks to enable them to function autonomously without political meddling
(Jenkins & Khadka, 2000).

48
This discussion raises a very interesting scenario. The literature suggests that one of the reasons
why SARAs have not performed well against its intended objective of controlling corruption is due
to its inability to effectively deal with patronage pressures. This might be due to its ‘Semi-
Autonomous’ status resulting in insufficient autonomy. Also, the proposed solution lies in making
revenue authorities more autonomous, similar to central banks. On the other hand, as discussed
in previous section, SARAs have been conceptualized to be based upon the organizational form
of executive agencies, which in contrast to More Autonomous Bodies (MABs) or central banks,
are semi-autonomous. So, despite consensus reached in this study and in literature (Polidano et
al., 1999a; McCourt, 2001a; Minogue, 2001; Polidano, 2001) that by design SARAs have been
conceptualized more like executive agencies than MABs, with only semi-autonomy, the failure of
SARA reform in literature, especially against or leading to corruption, has urged some
researchers to question the suitable organizational form of SARAs to control corruption. They
have questioned the appropriate levels of autonomy granted to SARAs, and to analyse if the
solution to SARAs implementation problem against corruption and patronage lies in changing the
autonomy levels of SARAs.
This observation it is argued can be tackled in either of the two ways described below. Either the
solution to SARAs performing poorly against corruption in tax administrations in developing
countries lies in reforming them into MABs by increasing autonomy and disaggregating further
away from parent MoFs, or conversely, it is more suitable to keep SARAs semi-autonomous on
the lines of executive agencies.
The review of mostly practitioner’s literature on modern agencies highlighted a tripod of doctrines
for agencification with three component dimensions including disaggregation, autonomization and
49
contractualization (Talbot, 2004). Due to the collective mechanics of all three doctrinal
components, theoretically, agency is controllable enough arising due to contractual mechanisms
in place (contractualization), yet independent enough to manage itself for efficient service
delivery (autonomization), and also devolved enough to pay attention towards attainment of main
tasks (disaggregation). It is argued here, that it might be the improper or incomplete
implementation of the agencification elements that has led to failure of SARA reform in some
developing countries, especially when dealing with patronage pressures. A lot of literature exists
in case of agencification in developing countries, where reform ideas appearing very good on
paper and theory, fail at the implementation stage of reform (Pollitt & Talbot, 2004). This failure
has been attributed to several factors including different contextual factors undermining uniform
reform implementation patterns for developing countries (Dunleavy & Hood, 1994; Polidano,
2001; Pollitt, 1995; Shah, 2007).This is in line with McCourt (2001c) who highlights the need for
uptake of a ‘contingency approach’ when it comes to choice of appropriate reform model for
SARAs in developing countries. He highlighted varied experiences with enclave ‘revenue
authority model’ in experimented countries, where corruption reduced in one country, but almost
identical model failed to do so in another country. So same SARA model, when applied in
different contexts, can lead to varied even surprising consequences.
Also, the experience with agencification has highlighted that in practice the tripod model of
doctrines of agencification has on many occasions failed to adopt all three concepts of
disaggregation, autonomization and contractualization in prescribed proportions, with either or
more of the elements not implemented properly, leading to failure in successfully and completely
implementing agencification (Pollitt et al., 2004). Continuing with this argument, it can be
postulated that the reason behind SARAs not performing well against corruption in developing
countries, which is blamed to insufficient autonomy, might be attributed to failure of proper

50
adoption of autonomization dimension of the tripod model of agencification. This is proposed to
have resulted in absence of adoption of even prescribed levels of ‘semi-autonomy’. This in turn is
proposed to have resulted in false perception of autonomy levels of SARAs being insufficient for
controlling corruption, although the case being autonomy (though only semi) not being
implemented properly. Also, calls for even higher levels of autonomy for SARAs have been
considered unsuitable and even dangerous for developing countries. This is because highly
autonomous SARAs become hub of specialist knowledge, and may lead to the seizure of the
government or parent ministry, where the SARA starts to direct and manage the ministry (Pollitt
et al., 2004).
So based upon discussion highlighted above, SARAs might have been suitably designed as
similar to executive agencies with only ‘semi-autonomous’ status. There might be nothing wrong
with the semi-autonomous status of SARAs, and its patchy implementation record could be
blamed to less than desired motivation and poor adaptability of SARA reform to their peculiar
contexts by developing countries. In such a case, the appropriate position of SARAs on the
autonomy continuum would be with agencies at box 1as illustrated in Figure 2.2 below. This
figure illustrates that the basic difference between agencies and (further out from ministries)
MABs is the level of autonomy granted by parent ministries to both organizational forms. This
figure illustrates that agencies enjoy more autonomy, in comparison to departments working
under ministerial control, hence the gap between ministry and agency in box 1. But in comparison
to MABs, agencies remain under ministerial control and hence are put closer to ministerial control
on the autonomy continuum in box 1, then MABs in box 2.

51
Figure 2.2 Nomenclature
‘Closer to’ (political control) ‘Further out’ (from political control)
More Autonomous
Bodies (MABs) e.g.
Central Banks
Ministries Agencies
Semi-autonomous
State Enterprises
and other
Corporate
Commercial
Organizations
Source: (Adapted from Pollitt et al., 2004)
On the other hand, a contrasting position to theory-practice paradox might lie in asserting that
semi-autonomous status and resulting insufficient levels of autonomy for SARAs are the problem
which renders them ineffective against corruption. So, the solution for SARAs to deal with
patronage and corruption problems lies in improving its autonomy on the lines of MABs and
placing them further away from span of ministerial control. In such a case, the appropriate
position of SARAs on the autonomy continuum would be with MABs at box 2 as illustrated in
Figure 2.2 above. This figure shows that MABs enjoy complete autonomy from ministerial control
and have far more independence then agencies. In terms of level of autonomy granted, MABs
are put further away from ministerial control, then agencies, on the autonomy continuum in box 2.
52
2.5.2 How much Disaggregation for SARAs?
The previous section questioned appropriate levels of two doctrinal components of agencification
including autonomization and disaggregation of SARAs to effectively control corruption. While
this section specifically focuses on analysing appropriate level of disaggregation of SARAs from
parent MoFs, which might enable them to effectively control corruption in tax administrations. It is
stressed here that development of both these sections takes discussion about appropriate level
of disaggregation of SARAs from parent MoFs further, with previous section questioning if
SARAs might be more effective against corruption if they have more disaggregation from parent
MoFs. While this section takes this discussion forward by proposing if effectiveness of SARAs
against corruption might be improved by completely disaggregating them from any accountability
link with parent MoFs.
It is questioned in this section whether not-enough disaggregation from MoFs’ span of control
may lead to ineffectiveness of SARAs in controlling corruption in tax administrations in
developing countries? In order to achieve this objective, this section of the thesis aims to analyse
SARA conceptual model proposed by Taliercio (2004a), not only to throw some light on theories
behind the SARA reform idea, but to look for answers to the question set out above. This model
represents one of the most prominent conceptual models developed in literature to improve our
understanding of the SARA reform model in developing countries. Specifically, Taliercio’s SARA
framework will be analysed to identify potential weaknesses and paradoxes against controlling
corruption.
2.5.2.1 SARAs for Credible Commitments
Taliercio (2004a) studied why politicians in diverse developing countries have been lending
support to creation of SARAs by granting them autonomy? He argued in his theoretical

53
framework that the instrument of SARAs, as adopted in many developing countries, is aimed to
extend credible commitments to reform targeted towards taxpayers by politicians. The motive for
politicians to make credible commitments to support SARA creation lies in possible increased tax
compliance by taxpayers, arising due to positive perceptions about tax reform, eventually
resulting in increased tax revenues available to politicians. Taliercio argue that politicians support
the creation of SARAs to indicate credible commitment to taxpayers from their side, maintaining
that such newly created SARAs will result in more competent, effective and fair tax
administrations. The level of autonomy granted to SARAs enables the politicians to make his/her
commitment credible in the minds of taxpayers, with the prospect of greater tax compliance by
taxpayers and thus increased tax revenues available to politicians. Taliercio also highlighted that
the politicians needed to make his/her commitment credible in the minds of taxpayers, owing to
problem of time consistency. Time consistency problem can lead to certain policies being
perceived as inconsistent by those affected by the policies, believing that politicians have too
much discretion and will have motivations in the future to back off from policy intentions stated in
the past. Taliercio (2004a) further proposed a solution for politicians which they can utilize to
change taxpayers’ perception about time consistency problem and to make politicians’
commitments credible, arising primarily due to taxpayers existing perception of too much
discretion accorded to politicians. He proposed that politicians supporting the creation of SARAs
can make use of concept of ‘commitment technology’ by adopting three such technologies of (1)
formation of corporate bodies, (2) delegation of authority to third parties, and (3) reputation
(Persson & Tabellini, 1994).
Also, while further elaborating these commitment technologies for use by politicians, Taliercio
(2004a, p. 216) states:
“By delegating authority to bureaucrats and giving taxpayers an oversight role, the effect
of the SARA reform is to insulate, to some extent, tax administration from politicians”.
54
As per Taliercio the objective of insulation from politicians is built into the institutional design of
newly created SARAs, firstly by delegating authority (previously held by politicians) to suitable
third parties. These third parties are expected to be autonomous professional bureaucrats
working for earning competitive wages and rewards for better performance. Secondly politicians
influence in SARAs working is insulated by giving taxpayers an oversight role through adoption of
instrument of Revenue Boards.
2.5.2.2 Delegation of Authority to Third Parties: Autonomous Bureaucrats
Taliercio (2004a) proposed the first commitment technology which can be employed by politicians
to make credible commitment to SARA reform is related to delegation of authority to third parties,
such as bureaucrats, delegated with the responsibility of making tax administration more efficient,
effective, and fair. Taking the form of managerial autonomy, power is delegated to bureaucrats,
to send a signal to taxpayers of diminished influence of politicians in SARAs’ operations. This
leads to taxpayers developing perceptions that these empowered bureaucrats are professionals
and have incentives to make fair choices for SARAs, thus improving taxpayers perceived fairness
levels of SARAs. Literature points out to several developing countries, such as Zambia, Kenya,
Bolivia, Argentina and Ghana, where managerial autonomy has been employed to convince
taxpayers of professional functioning of tax administrations (Jenkins, 1994). Taliercio (2004a)
further highlights that the success of the third-party delegation solution depends upon important
pre-conditions of suitable selection of third party, such as selection of credible professional
managers, and providing them with incentives to continually improve performance. By utilizing
the instrument of delegation, politicians try to signal to taxpayers that SARAs have been insulated
from political interference and resulting politicization of civil servants.

55
But what if taxpayers don’t believe that SARAs have been insulated from political interference
and resulting politicization of civil servants, due to not-enough disaggregation from previous
political principals in MoFs, with politicians in MoFs still retaining accountability responsibility?
Taliercio (2004) proposes a hierarchical multi-layered accountability relationship between SARAs
and several oversight bodies including Ministries of Finance, in addition to Revenue Boards, the
Comptroller/Auditor General, and Parliament. What if taxpayers don’t believe in the credibility and
suitable selection of bureaucrats, who are chosen to be empowered under the third party
delegation solution, arising due to persistent accountability link with previous political principals in
MoFs? What if taxpayers believe that bureaucrats who are given autonomy might still potentially
be hand-picked by previous political principals in MoFs, because politicians might have
transferred their autonomy to bureaucrats, but continue to retain accountability authority? It is
proposed here that as politicians in MoFs, whose influence was supposed to diminish due to
transfer of authority to bureaucrats, remain part of accountability equation of SARAs, this may
hinder development of taxpayers’ perception of meaningful insulation from political interference.
The basic concern here is why retain some oversight because this might counteract development
of perception of reduced influence of politicians in SARAs. This is particularly applicable when
SARAs are employed to reduce corruption in tax administrations which is occurring due to
politicization of tax officials by politicians. If the objective behind SARA reform is specifically to
control corruption, in addition to improvement in revenue collection, and this is sought by
delegating more authority to bureaucrats to diminish political interference, then why continue with
same past accountability relationship with parent MoFs which gave rise to politicization in the first
place? It is proposed that in such cases efforts to break corrupt alliance between two parties by
only taking away and delegating authority from politicians to bureaucrats might be perceived as
insufficient by taxpayers, especially if politicians continue to hold accountability authority.

56
The central criticism to Taliercio framework is that if MoFs continues to be part of the
accountability equation of SARAs, even after transferring autonomy to bureaucrats, how can
taxpayers develop perceptions that politicians in MoFs do not have any incentives to turn to their
old ways of doing things? How can taxpayers perceive that politicians will not use their oversight
power negatively for political gains, by trying to politicize tax officials to make them do what
pleases their accounting masters? It is argued here that this dichotomy by politicians of trying to
signal diminished influence in SARAs’ operations by giving away some of their autonomy, but
continuing to retain accountability, might lead to taxpayers questioning even any meaningful
autonomy transferred to tax officials. This is because tax officials might be conceptually more
autonomous than pre-reform, but this increased autonomy is kept under check by the same
politicians who transferred this autonomy, thus creating a potential avenue for misuse by
politicians. Also taxpayers might perceive that granting more autonomy to bureaucrats, while they
continue to be held accountable by previous political masters is undeserved, resulting in no
ownership and even resistance to reform by taxpayers, thus affecting tax compliance.
Taliercio (2004a) also highlighted that the success of third-party delegation depends upon
important pre-conditions of suitable selection of third party, such as selection of credible
professional managers, and providing them with incentives to continually improve performance.
In order to increase taxpayers’ compliance, they need to be convinced about professional and fair
reputation of the bureaucracy involved, resulting in decreased perceptions of corrupt practices
undertaken by bureaucrats, even in the face of increased autonomy. Referring to the argument of
how not-enough disaggregation from MoFs might undermine any meaningful transfer of authority
to bureaucrats in the minds of taxpayers. It is proposed here that not-enough disaggregation from
politicians in MoFs might undermine any meaningful transfer of authority to bureaucrats by
affecting the pre-condition of suitable selection of third parties as proposed by Taliercio (2004a).
57
In this case not-enough disaggregation from MoFs, with politicians continuing to retain
accountability authority, might lead to taxpayers questioning the credibility of suitable selection of
third parties i.e., bureaucrats, who are selected to be empowered by transferring authority from
politicians. This is particularly possible when politicians can make use of their accountability
authority to influence which tax officials get selected to be empowered under third party
delegation solution. The existence of colluding networks between politicians and bureaucrats are
not uncommon in developing countries with widespread perception of only extremely politically
aligned bureaucrats getting plum postings due to politicians support. This is the basic criticism to
theoretical arguments of Taliercio (2004a). It is postulated that the ‘delegation of authority to third
party’ solution might conceptually be insufficient to convince taxpayers of suitability and credibility
of bureaucrats as third parties, due to persistent accountability link with previous political
principals in former parent MoFs.
After highlighting a potential weakness in theoretical model by Taliercio (2004a), the next
question is how to improve adoption of ‘third-party delegation solution’ by improving taxpayers’
perceptions of de-politicization, suitable selection, and capability of bureaucrats to act as genuine
third parties. Drawing on the discussion above leads to the following potential solution. It is
proposed, why can’t politicians distance themselves further away from SARAs’ operations by not
only giving away autonomy to suitable third parties, but also transferring their accountability
authority to other suitable oversight bodies? It is proposed that SARAs might be completely
disaggregated from MoFs, such that previous political principals do not retain any accountability
authority. It is argued that complete disaggregation of SARAs from MoFs might result in stronger
taxpayers’ perception of insulation of SARAs from political influence.

58
2.5.2.3 Formation of Corporate Bodies: Accountability by Taxpayers in Revenue Boards
Taliercio (2004a) proposed that the second ‘commitment technology’ which could be utilized by
politicians to create credible commitment for SARA reform is related to formation of corporate
bodies i.e., Revenue Boards. He proposed that transfer of accountability functions to taxpayers in
the form of Revenue Boards, can also aid towards development of taxpayers perception of
diminished politicians influence in SARAs operations, hence leading to increasing insulation from
politics. By utilizing the instrument of Revenue Boards, politicians also try to signal increased
involvement of taxpayers in policy making process as stakeholders (via taxpayers
representatives sitting in Revenue Boards), thus utilizing Revenue Boards as institutions of joint
decision making. Taliercio proposed that time consistency problem of SARA reform (as described
above) can be solved by giving power to taxpayers to take part in joint decision-making, by
adoption of instrument of Revenue Boards, with representatives of private sector on board. Such
‘corporate body’ solution have been witnessed in numerous developing countries, such as
Zambia, Kenya, Malaysia, Uganda and Singapore, whereby Revenue Boards comprise of several
members belonging to top private sector organizations, such as representatives from Chambers
of Commerce and Industry and Institution of Chartered Accountants etc. (Hall & Jenkins, 1995).
The instrument of ‘formation of corporate bodies’ works in the same manner as ‘delegation of
authority to third parties’, by signalling to taxpayers insulation of SARAs operations from political
influence. This is signalled by forming Revenue Boards by transferring accountability functions,
previously held by politicians in centralized MoFs, to taxpayers sitting as board members in
Revenue Boards. The adoption of Revenue Boards to signal to taxpayers SARAs’ insulation from
political influence, rests on the logic that taxpayers hold past perceptions about ineffectiveness of
pre-SARA accountability arrangements, failing to control corruption due to existence of colluding
corruption networks between politicians and bureaucrats. As per this argument if instrument of
59
Revenue Boards is to be effective for insulating SARAs from politics, then taxpayers should
perceive a break in pre-SARA accountability arrangements. This is proposed to lead to
taxpayers’ perception of meaningful transfer of accountability functions from politicians to
taxpayers sitting in Revenue Boards. But what if taxpayers don’t perceive a break in pre-SARA
accountability arrangements, due to not-enough disaggregation of SARAs from previous political
principals in MoFs, with politicians still retaining accountability responsibility?
As noted above, Taliercio (2004) proposed hierarchical multi-layered accountability relationships
between SARAs and several oversight bodies including Ministries of Finance, in addition to
Revenue Boards, the Comptroller/Auditor General, and Parliament. It is proposed here, that as
politicians in MoFs, whose influence was supposed to diminish due to transfer of accountability
authority to taxpayers in Revenue Boards, remain part of accountability equation of SARAs, this
may hinder development of taxpayers’ perception of meaningful insulation from political
interference.
How to improve adoption of ‘Revenue Boards’ by improving taxpayers’ perceptions about
meaningful transfer of accountability functions from politicians to taxpayers and insulation from
politics. It is proposed that if Revenue Boards are to be employed specifically for control
corruption then they should replace, rather than augment, past accountability relationship of
SARAs with MoFs. In other words, establishment of a visible accountability link between SARAs’
bureaucrats and taxpayers in Revenue Boards, as opposed to previous accountability link
between pre-SARA bureaucrats and politicians in MoFs, is conceptualized to aid taxpayers’
perception of insulation of SARAs from politics. Also, if SARAs are continued to be held
accountable by MoFs, in addition to Revenue Boards, then taxpayers might question the
capability and value added by creation of Revenue Boards.

60
While discussing appropriate levels of disaggregation of SARAs from MoFs in the light of SARA
reform framework proposed by Taliercio, this section critiqued his framework. This critique rested
on the ground that Taliercio’s framework might be effective in achieving one of the objectives
behind SARA reform adoption of improvement in revenue performance, but displays a weakness
when applied to achieve control of corruption, by failing to convince taxpayers about insulation of
SARAs from politics, due to persistent accountability link with politicians in MoFs. It was proposed
that politicians in MoFs needed to completely disaggregate themselves from SARAs’ operations
to enable taxpayers to perceive meaningful insulation from politics. It is stressed that the
proposed solution to identified weakness in Taliercio’s framework is conceptualized to improve
SARA reform’s robustness against corruption, in addition to revenue improvement.
2.5.3 Balancing Autonomy with Accountability for Controlling Corruption
Section 2.5.1 specifically focused on analyzing appropriate level of autonomization of SARAs to
effectively control corruption; whilst, section 2.5.2 specifically focused on analyzing appropriate
level of disaggregation of SARAs from parent MoFs, which might enable them to effectively
control corruption in tax administrations in developing countries. This section aims to further
develop this line of argument by analyzing balance between increased level of autonomy with
increased level of effective accountability, to enable SARAs to control opportunities for
corruption, by reducing taxpayers’ perceptions about discretionary and monopoly powers
available to tax officials. In addition, this section also proposes that if there is not enough
disaggregation of SARAs from MoFs, then it is conceptualized to hinder balance between
increased autonomy with increased effective accountability to occur to control opportunities for
corruption after SARA reforms.

61
It is questioned in this section if and how failure to balance increased level of autonomy with
increased level of effective accountability might lead to ineffectiveness of SARAs in controlling
opportunities for corruption in tax administrations in developing countries? This question points to
a need to closely analyze theories employing autonomy and accountability concepts to control
opportunities for corruption. In order to achieve this objective, this section of the thesis aims to
analyze corruption analytical framework by Klitgaard (1988, 1997), not only to throw some light
on theories behind controlling corruption, but to look for answers to the question highlighted
above. This framework represents one of the most prominent conceptual models developed in
literature to improve our understanding of how autonomy and accountability concepts might be
utilized to control corruption. Specifically, Klitgaard’s framework will be analyzed to understand
how autonomy and accountability concepts might be employed for SARA reform to control
opportunities for corruption. It questions whether the solution to SARAs performing poorly against
corruption in tax administrations in developing countries lies in ensuring balance between
increased level of autonomy with increased level of effective accountability? This is to enable
SARAs to control opportunities for corruption by reducing taxpayers’ perceptions about
discretionary and monopoly powers available to tax officials.
2.5.3.1 Corruption equals Monopoly plus Discretion minus Accountability
There is a very important conceptual difference between the concepts of discretionary and
monopoly powers available to tax officials under a centralized tax administration system; and
autonomy granted to tax officials under decentralized SARA reforms. Discretionary and monopoly
powers imply a sort of unofficial autonomy exploited by tax officials in the face of a complex tax
system which is not sufficiently understood by taxpayers. This increases the monopoly power of
tax officials to make discretionary decisions, leading to increases in opportunities for corruption.
These discretionary powers are generally exploited by tax officials pre-SARA reform where they
62
are governed by ineffective centralized accountability mechanisms (Klitgaard, 1988, 1997).
These discretionary powers of tax officials with propensity to increase opportunities for corruption
can also be thought of as an un-checked form of autonomy, which is not sufficiently checked due
to ineffective accountability mechanisms. On the other hand, the autonomy granted to tax officials
after SARA reforms should not be confused with the discretionary and monopoly powers
available to tax officials before SARA reforms. This assertion rests on ground that increased level
of official autonomy, granted to tax officials after SARA reform, is conceptualized to be kept under
check by increased level of effective accountability mechanisms. As Schick (1998) pointed out,
the underlying principle behind the creation of agencies is a combination of ‘letting managers
manage’ and ‘making managers manage’, which in essence is about granting increased
autonomy in operations in exchange for greater accountability for results.
The conceptual difference between ‘discretionary and monopoly powers’ and ‘autonomy’ of tax
officials has an important bearing for SARA reforms for having propensity to increase
opportunities for corruption. It is proposed that when SARA reforms have been implemented
improperly with more focus on ensuring autonomy without balancing with effective accountability.
This can lead to conversion of and relapse of autonomy concept under SARA reforms back into
discretionary and monopoly powers available to tax officials pre-SARA reform. This proposition
rests on the argument that monopoly and discretionary powers exploited by tax officials are
conceptualized as that form of unofficial autonomy which arises in the face of ineffective
accountability mechanisms. Rather this unchecked autonomy, becoming available to tax officials
due to improper implementation of SARA reforms, could be considered more dangerous than
monopoly and discretionary powers available to tax officials pre-SARA reforms. Because after
adoption of SARA reforms tax officials are entrusted with more autonomy then was previously
available to them. In the wake of ineffective accountability mechanisms, the increased level of un-
63
checked but official autonomy is conceptually and comparatively more capable of increasing
opportunities for corruption for tax officials, than were possible pre-SARA reform. This can be
further elaborated by analysing the corruption framework proposed by Klitgaard (1988, 1997).
Klitgaard (1988, 1997) presented a simple and functional analytical framework modelling
corruption, by focusing on corrupt systems rather than corrupt individuals. He highlighted three
broad factors affecting opportunities for corruption, including: Monopoly powers of public officials
over clients (M), Discretionary control over provision of services (D), and Accountability levels for
public officials (A). According to this framework, the probability of incidence of corruption is
represented by the following simple equation (a):
This equation suggests the greater the degree of monopoly and discretionary powers accorded to
public officials, the greater will be the size and number of opportunities to engage in corruption,
minus the level of accountability mechanisms in place (Fjelstad, 2005b; Purohit, 2007). In order
to reduce opportunities for corruption available to public officials, an anti-corruption strategy
based on this framework needs to focus on reducing monopoly and discretionary powers
available to public officials, while simultaneously enhancing the accountability mechanisms in
place (Rose-Ackerman, 1999; Purohit, 2007). It is highlighted here that the above equation
represents a simplified abstract tool and is not a true representation of reality.
At first glance this framework of corruption along with its explanation by Klitgaard (1988, 1997)
seems to suggest that if the objective of reform is to reduce corruption, then autonomy granted to
tax officials (which in this definition is a combination of monopoly plus discretionary powers
64
accorded to tax officials) needs to be reduced, while simultaneously enhancing accountability of
tax officials. But isn’t this proposal at odds with the basic principle of SARA reforms propagating
both increased levels of autonomy as well as accountability? Why is this framework of corruption
propagating a decrease in the level of autonomy granted to tax officials in order to control
opportunities for corruption? Is this framework also pointing out that since agencification under
NPM propagates increased level of both autonomy and accountability, and as Klitgaard’s
corruption framework propagates decreased level of autonomy while simultaneously increased
level of accountability for tax officials to control corruption. This leads to the proposition that
agencification in NPM has propensity to increase opportunities of corruption since it entails
increased level of autonomy accorded to tax officials? Is this observation lending support to
literature (Kernaghan, 1992; Doig, 1997; Dunleavy & Hood, 1994; Greenaway, 1995; Haque,
2000; Peters, 1996; Theobald, 1997) which highlights that agencification in NPM has propensity
to increase opportunities for corruption arising due to decentralization?
There are a number of possible explanations to the questions raised above. Although, at first
glance, it does seem that this corruption framework is suggesting a decrease in the level of
autonomy along with simultaneous increase in accountability in order to control corruption. But
this corruption framework can be analyzed differently in the following way. It is proposed that
Klitgaard’s corruption framework is mainly developed for and targeted towards centralized tax
administrations, where generally tax officials are entrusted with limited autonomy along with tight
centralized accountability mechanisms. In these cases opportunities for corruption arise when
centralized tax administrations fail to control tax officials from exercising unofficial autonomy due
to ineffective accountability mechanisms. In these cases, Klitgaard’s corruption framework
suggests, as highlighted in equation (a) above, decreasing the unofficial autonomy exercised by
tax officials, in addition to simultaneously taking steps to tighten accountability. This corruption
65
framework suggests reducing the ‘un-official’ form of autonomy, which is exploited by tax officials
and was never entrusted to them, and only arisen due to ineffective accountability mechanisms.
So, Klitgaard’s corruption framework does not propagate reducing autonomy of tax officials in
order to control opportunities for corruption. Rather this framework posits reducing discretionary
and monopoly powers of tax officials, which in this study are conceptualized as different from
autonomy concept, and represents that instance of un-official autonomy not governed by
effective accountability. Literature also highlighted that it is very much possible that individuals
and organizations are capable of exercising great levels of freedom and autonomy, although
there has been no disaggregation from formal hierarchy. In other words, there is increase in
autonomization without any formal disaggregation of the organization from formal hierarchy. This
observation becomes particularly evident when analyzing public bureaucracies in developing
countries, where ever-increasing regulatory pressures ironically result in even greater
opportunities for exercising unofficial discretion and informal autonomy, many times associated
with corrupt activities (de Soto, 2000).
2.5.3.2 Corruption equals Autonomy minus Accountability for SARAs
The previous section aided in highlighting how Klitgaard’s corruption framework is better suited to
shed light on how to control opportunities for corruption in centralized tax administrations by
decreasing unofficial autonomy, in addition to simultaneously taking steps to tighten
accountability. This section aims to adapt Klitgaard’s corruption framework to also make it
suitable to analyze control of corruption in SARA reforms by analyzing it through a different lens.
It is proposed that Klitgaard’s corruption framework can be adapted as follows to also make it
suitable to analyze how to control opportunities for corruption in decentralized tax administrations
after SARA reforms. By combining both monopoly and discretionary powers into total autonomy
entrusted to tax officials, and converting the following framework as described below, Klitgaard’s
66
corruption framework can also be utilized to analyze how to control opportunities for corruption in
decentralized tax administrations after SARA reforms.
Klitgaard’s analytical framework for corruption control proposed in equation (a):
The concepts of monopoly and discretionary powers represent the total unofficial autonomy
exploited by tax officials. If we amend the above equation (a) by combining both monopoly and
discretionary powers into total unofficial autonomy available to tax officials, it would be presented
as follows in equation (b):
Now integrating this amended autonomy concept from equation (b) back into equation (a) by
replacing (Monopoly + Discretion) in equation (a) with (Autonomy) from equation (b), or by
replacing (M + D) in equation (a) with (A) from equation (b), leads to conversion of equation (a)
into equation (c) as follows:
As per equation (c) it is further proposed here that as long as SARAs’ level of autonomy (no
matter how much higher than before) is balanced (or nullified) by equal level of effective
67
accountability, then hypothetically corruption can be reduced to zero. In other words, corruption
can be controlled by balancing equal level of autonomy with proportionate effective accountability
as shown below in equation (d):
Corruption = Autonomy – Accountability
(C = A – A)
Corruption = 0
There are two caveats here. First, equation (d) is not proposing that by balancing equal level of
autonomy with proportionate effective accountability, corruption level can be brought down to
zero in real terms. As review of literature highlighted that from an economic stand-point it is
impossible to bring corruption level equal to zero. Rather zero in equation (d) signifies that by
balancing equal level of autonomy with equivalent effective accountability, corruption can be
reduced to an acceptable level. Second, by proposing in equation (d) that if equal level of
autonomy is balanced with equivalent effective accountability then both nullify the effect of each
other, is not meant to propose that effect of both concepts is cancelled or nullified by each other.
Rather it is to suggest that there is no un-accountable part of autonomy concept which is not
checked by accountability concept. Also as both autonomy and accountability concepts are
balanced with each other, there is no un-accountable autonomy left, which prevents opportunities
for corruption from being controlled.
The previous section highlighted how Klitgaard’s corruption framework can be adapted to make it
suitable to examine control of corruption under decentralized SARA reforms, although Klitgaard’s
corruption framework was originally developed to examine control of corruption under centralized
tax administrations. Importantly the arguments highlighted above can be utilized to explain why
68
some developing countries have not been able to effectively control opportunities for corruption
even after SARA reforms. Leading from arguments above, it is proposed that those SARA cases
which have not been effective against controlling opportunities for corruption, they might be
represented by a case where increased level of autonomy has not been balanced with increased
level of effective accountability mechanisms. Rather autonomy levels granted to tax officials were
greater than as could be effectively controlled with available accountability. This occurrence is
proposed to lead to conversion of increased autonomy under SARA reform into a form of
unofficial autonomy (i.e., discretionary and monopoly powers) available to tax officials in the face
of ineffective accountability mechanisms, which is proposed to lead to increases in opportunities
for corruption.
The above case highlighting imbalance between both autonomy and accountability concepts
under SARA reforms, leading to failure in controlling increasing opportunities for corruption, and
bringing corruption hypothetically equal to zero, can be illustrated in equations below. If under
SARA reforms there is more focus on ensuring increased autonomy for tax officials than also
balancing it with proportionate accountability. In this case it is proposed that level of autonomy is
higher than as could be safeguarded with effective accountability mechanisms to control
opportunities for corruption. In other words increased level of autonomy is greater than existing
level of accountability as illustrated in equation (e) below:
Also since equation (b) has already proposed above that both concepts of monopoly and
discretionary powers represent the total unofficial autonomy exploited by tax officials as
reproduced below:
69
Now integrating this amended autonomy concept from equation (b) back into equation (e) by
replacing (Autonomy) in equation (e) with (Monopoly + Discretion) from equation (b), or by
replacing (A) in equation (e) with (M + D) from equation (b), leads to conversion of equation (e)
into equation (f) as follows:
As per equation (f) since both monopoly and discretionary powers available to tax officials are
greater than as could be effectively controlled by accountability mechanisms. This leads to
ineffectiveness of SARA reform to effectively control opportunities for corruption and bringing
corruption level hypothetically equal to zero as follows in equation (g):
It is important to note that in case of improper implementation of SARA reforms where increased
autonomy is not balanced with proportionate effective accountability {as shown above in equation
(e)}, it is proposed to lead to not only increases in opportunities for corruption {as shown above in
equation (g)}. But also proposed to lead to conversion of autonomy concept, as employed after
SARA reforms, into same monopolistic and discretionary powers exploited by tax officials under
70
centralized tax administrations pre-SARA reforms {as shown above in equation (b)}. This
discussion rested on the ground that the differentiating property of autonomy concept, as
employed after SARA reforms, from the unofficial autonomy available under centralization, is that
although autonomy after SARA reforms is at a higher level than was entrusted to tax officials
under centralization, but is conceptually balanced with proportionate effective accountability {as
shown above in equation (d)}. So, those SARA reform countries which have not been able to
effectively control increasing opportunities for corruption could be represented by a case where
autonomy levels granted to tax officials after reforms are greater than as could be effectively
controlled by existing accountability levels. In this case, this un-checked autonomy concept is
conceptualized to transform into a combination of monopoly plus discretionary powers available
to tax officials, as not controlled by effective accountability mechanisms. Consequently corruption
in this case becomes equal to monopoly plus discretionary powers exercised by tax officials,
which are greater than effective accountability mechanisms in place {as shown above in equation
(f)}.
From the above discussion, it can also be deduced that if SARAs are not completely
disaggregated from any accountability link with parent MoFs, such that presence of MoFs in
SARAs’ accountability equation continues to undermine both autonomy (by undermining
delegation of authority to third parties) and accountability (by undermining oversight by Revenue
Boards) of SARAs. Consequently, this is proposed to lead to failure to balance increased level of
autonomy with increased level of effective accountability mechanisms, with imbalance arising due
to persistent accountability link of SARAs with MoFs. This is proposed to lead to ineffectiveness
of accountability mechanisms of SARAs to control opportunities for corruption, by failing to
reduce taxpayers’ perceptions about discretionary and monopoly powers available to tax officials
even after reforms.

71
This section not only highlights the ‘Why’ and ‘How’ questions of conversion of unofficial
autonomy concept into monopoly and discretionary powers available to tax officials before SARA
reforms. But this section also took this argument further and applied and adapted the similar
argument to ‘Why’ and ‘How’ increased level of official autonomy after SARA reforms, in the face
of ineffective accountability mechanisms, was proposed to transform into similar monopolistic and
discretionary powers available to tax officials before reforms.
2.6 SUMMARY AND CONCLUSION
This chapter has explored various themes about adoption of SARAs in developing countries with
particular focus on controlling corruption. The five major areas covered included introduction to
SARAs, introduction to corruption, motivations and opportunities for corruption in revenue
administration, discussion about theoretical underpinnings of SARAs and how to to better control
opportunities for corruption in SARAs by analysing prominent conceptual frameworks in this field.
The specific contribution of chapter two to the overall thesis argument lies in locating the thesis
within existing knowledge about SARAs in developing countries for controlling corruption
The discussions raised in this chapter will be utilized in next chapter to develop an Anti-
Corruption SARA Framework by analyzing individual SARA design components towards
controlling motivations and opportunities for corruption in tax administrations in developing
countries. This analytical framework will be developed by utilizing two SARA frameworks
including SARA design components proposed by Taliercio (2004) and control of corruption
framework for tax administrations by Martinez-Vazquez et al. (2004).The development of
analytical framework will result in formulation of two research hypotheses in the next chapter.
72
CHAPTER THREE
DEVELOPMENT OF ANALYTICAL FRAMEWORK FOR CONTROLLING CORRUPTION IN
SARAs
3.0 INTRODUCTION
Chapter two analysed prominent SARA conceptual models in literature, which not only threw light
on theories behind SARA reforms, but also looked for answers to the theory-practice paradox.
SARA frameworks were analysed to explain why experiences of SARA reforms in developing
countries deviated from conceptual prescriptions. It was specifically questioned why some
SARAs have failed to effectively control opportunities for corruption.
Drawing from discussions raised in chapter two, this chapter seeks to develop an analytical
framework for SARAs by incorporating potential solutions to identified problems, which came to
light while critically analysing SARA models. The analytical framework will be developed by
answering questions such as: which SARA design components contribute towards reducing
corruption?; what are the processes by which this outcome is achieved?; and are certain SARA
design components better suited to curb corruption than others? The development of this
framework is aimed to contribute towards explaining how existing SARA model can be improved
not only to effectively control corruption in tax administrations in developing countries, but also to
improve revenue performance as a consequence. The analytical framework developed in this
chapter will be used for conducting two-staged analysis of SARAs in chapters 5 and 6 of the
thesis, where the analytical framework will be tested and revised for SARA country cases.
73
The rest of this chapter is organized as follows. Drawing onto discussions in chapter two lead to
development of analytical framework and two research hypotheses set out in section 3.1. Both
research hypotheses are formulated to lend answer to question why some SARAs have failed to
effectively control opportunities for corruption. Research hypothesis 1 proposes in section 3.1.5
that by preferring personnel autonomy over effective accountability, SARAs have been effective
against motivations but not opportunities for corruption. Also research hypothesis 2 proposes in
section 3.1.6 that SARAs have been ineffective in controlling opportunities for corruption due to
ineffective accountability by MoF. Section 3.2 concludes the chapter with a general summary and
discussion. The structure of Chapter 3 is shown in Figure 3.1.

74
3.0
INTRODUCTION
3.1
AUTONOMY VS. ACCOUNTABILITY IN SARAs
3.1.1
Preference of Autonomy over Accountability in
SARAs
3.1.2
Preference of Autonomy for Controlling
Motivations and Accountability for Controlling
Opportunities for Corruption in SARAs
3.1.3
Preference of Personnel Autonomy for
Controlling Motivations for Corruption in SARAs
3.1.4
Preference of Effective Accountability for
Controlling Opportunities for Corruption in SARAs
3.1.5
Preference of Personnel Autonomy over Effective
Accountability for Controlling Corruption in
SARAs
3.1.6
Why Ineffective Accountability for Controlling
Corruption in SARAs?
3.2
75
3.1 AUTONOMY VS. ACCOUNTABILITY IN SARAs
Keeping in mind the discussions presented in theory-practice paradox section in chapter two, this
section aims at developing an Anti-Corruption SARA Framework by analyzing individual SARA
design components towards controlling motivations and opportunities for corruption in tax
administrations in developing countries. This analytical framework is developed by utilizing two
SARA frameworks including SARA design components proposed by Taliercio (2004) and control
of corruption framework for tax administrations by Martinez-Vazquez et al. (2004).
What follows in this section is a logical flow of discussion towards development of research
hypotheses, which not only aids further development of the thesis argument in the form of
analytical framework, but also lending clarity to the sequence of arguments. It is stressed here
that the development of research hypotheses is an iterative process and not linear development.
The following sections explain process of development of research hypotheses through iterative
reasoning.
3.1.1 Preference of Autonomy over Accountability in SARAs

As highlighted in chapter two before, SARA reform has been adopted by developing countries for
not only efficiency improvement but also specifically for controlling corruption in tax
administrations (McCourt, 2002; Fjeldstad, 2003, 2005a, 2005b; Kidd & Crandall, 2006; Mann,
2004; Martinez-Vazquez et al., 2004; McCarten, 2006; Taliercio, 2004; Zuleta, 2007). Although
control of corruption is one of the objectives behind SARA reform adoption, the literature points to
poor empirical record of SARAs against sustainably controlling corruption in tax administrations
(Fjeldstad, 2003, 2005a, 2005b; Mann, 2004). This has resulted in researchers arguing against
developing countries (Martinez-Vazquez et al., 2004; Fjeldstad, 2005a; Kidd & Crandall, 2006;
76
McCarten, 2006). It is questioned here if the poor record of SARAs against corruption can be
attributed to its failure to balance increased autonomy with proportionate accountability, leading
to increase in opportunities for corruption in tax administration in developing countries. In other
words, has SARA reforms resulted in preference for adopting autonomy-enhancing design
components but ineffective in ensuring proportionate accountability-enhancing design
components for corruption control?
After highlighting the above question, the following sections aim to look for answers and deducing
research hypotheses. Before moving onto next sections to formulate research hypotheses, first
the following section aims to briefly draw attention to the definitions of SARA reform and how it
has utilized its design components to enhance autonomy and accountability to control corruption.
The discussion above highlighted the need to examine the balance between autonomy and
accountability- enhancing design components in the SARA reform model. This is to examine
which design component has compromised its ability to effectively control corruption. It is justified
to assume that SARA model is constituted of separate design components with some enabling
more autonomy in the model, while others employed to tighten the accountability of the reformed
revenue authority, ensuring the balance between autonomy and accountability. According to
Taliercio (2004, page 46) “Semi-Autonomous Revenue Authorities have been defined as tax
administrations that have greater than usual autonomy along several organizational design
dimensions, including: legal character, corporate governance, financing and budgeting, personnel
policy, procurement policy, and accountability relationships”. As per this definition, SARA design
components such as legal character, financing and budgeting mechanisms, procurement policy
mechanisms and personnel policy mechanisms are incorporated in the reform design to extend
more autonomy to the reformed SARAs. While SARA design components such as corporate
77
governance mechanisms and accountability relationships with oversight bodies are instituted in
the reform design to extend more accountability of the reformed SARAs.
Since the on-going discussion is primarily aimed towards highlighting how SARAs utilize
autonomy and accountability-enhancing design components to control corruption. It is worthwhile
to highlight here that in order to understand the incidence, extent and mechanisms of corruption
in tax administrations, the literature points to numerous studies which have differentiated
between two distinct types of contributing factors to corruption, separately affecting either
motivations or opportunities for corruption (World Bank, 1999, Martinez-Vazquez et al., 2004). A
detailed discussion of these factors has already been done in chapter two, section 2.3. A brief
summary of these factors is reproduced here with an aim of placing the development of research
hypotheses in context.
The motivating factors to corruption influences the motivation of tax officials to indulge in
corruption given that an opportunity for corruption arises and answers the question of what
factors affect the decision by some tax officials to engage in corruption when facing an
opportunity for corruption? Also, Martinez-Vazquez et al. (2004) identified numerous preventive
strategies to control motivations for corruption in tax administrations including: instilling ethics in
tax officials; increases in the probability of detection; increases in and stricter enforcement of
penalties for corruption; and increases in wages in the public sector and the establishment of
incentive compatible compensation mechanisms.
In addition to motivations of tax officials to engage in corruption, the factors affecting the ‘ability’
of tax officials to indulge in corruption represents the ‘windows of opportunity’ for corruption. A tax
official might be willing and perceive incentives and motivations to engage in corruption, but in
78
order to engage in corruption, it is vital that an opportunity for corruption also arises. The
preventive strategies aimed at limiting the opportunities for tax officials to indulge in corruption
include: introduction of oversight mechanisms; reduction of discretionary powers of tax officials;
and de-politicization of tax officials (Martinez-Vazquez et al., 2004).
3.1.2 Preference of Autonomy for Controlling Motivations and Accountability for
Controlling Opportunities for Corruption in SARAs
This section is aimed to elaborate how SARA reform conceptually uses its autonomy and
accountability-enhancing design components to control corruption in tax administration. It is
proposed that both autonomy and accountability-enhancing design components of SARA reform
model are capable of checking corruption in tax administrations by specifically affecting either
motivations or opportunities of corruption or both. Taking this argument further for specifically
controlling motivations for corruption in tax administrations, it is conceptualized that autonomy-
enhancing design components of SARA reform model (such as funding mechanisms and
personnel management) are better equipped to control motivations for corruption in tax
administrations. These design components are better equipped to control corruption by
specifically affecting implementation of preventive mechanisms for controlling motivations for
corruption (as identified by Martinez-Vazquez et al., 2004) of: (1) increases in wages and
rewards; (2) increases in the probability of detection; (3) increases in and stricter enforcement of
penalties for corruption; and (4) instilling ethics in tax officials. This proposition rests on the logic
that by utilizing the autonomy-enhancing design components of personnel and financial
autonomy to, for example, increase wages and rewards of tax officials, their motivations for
corruption coming from lack of competitive compensation can be controlled. In order to control
motivations for corruption, both personnel autonomy and financial autonomy are autonomy-
enhancing design components of SARA reform playing their part. This is to ensure that SARA
79
reform is capable of controlling motivations for corruption by increases in wages and rewards
because it is governed by a personnel system which is free from general civil service regulations
(coming from personnel autonomy). Also, it has secure source of funding available to make such
increases in wages and rewards a reality for SARA employees (coming from financial autonomy).
In this case, both personnel autonomy and financial autonomy are autonomy-enhancing design
components of SARAs, incorporated in the reform model not only to extend autonomy, but this
increased level of autonomy, on these identified dimensions, also serves to control motivations
for corruption. This consideration of personnel and financial autonomy to control motivations for
corruption is in line with Martinez-Vazquez et al., (2004) who argue that an effective preventive
mechanism for controlling motivations for corruption for tax officials includes increasing in wages
and rewards for tax officials. The series of arguments highlighted above helped to establish that
the autonomy-enhancing design components of SARA reform are conceptualized to control
motivations for corruption.
Taking this argument further for specifically controlling opportunities for corruption in tax
administrations, it is conceptualized that accountability-enhancing design components of SARA
reform model are better equipped to control opportunities for corruption. The accountability-
enhancing design components referred here are defining SARA components such as corporate
governance mechanisms and effective ‘accountability relationships with oversight bodies’ as
highlighted in SARA definitions above. These design components are better equipped to control
corruption by specifically affecting implementation of preventive mechanisms for controlling
opportunities for corruption (as identified by Martinez-Vazquez et al., 2004) of (1) introduction of
oversight mechanisms; (2) de-politicization of tax officials, and (3) reduction of discretionary
powers of tax officials. This proposition rests on the logic that by utilizing accountability-
enhancing SARA design components of corporate governance mechanisms and effective

80
‘accountability relationships with oversight bodies’ to, for example, introduce improved oversight
mechanisms for SARAs, the opportunities for corruption arising out of weak oversight
mechanisms can be controlled. Note that in order to control opportunities for corruption both
corporate governance mechanisms and effective ‘accountability relationships with oversight
bodies’ are accountability-enhancing SARA design components playing their part. This is to
ensure that SARA reform is capable of controlling opportunities for corruption by introduction of
effective oversight mechanisms because it is not only supervised by and accountable to a
Revenue Board (coming from corporate governance mechanism), but it is also accountable to
several other supervising bodies including MoF and Parliament etc. (coming from effective
‘accountability relationships with oversight bodies’). So, in this case both corporate governance
mechanisms and effective ‘accountability relationships with oversight bodies’ are accountability-
enhancing SARA design components that are incorporated in the SARA reform model not only to
extend accountability. But it is also proposed that this increased level of accountability on these
different identified dimensions also serves to control increasing opportunities for corruption. This
consideration of corporate governance mechanisms and effective ‘accountability relationships
with oversight bodies’ to control opportunities for corruption is in line with Martinez-Vazquez et
al., (2004) who argue that an effective preventive mechanism for controlling opportunities for
corruption for tax officials includes introduction of effective oversight mechanisms. The series of
arguments highlighted above helped to establish that the accountability-enhancing design
components of SARA reform are conceptualized to control opportunities for corruption.
3.1.3 Preference of Personnel Autonomy for Controlling Motivations for Corruption in
SARAs
The previous section discussed that some SARA design components are included in the model
to extend autonomy (such as personnel management and funding mechanisms), whilst other
81
design components are included to extend accountability (such as corporate governance
mechanisms and effective ‘accountability relationships with oversight bodies’). It was also
proposed that autonomy-enhancing design components are conceptualized to control motivations
for corruption; while accountability-enhancing design components are conceptualized to control
opportunities for corruption in tax administrations.
This section now aims to discuss that out of all autonomy-enhancing design components of
SARA reform; conceptually which component has the most intense impact on controlling
motivations for corruption in tax administrations. Similarly, next section will highlight that out of all
accountability-enhancing design components of SARA reform; conceptually which component
has the most intense impact on controlling opportunities for corruption in tax administrations. In
doing so, this discussion serve to highlight that all distinctive defining features of SARA reform,
with some enabling more autonomy while others improving accountability, do not affect
motivations and opportunities of corruption in the same way. That is to say that though many
design components of SARA reform are included in the reform model to improve autonomy, not
all of them affect motivations for corruption in the same way. Similarly for accountability, it is to
highlight that although multiple design components of SARA reform are included in the reform
model to improve accountability, but not all of them affect opportunities for corruption in the same
way.
It is proposed that out of all autonomy-enhancing SARA design components (including legal
autonomy, personnel autonomy and financial autonomy), personnel autonomy has the most
notable and strong impact towards controlling motivations for corruption in tax administrations.
This is owing to reason that personnel autonomy is the single autonomy-enhancing SARA design
component, in comparison to other components, which has the most pronounced and visible
82
effect on implementing each preventive strategy to control motivations for corruption in tax
administrations. (The preventive strategies in question here are as proposed by Martinez-
Vazquez et al., (2004) and were already highlighted above in the earlier section 3.1.1). In other
words, personnel autonomy is proposed to exert most pronounced and direct effect towards
controlling motivations for corruption in tax administrations because all identified preventive
strategies to control motivations for corruption can be implemented by increasing personnel
autonomy. This importance accorded to personnel autonomy in relation to other autonomy-
enhancing SARA design components is in line with the findings obtained in relevant other studies
on SARA reforms in developing countries. In particular, Taliercio (2004) concluded that the effect
of personnel autonomy towards improvement in SARAs’ performance, in comparison to other
components of autonomy, was most intense. Similarly, it is proposed here that the impact of
personnel autonomy on controlling motivations for corruption in tax administrations is proposed to
be most intense due to its most noticeable effect on implementing all preventive strategies to
control motivations for corruption in tax administrations. The series of arguments highlighted
above helped to establish that SARA reform, out of all its design components, is conceptualized
to utilize personnel autonomy most intensely to control motivations for corruption.
Next it is discussed how personnel autonomy contributes towards controlling motivations for
corruption by directly affecting implementation of different preventive strategies to control
corruption in tax administrations. Personnel autonomy is proposed to control for motivations for
corruption in tax administrations by affecting implementation of preventive strategy of increases
in wages and rewards (with the indicators being increases in wages for tax officials and utilization
of performance-linked bonuses). Also, by increases in the probability of detection (with the
indicators being increases in the quality and frequency of internal audits and probes by
establishing internal audit and anti-corruption divisions; increases in the number of supervisory
83
personnel assigned to internal audit and anti-corruption divisions; and statutory condition of
declaration of assets by all employees). Moreover, by increases in and stricter enforcement of
penalties for corruption (with the indicators being increases in imposition of high levels of
penalties including monetary sanctions, job dismissals and prison sentences); and lastly by
instilling ethics in tax officials (with the indicators being increases in ethics training and adoption
of code of conduct). The series of arguments highlighted above helped to propose that SARA
design component of personnel autonomy is conceptualized to control motivations for corruption
by 1) increases in wages and rewards; 2) increases in the probability of detection; 3) increases in
and stricter enforcement of penalties for corruption; and 4) instilling ethics in tax officials.
3.1.4 Preference of Effective Accountability for Controlling Opportunities for Corruption in
SARAs
The discussion in the previous section highlighted both why and how personnel autonomy is
proposed to control motivations for corruption in tax administrations. This section aims to
highlight similar line of argument discussing why and how effective accountability is proposed to
control opportunities for corruption in tax administrations.
It is proposed that out of all accountability-enhancing SARA design components (including
corporate governance mechanisms and ‘accountability relationships with oversight bodies’),
effective accountability has the most notable impact on controlling opportunities for corruption in
tax administrations. This is because effective accountability is the single accountability-enhancing
SARA design component, in comparison to other components, which has the most pronounced
and visible effect on implementing each preventive strategy to control opportunities for corruption
in tax administrations. (The preventive strategies in question here are as proposed by Martinez-
Vazquez et al., (2004) and were already highlighted above in the earlier section 3.1.1) In other
84
words, effective accountability is proposed to exert most pronounced and direct effect towards
controlling opportunities for corruption in tax administrations because all identified preventive
strategies to control opportunities for corruption can be implemented by improving accountability.
This importance accorded to effective accountability in relation to other accountability-enhancing
SARA design components is in line with the findings obtained in relevant other studies on SARA
reform in developing countries. In particular, Taliercio (2004) concluded that the impact of
effective accountability towards improvement in SARA performance was very intense. Similarly, it
is proposed here that the impact of effective accountability on controlling opportunities for
corruption in tax administrations is proposed to be most intense due to its most noticeable effect
on implementing all preventive strategies to control opportunities for corruption in tax
administrations. The series of arguments highlighted above helped to establish that SARA
reform, out of all its design components, is conceptualized to utilize effective accountability most
intensely to control opportunities for corruption.
Next it is discussed how effective accountability contributes towards controlling opportunities for
corruption in tax administrations by directly affecting implementation of different preventive
strategies to control corruption in tax administrations. Effective accountability is proposed to
control for opportunities for corruption in tax administrations by affecting implementation of
preventive strategy of introduction of oversight mechanisms (with the indicators being effective
accountability relationships with all supervising bodies including MoF, Revenue
Boards/Superintendents (whichever applicable), and Parliament etc.). Also, by de-politicization of
tax officials (with the indicators being effective accountability relationships with all supervising
bodies especially Revenue Boards/Superintendents (whichever applicable), and Parliament etc.,
in addition to MoF). And, lastly by reduction of discretionary powers of tax officials (with the
indicators being balancing effective accountability relationships between all supervising bodies
85
especially between MoF and other supervising bodies including Revenue
Boards/Superintendents (whichever applicable) and Parliament etc.). The series of arguments
highlighted above helped to propose that SARA design component of effective accountability is
conceptualized to control opportunities for corruption by 1) introduction of oversight mechanisms;
2) de-politicization of tax officials; and 3) reduction of discretionary powers of tax officials.
3.1.5 Preference of Personnel Autonomy over Effective Accountability for Controlling
Corruption in SARAs
Section 3.1.1 questioned if SARA reforms resulted in preference for adopting autonomy-
enhancing design components but ineffective in ensuring proportionate accountability-enhancing
design components for corruption control? The previous two sections have already discussed
that SARA reform, out of all its design components, is conceptualized to utilize personnel
autonomy most intensely to control motivations for corruption. Also, SARA reform, out of all its
design components, is conceptualized to utilize effective accountability most intensely to control
opportunities for corruption. Based on these two discussion points, the previous question can be
updated into proposing that SARA reforms has resulted in preference for adoption of personnel
autonomy but ineffective in ensuring proportionate accountability for corruption control.
Taking this discussion further, as personnel autonomy is conceptualized to control motivations for
corruption (section 3.1.3), and effective accountability is conceptualized to control opportunities
for corruption (section 3.1.4). So, integrating all this discussion together leads to the formation of
following research hypothesis:
Research Hypothesis 1: By preferring personnel autonomy over effective

accountability, SARAs have been effective against motivations but not opportunities for
corruption.
86
Research hypothesis 1 is formulated to lend answer to the question of why some SARAs have
failed to effectively control opportunities for corruption. It is proposed that SARA reform in some
instances might have failed to balance increased autonomy with proportionate accountability.
This, in turn, is proposed to have led to some progress in controlling motivations for corruption,
but failing to control opportunities for corruption.
Referring back to section 3.1.3, it was argued that personnel autonomy is conceptualized to
control motivations for corruption by 1) increases in wages and rewards, 2) increases in the
probability of detection, 3) increases in and stricter enforcement of penalties for corruption, and 4)
instilling ethics in tax officials. As a next step, all four of these preventive strategies for controlling
motivations for corruption are integrated with research hypothesis 1 above. This leads to
formulation of following four sub-hypotheses (with each one focusing on one preventive strategy
for controlling motivations for corruption):
Research Hypothesis 1(a): By increases in wages and rewards, SARAs have been
effective against motivations but not opportunities for corruption.
Research Hypothesis 1(b): By increases in the probability of detection, SARAs have

been effective against motivations but not opportunities for corruption.
Research Hypothesis 1(c): By increases in and stricter enforcement of penalties for

corruption, SARAs have been effective against motivations but not opportunities for
corruption.
Research Hypothesis 1(d): By instilling ethics in tax officials, SARAs have been
The underlying arguments behind each of these sub-hypotheses have been discussed before in
section 3.1.3. It is hoped that formulation of these sub-hypotheses will help to add detail and
focus to the examination of research hypothesis 1. Next, Figure 3.2 conceptualizes the sub-
hypotheses into a sign graph diagram.

87
Figure 3.2 Sign Graph Diagram for Hypothesis 1
Hypothesized relationship between Personnel Autonomy and Corruption in SARAs
Personnel
Autonomy
Wages and Probability of Enforcement of Ethics in Tax

Rewards Detection Penalties Officials
Motivations for Opportunities

Corruption for Corruption
Corruption
The sign diagram above proposes that by using personnel autonomy to increase wages and
rewards, probability of detection, stricter enforcement of penalties and by instilling ethics in tax
officials, SARAs are proposed to control for motivations but not opportunities for corruption. So as
per research hypothesis 1, SARAs have preferred adoption of personnel autonomy which is
effective against controlling motivations but not opportunities for corruption.

88
3.1.6 Why Ineffective Accountability for Controlling Corruption in SARAs?
The research hypothesis 1 above focused more on personnel autonomy and how it has been
adopted more to control motivations for corruption. Following this same logic, but now focusing
more on accountability, it can also be proposed that SARAs have been ineffective in controlling
opportunities for corruption due to ineffective accountability. This idea is an extension of the
same logic which was used to develop research hypothesis 1. In other words, it is to say that
when research hypothesis 1 proposes that adoption of personnel autonomy has been preferred,
then it also means to say that such preference has not been extended to adoption of effective
accountability. The next logical question which comes to mind is ‘Why’ SARAs have been unable
to adopt effective accountability? The discussion in chapter two section 2.5.2 argued against
inclusion of MoFs in the accountability arrangements of SARAs to ensure de-politicization of tax
officials. Taking this argument further, it is proposed that SARAs have been unable to adopt
effective accountability due to presence of MoFs in its accountability arrangements and hence
leads to formation of research hypothesis 2:
Research Hypothesis 2: SARAs have been ineffective in controlling opportunities for

corruption due to ineffective accountability by MoF.
Research hypothesis 2 is also formulated to lend answer to the question of why some SARAs
have failed to effectively control opportunities for corruption.
Also section 3.1.4 above stated that SARA design component of effective accountability is
conceptualized to control opportunities for corruption by 1) introduction of oversight mechanisms;
2) de-politicization of tax officials; and 3) reduction of discretionary powers of tax officials. It is
also proposed that presence of MoFs in the accountability arrangements renders SARAs
ineffective for all three preventive strategies to control opportunities for corruption. Following are
possible explanations for this discussion.

89
The first preventive strategy to control opportunities for corruption relates to introduction of
effective oversight mechanisms. It is proposed that presence of MoFs in the accountability
equation, in addition to accountability powers extended to supervising revenue boards,
undermines effective oversight mechanisms, due to tensions between these two bodies. This
underlying argument has already been discussed in detail in chapter two, section 2.5.2.3. It
discussed how continuing accountability link to MoFs can serve to undermine corruption control
in SARAs by negatively affecting both delegation of authority to third parties and transfer of
oversight to taxpayers (i.e., Revenue Boards).
The second preventive strategy to control opportunities for corruption relates to de-politicization
of tax officials. It is proposed that presence of MoFs in the accountability equation results in
continuation of politicization of tax officials in the guise of accountability by MoFs. This underlying
argument has already been discussed in detail in chapter two, section 2.5.2.2.It discussed why
centralized MoFs can be inclined to utilize their accountability powers negatively to politicize tax
officials, not only to make up for lost autonomy, but also for gaining lost patronage benefits that
were available to MoFs before SARA reform.
The third preventive strategy to control opportunities for corruption relates to reduction of
discretionary powers of tax officials. It is proposed that presence of MoFs in the accountability
equation, in addition to accountability powers extended to revenue boards, undermines reduction
of discretionary powers of tax officials, due to tensions between these two bodies. This
underlying argument has also been already discussed in detail in chapter two, section 2.5.3.2.
It discussed how tensions due to simultaneous accountability by both MoFs and revenue boards
resulted in imbalance between increased autonomy and effective accountability. This was
90
proposed to lead to conversion of autonomy concept into un-checked discretionary powers
available to tax officials in the face of ineffective accountability.
In the paras above, it was discussed that presence of MoFs in the accountability arrangements
renders SARAs ineffective for all three preventive strategies to control opportunities for
corruption. As a next step, all three of these preventive strategies for controlling opportunities for
corruption are integrated with research hypothesis 2 above. This leads to formulation of following
three sub-hypotheses (with each one focusing on one preventive strategy for controlling
opportunities for corruption):
Research Proposition 2(a): SARAs have been ineffective in controlling opportunities for
corruption due to ineffective accountability by MoF by not introducing effective oversight
mechanisms.
Research Proposition 2(b): SARAs have been ineffective in controlling opportunities for
corruption due to ineffective accountability by MoF by not de-politicizing tax officials.
Research Proposition 2(c): SARAs have been ineffective in controlling opportunities for
corruption due to ineffective accountability by MoF by not reducing discretionary powers
of tax officials.
It is hoped that formulation of these sub-hypotheses will help to add detail and focus to the
examination of research hypothesis 2. Next, Figure 3.3 conceptualizes the sub-hypotheses into a
sign graph diagram.

91
Figure 3.3 Sign Graph Diagram for Hypothesis 2
Hypothesized relationship between Accountability and Corruption in SARAs
Interference
from MoF
Effectiveness of Politicization of Discretionary

Oversight Tax Officials Powers of Tax
Mechanisms Officials
Opportunities
for Corruption
Corruption
The sign diagram above proposes that the continuing interference by MoFs in the accountability
arrangements of SARAs undermines not only the effectiveness of oversight mechanisms but
continue to politicize and increase discretionary powers of tax officials and hence opportunities
for corruption are not controlled.

92
In conclusion, this chapter sets out why some SARAs have failed to effectively control
opportunities for corruption. To answer this it has set out an Anti-Corruption SARA Framework by
analyzing individual SARA design components towards controlling motivations and opportunities
for corruption in tax administrations in developing countries. This analytical framework is
developed by utilizing two SARA frameworks including SARA design components proposed by
Taliercio (2004) and control of corruption framework for tax administrations by Martinez-Vazquez
et al. (2004). One of the advantages of this framework lies in conceptualizing how individual
SARA design components can be utilized to separately control motivations and opportunities for
corruption. In addition, another important contribution behind developing this framework lies in
bringing together ideas by Taliercio (2004) and Martinez-Vazquez et al. (2004), which have
previously remained separate. In a nutshell, the specific contribution of chapter three to the
overall thesis argument lies in aiming to answer the question set out above by conceptually
developing an Anti-Corruption SARA framework by drawing on literature reviewed in chapter two.
The development of analytical framework resulted in formulation of two research hypotheses.
Research hypothesis 1 proposed that by preferring personnel autonomy over effective
accountability, SARAs have been effective against motivations but not opportunities for
corruption. Also, research hypothesis 2 proposed that SARAs have been ineffective in controlling
opportunities for corruption due to ineffective accountability by MoF. With an aim to add focus
and specificity to both research hypotheses, so as to make the forthcoming analysis of SARA
cases more focused and to the point, sub-hypotheses were also developed. All sub-hypotheses,
with each one referring to one specific preventive strategy to control motivations and
opportunities for corruption, helped to logically break down both research hypotheses. For the
purpose of clarity both research hypotheses along with sub-hypotheses are reproduced below.
93
Figure 3.4 Research Hypothesis 1 along with sub-hypotheses
Research Hypothesis 1: By preferring personnel autonomy over effective accountability,

SARAs have been effective against motivations but not opportunities for corruption.
Research Hypothesis 1(a): By increases in wages and rewards, SARAs have been effective
against motivations but not opportunities for corruption.
Research Hypothesis 1(b): By increases in the probability of detection, SARAs have been
Research Hypothesis 1(c): By increases in and stricter enforcement of penalties for

corruption, SARAs have been effective against motivations but not opportunities for corruption.
Research Hypothesis 1(d): By instilling ethics in tax officials, SARAs have been effective
against motivations but not opportunities for corruption.
Figure 3.5 Research Hypothesis 2 along with sub-hypotheses
Research Hypothesis 2: SARAs have been ineffective in controlling opportunities for

corruption due to ineffective accountability by MoF.
Research Hypothesis 2(a): SARAs have been ineffective in controlling opportunities for
corruption due to ineffective accountability by MoF by not introducing effective oversight
mechanisms.
Research Hypothesis 2(b): SARAs have been ineffective in controlling opportunities for
corruption due to ineffective accountability by MoF by not de-politicizing tax officials.
Research Hypothesis 2(c): SARAs have been ineffective in controlling opportunities for
corruption due to ineffective accountability by MoF by not reducing discretionary powers of tax
officials.
94
The development of Anti-Corruption SARA Framework highlighted the need to map existing
SARA cases as per research hypotheses developed in this chapter. The development of sub-
hypotheses helped to make the thesis argument more focused. In order to analyze the validity of
the framework through analysis of SARA cases, the forthcoming analysis in chapters 5 and 6 of
the thesis will be examined in terms of these sub-hypotheses. Analysis of sub-hypotheses will be
undertaken with an aim to test both research hypotheses 1 and 2, and any answers or findings as
per analysis of these sub-hypotheses will mean to respond to both research hypothesis 1 and 2.
Next, chapter 4 will also outline the methodology to be used for conducting the proposed analysis
of SARA cases and case study of tax administration reforms in Pakistan.

95
CHAPTER FOUR
RESEARCH METHODS
4.0 INTRODUCTION
Chapter three set out an Anti-Corruption SARA Framework by analyzing individual SARA design
components towards controlling motivations and opportunities for corruption in tax
administrations in developing countries. In order to analyze the validity of the framework through
analysis of SARA cases, a two-staged analysis of SARA cases will be conducted in chapters five
and six of the thesis, where the analytical framework will be tested and revised for SARA country
cases. For this reason, this chapter aims to elaborate multiple research methods for both macro
and micro level of analysis.
This rest of this chapter is divided into three main sections. Section 4.1 discusses research
methods by explaining why multiple methods are employed to undertake macro and micro level
of analysis and sequence and timing of data collection. Section 4.2 details how systematic review
is conducting for macro level of SARA analysis. Section 4.3 explains how both interview data and
documentary sources are combined for conducting micro level of analysis for Pakistan. Section
4.4 concludes the chapter. The structure of chapter 4 is elaborated in Figure 4.1 below:
96
Figure 4.1: Structure of Chapter 4
4.0
INTRODUCTION
4.1
RESEARCH METHODS
4.2
METHODS FOR MACRO ANALYSIS
OF SARAS
4.3
METHODS FOR MICRO ANALYSIS
OF PAKISTAN
4.3.1 4.3.2
Interviews Document Analysis
4.4
97
4.1 RESEARCH METHODS

This section aims to elaborate multiple research methods for both macro and micro level of
analysis. Chapter five will include the macro analysis of SARA reforms while chapter six will
comprise of micro analysis of tax administration reforms in Pakistan. Both ‘macro’ and ‘micro’
terms refer to their literal meanings with macro meaning large-scale, overall and micro meaning
small or focused. The term ‘macro’ is adopted to denote the broad level of analysis including
many SARA developing countries undertaken to develop a broader picture of SARA
implementation against corruption. The term ‘micro’ is adopted to denote the focused and specific
level of analysis including only Pakistan. Both ‘macro’ and ‘micro’ analyses are utilized not only to
paint a picture of trend so far in SARA developing countries but also examining one case
(Pakistan) specifically for focus. The multiple methods refer to selection of appropriate research
methods at two levels of change, the macro (SARAs) and micro (Pakistan). At the macro level a
detailed mapping of SARA cases will be undertaken for conceptually testing the Anti-Corruption
SARA framework for multiple SARA countries by conducting a systematic review in the form of
analysing secondary literature (published papers, reports and books). At the micro level a
detailed case study of tax administration reforms in Pakistan will be undertaken for conceptually
testing the Anti-Corruption SARA framework for Federal Board of Revenue (FBR) by utilizing a
combination of semi-structured interviews and secondary sources of data.
One of the advantages of conducting two-staged analysis of SARAs contributes towards
increasing the robustness of thesis findings. Although the micro level of analysis, undertaking
case study of tax administration reform in Pakistan, is cross-sectional in nature as data collection
occurred in one point in time; but in terms of macro SARA analysis, the examination of secondary
literature spanned different points in time/stages of reform due to different developing countries
undertaking SARAs at different times, demonstrating a sort of longitudinal dimension of analysis.

98
Hence, undertaking both macro and micro level of analyses of analytical framework, and making
an effort towards a cross-sectional as well longitudinal inquiry of research, increases the
robustness of research findings. Another advantage of two-staged analysis entails combination of
analysis with macro approaching old evidence (secondary literature) in a new way (analytical
framework); while, the micro approaching new evidence (semi-structured interviews) in an old
way (Case Study method). There is an argument which proposes that old evidence makes study
relevant, while new evidence makes study useful for others. Thus, one of the significance of two-
staged analysis involves combining both old evidence (macro) and new evidence (micro) and
making this study not only relevant but also useful for future research.
The sequence and timing of data collection and analysis, like the plan of the thesis, was
determined by the logic of needing one piece of information in order to find another. The first
pieces were found at the macro level of analysis, and the last pieces were found at the micro
level of analysis. In other words, the sequence and timing of data collection and analysis moved
from the macro level of analysis examining SARAs to the micro level of analysis undertaking
case study of tax reforms in Pakistan. There was also reflexivity in that earlier work was
revisited. The literature reviews were continuously updated. The macro and micro level findings
were continuously reflected on to increase the sophistication of interpretation. As illustrated in
figure below, this research on SARAs is Descriptive (as it aims to provide information and facts
about SARAs), Analytical (as it aims to apply analytical framework to literature and data and
through macro and micro analysis also aims to re-organize information (data) for comparing
SARAs, finding patterns against controlling corruption, and hence coming up with categories of
SARAs with differences against corruption), Persuasive (as have taken a position regarding
autonomy and accountability of SARAs, making a claim about interference from MoFs and
recommending to keep MoFs out of accountability arrangements of SARAs) and Critical (as
99
evaluated Taliercio and Klitgaard’s work and entered a debate about SARAs’ capability to control
corruption).
Figure 4.2: The Onion Model of Academic Discourse
Critical
Persuasive
Analytical
Descriptive
Descriptive: Providing information and facts

Analytical: Re-organizing information: applying models to data, comparing, finding patterns and categories
Persuasive: Taking a position, making a claim or recommendation, interpreting, developing an argument
Critical: Evaluating others’ work, entering a debate, considering alternatives- at least 2 positions, including yours
Source: Humphrey, S., Charles, C., Economou, D. & Drury, H., (2015)
The evidence for this research has been collected from both qualitative and quantitative data by
employing primary and secondary sources. The reliance of this thesis is tilted more towards
qualitative than quantitative data because quantitative methods have restricted capability towards
exploring the reasons and perceptions behind failure of SARA reforms towards controlling
corruption in developing countries. Miles and Huberman (1994) highlighted that as qualitative
methods emphasize people’s ‘lived experience’, they are primarily suitable for tracing the
meanings people attribute towards events, processes, and structures in their lives, along with
their understandings, and for linking these understandings to the outer world. McEvoy and
100
Richards (2006) also proposed that qualitative methodology can help explain complex concepts
and relationships that are probable to be captured by prearranged response categories or
standard qualitative measures.
First the research strategy employed for conducting the macro level of SARA analysis is
described below, followed by illustrating the research strategy for conducting micro level of
analysis for FBR, Pakistan.
4.2 METHODS FOR MACRO ANALYSIS OF SARAS
At the macro level a detailed mapping of SARA cases was done for conceptually testing the Anti-
Corruption SARA framework for multiple SARA countries. The macro analysis entailed analyzing
SARA cases which have not performed well against corruption. Although review of literature in
chapter two highlighted quite a few SARAs fitting this purpose, but were geographically dispersed
all around the globe. Since the researcher lacked resources of time and money to carry out in-
depth case studies of selected SARA cases in pursuit of primary data to analyze analytical
framework, the researcher was left with no other choice but to rely on secondary literature to
conduct macro level of analysis. Hence, the macro level of SARA analysis was aimed at
theorizing about what is going wrong with practice by approaching old evidence (secondary
literature on SARAs) in a new way (i.e., analytical framework). Also this research, especially the
macro level of analysis relying heavily on use of secondary literature, is limited by the lack of
publicly available data. Therefore, the macro level of SARA analysis can only be as good as the
quality and quantity of the secondary literature on which it has relied. As is common in such
research, individuals and corporate organisations alike seldom offer information on how to pay
bribes or engage in corruption (Johnston, 2002).

101
The secondary literature was acquired through published documentary sources including books,
journal articles, archived documents, newspapers, government reports, reports by international
donor organizations, reports of NGOs, and reports and information available at official websites
of SARAs and ministries in developing countries to facilitate macro level of analysis. Unavoidably
these sources of data have gaps, as true for any kinds of socially generated data, but they are
also capable to offer insights into how corrupt practices might have developed. The documentary
sources were selected on the basis of criteria of accessibility, comparability and association with
SARAs and Corruption. One of dominant methodology in the field of public administration has
been comparative study for over five decades (Guess, 1989). It has assumed centre stage in the
field of social science and particularly in the field of public administration research. For this
reason, Hood (1989) noted "Today, public administration scholars live in what is much more a
global village conceptually, in that it would be hard to write an acceptable research degree thesis
in the subject today which did not draw on an international literature for its conceptual framework.
It is hard to see this trend going into reverse". Moreover, Bekke et al. (1996) argued,
"comparative analysis is necessary for the identification of key concepts, of relations among
concepts, and of the underlying logic or dynamic of the relations. Comparative research is also
an antidote for the narrowness sometimes associated with studying a single system".
Secondary literature analysis was conducted because they "provide a key source of data on
events or groups too small or scattered or otherwise difficult to trace for national interview
surveys to be a realistic possibility." (Hakim, 1993, p.133). Hakim (1993) highlighted the suitability
of secondary literature "for research on the policy process itself ... Albeit incomplete accounts,
are part of the reality being studied" (p. 134). Content analysis can be particularly beneficial for
conducting cross-sectional analysis requiring frequencies from qualitative or unstructured data
(Carney, 1972; Jupp & Norris, 1993). It can also be beneficial when inferences are to be made
about "what is 'written between the lines' (Carney, 1972, p. 25). But there is limitation for content
102
analysis as it only "gets the answers to the question with which it is supplied." (Carney, 1972, p.
283). Content analysis also presumes the significance of discourse. Some scholars might debate
the ability of discourse towards producing valid results for reason that documentary data only
captures partial accounts of the reality being examined. This is correct, but so is the argument
that documents assume a key source of data on geographically spread events making
interviewing challenging, and they might be incomplete but still remain part of reality being
studied (Hakim, 1993).
The following figure illustrates the process followed to analyse secondary literature for macro
level of SARA analysis (Source: adapted from Hakim, 1993).
Figure 4.3: Systematic Evaluation Process for Macro Level of SARA Analysis
Stage 1: Planning the Analysis
Step 1: Mapping Field of Study
Step 2: Producing an Analysis Protocol
Stage 2: Identifying and Evaluating Studies
Step 3: Conducting a Systematic Search
Step 4: Evaluating Studies
Stage 3: Extracting and Synthesizing Data
Stage 5: Conducting Data Extraction
Step 6: Conducting Data Synthesis
Stage 4: Reporting
Step 7: Reporting the Findings

103
The stage one of planning the analysis in figure above involved mapping the field and becoming
clearer about its boundaries. ‘The technique of mapping is one useful way in which one can begin
to scope the research and become clearer about its boundaries’ (Jenkins, 2003). The step one of
the systematic review involved identification of secondary literature most relevant to this
research, as this thesis integrates three major areas including reformed tax administrations,
changes in autonomy and accountability and developing countries. For the purpose of mapping
the field, the most relevant secondary literature represented the overlap of all three major areas
in the form of SARAs as illustrated in the figure below, such that these literatures were related to
all three domains of tax administrations, autonomy and accountability and developing countries.
As the figure illustrates, SARAs represent a case of reformed tax administrations with changed
autonomy and accountability mechanisms in developing countries. In other words, as this
research aimed to analyse the interaction of reformed tax administrations with autonomy and
accountability in developing countries. Hence, SARAs represented a good example of analysing
an instance when tax administrations have been reformed in developing countries with changed
autonomy and accountability.

104
Figure 4.4: Mapping the Field for Macro Level of SARA Analysis
Not very relevant as Not very relevant as only
only involves tax involves tax administration and
administration and Autonomy and Accountability
Developing Countries
Tax Overlap representing
SARAs as examples of
Not very relevant as
Tax Administration,
only involves Autonomy Administration Autonomy and
and Accountability and Accountability and
Developing Countries Developing Countries
Developing Autonomy
Countries Accountability
In order to produce an analysis protocol (step two in figure 4.4 above) for conducting systematic
review, keywords were devised for running online literature searches for the purpose of mapping
the field and selection of relevant secondary literature. As a step three in figure 4.4 above, three
separate systematic searches were conducted. In the first phase all four keywords i.e., Tax
Administration + Developing Country + Autonomy + Accountability were used for finding
published literature which clearly dealt with or examined all four areas or keywords. In the second
phase, keyword searches were run with only Tax Administration + Developing Country to look for
published literature which dealt with both these areas or keywords. After this step, the relevant
material obtained in the second phase was critically evaluated for applicability or reference to
SARAs and only those studies were included in the analysis showing linkage to SARAs. In the
third phase, the keyword search only used the term ‘SARA’ or ‘semi-autonomous revenue
authority’ or ‘autonomous revenue authority’ to look for published data sources which dealt with
only this keyword. After this step, the relevant material obtained in the third phase was critically
evaluated for material on autonomy and accountability mechanisms after reform and only those
105
studies were included for the systematic review having material for both. In sum, for the purpose
of mapping the field online keyword searches were conducted in three layers following above
scheme.
It is worth highlighting here that only those secondary literature studies were included for the
systematic review in the macro analysis which fulfilled the above mentioned criteria. Special
attention was paid to evaluating studies (step four in figure 4.4) for relevance and only those
studies were included which were related to SARA reform in developing country and had material
on implementation of autonomy and accountability mechanism. For example those secondary
literature studies which had covered implementation of tax administration reform in developed
countries were excluded from macro level of analysis. Also those secondary literature sources,
which were although related to tax administration reforms in developing countries, but had not
covered the topics of autonomy and accountability, were also excluded from macro analysis.
Furthermore, these literature searches lasted the duration of writing up systematic review for
macro analysis (January 2013 to June 2013) and were continually updated over time in pursuit of
new and current literature. The University of Stirling online library, online resource database
JSTOR and Journal indexing services of Sage Journals and Emerald were used for data
collection and extraction. Also steps were undertaken to ensure the quality of selected literature
sources. Only published books, journal articles, government reports, reports by international
donor organizations, reports and information available at official websites of SARAs and
ministries in developing countries were used in this study. For the same reason, when internet
searches as per keywords returned links to documents whose credibility could not be verified,
although they appeared relevant to research topic, they were not selected to be included for
systematic review.
106
After the selection of secondary literature as per the keyword searches described above, each
and every literature source was analysed individually by following the procedure as described
below. This represents the step five of data extraction as highlighted in figure 4.4 above. The
softcopy of every literature source was read in detail by the researcher in pdf format. First, it was
made sure that the literature under review referred to SARA implementation in a developing
country. After this, the whole document was analysed for literature on personnel autonomy
especially for indicators of wages and rewards (sub-hypothesis 1a), probability of detection (sub-
hypothesis 1b), enforcement of penalties (sub-hypothesis 1c), and ethics in tax officials (sub-
hypothesis 1d). Also, the document was also analysed for material on accountability especially
for effective oversight mechanisms (sub-hypothesis 2a), politicization of tax officials (sub-
hypothesis 2b) and discretionary powers of tax officials (sub-hypothesis 2c). In doing so, the
researcher annotated and highlighted every secondary source extensively. Where the softcopy
was not available, especially for books from library, the researcher studied the hard copies
(photocopies) of these sources and highlighted and annotated the source as per procedure
described above. After conducting this time-intensive exercise for each selected secondary
literature, the researcher organized material for macro analysis as per each sub-hypothesis. For
this purpose, separate folders (both in computer and hard files) were made, maintained and
updated by the researcher for the whole duration of writing up macro analysis. For example, one
such folder was labelled ‘Sub-hypothesis1(a) Wages and Rewards’ and included literature on
SARAs having material on personnel autonomy in terms of changes in wages and rewards.
Similar folders (both in computer and hard files) were also maintained for the rest of all six sub-
hypotheses as well. A complete detail of selected secondary literature for each sub-hypothesis is
presented in individual tables in the next chapter while undertaking systematic review under
macro analysis. These tables are not reproduced here in order to avoid repetition. Please refer to
107
pages 130, 136, 142, 148, 155, 161, and 166 for summaries of selected literature sources for
each sub-hypothesis.
Not only the organization but the synthesis and analysis of individual secondary literature sources
(step 6 in figure 4.4 above) was also conducted as per sub-hypotheses. For example, in order to
analyse a particular literature source ‘X’ for sub-hypothesis 1(a), it was analysed if the literature
reported any increase in personnel autonomy in terms of wages and rewards after SARA reform.
If the literature did reported positive increase in wages and rewards, then it was taken to control
for motivations for corruption and supporting sub-hypothesis 1(a). Similar analysis was conducted
for all other six sub hypothesis and findings are reported in the next chapter.
Also individual indicators developed for all sub-hypotheses were also utilized to analyze selected
published sources. Individual indicators relating to each sub-hypothesis have already been
identified and discussed in chapter three (See Sections 3.1.3, 3.1.4). As a next step, each
indicator was utilized to evaluate the selected secondary literature source. In essence, the
combination of all indicators was used for conducting systematic review for analysing personnel
autonomy and effective accountability for controlling motivations and opportunities for corruption
in SARAs through macro level of SARA analysis.
As a next step detailed descriptive information in the form of key findings and results was
extracted from all selected secondary literature sources.This time-intensive exercise led to
identification of SARA countries with most availability of secondary literature. So, based upon the
logic of convenience sampling, those SARA countries with most number of secondary literatures
appear in the next chapter for macro analysis of SARA cases. The SARA countries included for
108
systematic review include Peru, Ghana, South Africa, Tanzania, Mexico, Guatemala, Uganda,
Kenya and Venezuela as highlighted in table 5.18 on page 176.
One of the limitations of acquiring secondary literature as per above procedure remains that
researcher has no control over which SARA countries have been selected for macro analysis.
Only those countries, for which better quantity and quality of secondary literature was available,
make up the cases for macro analysis. Since the researcher lacked resources of time and money
to carry out in-depth case studies of multiple SARA cases in pursuit of primary data to analyze
analytical framework, the researcher was left with no other choice but to rely on secondary
literature to conduct macro level of analysis.
4.3 METHODS FOR MICRO ANALYSIS OF PAKISTAN
To explore why tax administration reform failed to effectively control opportunities for corruption in
FBR Pakistan, Chapter six aims to conceptually test the Anti-Corruption SARA framework for
FBR, by conducting a detailed case study of tax administration reforms in Pakistan. The term
‘micro’ is adopted to denote the focused and specific level of analysis including only Pakistan, as
this proposed analysis will enable to examine in detail the current state of affairs and potential
and suitability of SARA reform for Pakistan. In this section the purpose of case study work and
the choice of research method will be discussed. Case study of tax administration reform was
used in this thesis to demonstrate and explain the reasons and extent of corruption weakening
Pakistan’s progress against optimum tax collection over years. There are a number of researches
which highlight the applicability and benefits of case study method in social science research
(Yin, 1981, 1984, 2003; Saunders, et. al., 2003). Two main sources of data were employed for
the case study at the micro level of analysis: document analysis (secondary literature) and semi-
structured interviews. The triangulated methodology collects two types of data (documentary and
109
interview) for cross-checking. Hence, the micro level of analysis for FBR was aimed at theorizing
what is going wrong with practice by approaching both new and old evidence (semi-structured
interviews and secondary literature) in a new way (i.e., analytical framework). The following
paragraphs will elaborate them further.
4.3.1 Interviews
The primary data source for case study on tax administration reforms in Pakistan was generated
by using one-to-one semi-structured interviews for obtaining an understanding of the social world
occupied by social actors, because corruption in tax administration is a social practice committed
by and concerning social actors within a specific social context. The interviews were conducted to
gather tax officials’ perception about reasons of failure of tax administration reforms towards
controlling opportunities for corruption in Pakistan. One of the challenges of this research has
been to stimulate the interpretations and understandings of the actors through interviews in order
to improve our understanding of why tax administration reforms failed to took off and control
corruption in Pakistan. One of the useful inquiry techniques involves observing behaviour and
can be used as a short-cut in search of answers to the research question by directly asking
people what is going on (Robson, 2002). Face-to-face interviews offer numerous advantages
over postal or self-administered questionnaires by offering the option of adjusting one's line of
query and following up interesting responses with more questions for aiding examination of
underlying motives. In order to gather the interpretations of individuals involved in tax
administration reforms, it was argued that semi-structured interviews would be one of the most
appropriate research techniques for gathering data. There are many research methods textbooks
which have detailed three main types of interviews involving structured, semi-structured and
unstructured interviews (Yin, 1994; Morris & Wood, 1991; Robson, 2002; Saunders, et al., 2003).
This technique permits the researcher to examine some of the primary motives more directly for
110
bringing out and undertaking ostensible contradictions (Norton et al, 2004). Semi-structured
interviews permitted the researcher to examine "themes", "trails" and "patterns" (Pettigrew, 1990,
p. 277).
In order to conduct semi-structured interviews of respondents, interview schedule was developed
in this study. This interview schedule was based on the hypotheses developed in the analytical
framework. The procedure of developing interview schedule is detailed below. The analytical
framework resulted in development of two research hypotheses and seven sub-hypotheses. In
order to analyze the validity of the framework through micro level of analysis, the forthcoming
analysis in chapters six will be examined in terms of these sub-hypotheses. Consequently the
development of the interview schedule was also undertaken with these sub-hypotheses in mind.
Hence, the interview schedule consists of seven broad sections with each one referring to one
specific preventive strategy to separately control motivations or opportunities for corruption in tax
administrations (Appendix 1). Individual indicators relating to each sub-proposition have already
been identified and discussed in chapter three (Sections 3.1.3, 3.1.4). As a next step, each
indicator was utilized to develop individual questions of the interview schedule for testing sub-
hypotheses. The combination of all questions made up the instrument of semi-structured
interview schedule for testing personnel autonomy and effective accountability for controlling
motivations and opportunities for corruption in FBR Pakistan through micro level of analysis. All
sub-hypotheses for the micro level of analysis, along with indicators and relevant interview
questions are illustrated in table 4.1 below.

111
Table 4.1 Overview of Sub-Hypotheses along with Indicators and Interview Schedule
Questions for Micro Level of Analysis
Sub-Hypotheses Indicators Interview Questions

Research hypothesis 1(a) proposes that by Increases in wages for tax officials and Q1, Q2, Q3
increases in wages and rewards, SARAs have utilization of performance-linked bonuses
been effective against motivations but not
opportunities for corruption.
Research hypothesis 1(b) proposes that by Increases in the quality and frequency of Q4, Q5, Q6, Q7, Q8, Q9
increases in the probability of detection, SARAs internal audits and probes by establishing
have been effective against motivations but not internal audit and anti-corruption divisions;
opportunities for corruption. Increases in the number of supervisory
personnel assigned to internal audit and anti-
corruption divisions;
statutory condition of declaration of assets by all
employees
Research hypothesis 1(c) proposes that by Increases in imposition of high levels of Q10, Q11
increases in and stricter enforcement of penalties penalties including monetary sanctions, job
for corruption, SARAs have been effective against dismissals and prison sentences.
motivations but not opportunities for corruption.
Research hypothesis 1(d) proposes that by Increases in ethics training and adoption of Q12, Q13, Q14, Q15
instilling ethics in tax officials, SARAs have been code of conduct.
effective against motivations but not opportunities
for corruption.
Research hypothesis 2(a) proposes that SARAs Effective accountability relationships with all Q16, Q17
have been ineffective in controlling opportunities supervising bodies including Ministry of
for corruption due to ineffective accountability by Finance, Revenue Boards/Superintendents
MoF by not introducing effective oversight (whichever applicable), and Parliament etc.
mechanisms.
Research hypothesis 2(b) proposes that SARAs Effective accountability relationships with all Q18, Q19
have been ineffective in controlling opportunities supervising bodies especially Revenue
for corruption due to ineffective accountability by Boards/Superintendents (whichever applicable),
MoF by not de-politicizing tax officials. and Parliament etc., in addition to Ministry of
Finance.
Research hypothesis 2(c) proposes that SARAs Balancing effective accountability relationships Q20, Q21
have been ineffective in controlling opportunities between all supervising bodies especially
for corruption due to ineffective accountability by between Ministry of Finance and other
MoF by not reducing discretionary powers of tax supervising bodies including Revenue
officials. Boards/Superintendents (whichever applicable)
and Parliament etc.
The researcher visited Pakistan to conduct interviews at FBR. The interviews were conducted
from 15th September 2009 to 20th December 2009 and March 2016 to June 2016. The researcher
conducted interviews with a number of tax officials, journalists, MoF officials, taxpayers and
businessmen/traders and relied on the opinions of interviewees who highlighted their own
personal experiences and perceptions, leading to some significant findings. In this study limited
numbers of people were interviewed (31 semi-structured interviews) owing to problems of access
and duration of fieldwork. 50 interviews were originally proposed but because of the sensitive
nature of the theme of the research, the researcher faced considerable constraints in the
112
willingness of respondents to be interviewed, which limited the total number of semi-structured
interviews to 31. Due to this reason, this study does not claim to offer a comprehensive account,
rather only providing some illustrations from a limited number of cases, representing the views of
some respondents about the main topic of this thesis. The views of interviewees were sought on
their perceptions about effectiveness of tax administration reform towards controlling corruption in
FBR. The researcher was of the opinion that these interviewees were actors who have been
critics of practice of corruption in FBR Pakistan.
Since this case study has been qualitative in nature exploring a sensitive topic, more emphasis
was given to acquiring a purposive sample of people, occupying key positions in relation to
research topic and willing to participate in research by giving interview. For this reason and
mainly due to the sensitivity of the research topic, this research cannot claim to have acquired a
representative sample of people rather purposive sampling is utilized to acquire interview data.
As Patton (1990) puts it, "there are no rules for sample size in qualitative inquiry", therefore it is
the availability and willingness of the interviewees and the aim of the research that determine the
size of the sample. The interviewees selected for the interviews not only belonged to FBR but
also included, journalists, MoF officials, taxpayers and businessmen/traders, hence following a
multi-stakeholder approach (Pettigrew, et al., 1992, p. 301).
All employees in FBR can be categorized into two categories: gazetted (Grade 17 or above) or
non- gazetted officers (Grade 1 till 16). There are total 22 grades for employees with clerical staff
only going up to grade 16 in FBR. This research only targeted the population of gazetted officials
in FBR (Grade 17 or above) either because in terms of seniority they occupied a lead position in
FBR, or because of their direct involvement with the reform process. There were 1500 such
officers in FBR (460 Customs officers, 1040 Inland Revenue Service officers) at the time of
113
writing this thesis (Personal communication with HR office, FBR). Out of this, 468 were working in
both Lahore and Islamabad). Tax officials belonged to different functional groupings within FBR
and occupied different hierarchical levels.
Table 4.2 lists the respondents who were interviewed from FBR.
Table 4.2 List of Tax Officials Interviewed
Respondent Civil Service Group Seniority* Region

Tax Official 1 Customs Group Junior Level Provincial Level
Tax Official 2 Inland Revenue Service Junior Level Provincial Level
Tax Official 3 Customs Group Mid-Senior Level Federal Level
Tax Official 4 Customs Group Mid-Senior Level Federal Level
Tax Official 5 Inland Revenue Service Mid-Senior Level Provincial Level
Tax Official 6 Inland Revenue Service Senior Level Federal Level
Tax Official 7 Customs Group Mid-Senior Level Provincial Level
Tax Official 10 Customs Group Senior Level Federal Level
Tax Official 12 Inland Revenue Service Junior Level Provincial Level
Tax Official 14 Inland Revenue Service Mid-Senior Level Federal Level
Tax Official 15 Inland Revenue Service Mid-Senior Level Federal Level
Source: Generated from Fieldwork
*Junior Level = Grade 17
Mid-Senior Level = Grade 18
Senior Level = Grade 19 and up
The interviewees from FBR represented not only different civil service groups in FBR including
Customs and Inland Revenue Service, but also varied in terms of seniority in organizational
hierarchy. These differences in groups and seniority were aimed to obtain a broader view on
research topic. As highlighted in table above, most interview respondents (11/16) belonged to the
newly integrated Inland Revenue Service. In addition, most of the interview respondents (9/16)
belonged to the mid senior level in FBR. The designations of Interviewees cannot be revealed to
protect their anonymity and was one of the conditions on which they agreed to be interviewed.
114
Some of the respondents were of the view that if the designations along with regional postings
would be revealed, it would be possible to identify the interviewees. All respondents belonged to
two cities of Pakistan including Lahore and Islamabad. The tax officials in Lahore belonged to the
regional office of FBR in Lahore, while the tax officials in Islamabad belonged to FBR
headquarters. Islamabad was chosen because the headquarters of FBR is based in the capital.
Lahore was chosen because it is the capital city of largest province Punjab and other than the
FBR headquarters in Islamabad, most tax officials are based in this city, helping with their
access. Also Lahore was chosen because the researcher resided here and had access to tax
officials in Lahore. Hence, all tax officials not belonging to Lahore or Islamabad were excluded
from being interviewed, hence the purposive sample.
As for the selection of tax officials in Lahore and Islamabad, the researcher had no choice but to
interview those tax officials who accepted the request for interview thus agreeing to be
interviewed. For this reason, there is self-selection bias in the sample of respondents which
means that those respondents who feel strongly or against corruption are over-represented in
this study. In order to take care of this bias, the researcher was very interested in interviewing
certain tax officials who had reputation/ record of possibly being involved in some corrupt
practices (newspaper sources) but those declined to be interviewed. The researcher would
forward request for interview to different tax officials at FBR offices through calls and official
emails. Access to first of the interviewees was assisted through mid-senior level tax officials
(also friends of researcher) at FBR headquarters in Islamabad, who assured their colleagues
about confidentiality and promise of anonymity of the interviewees. The interviewees varied on
the issue of cooperativeness. Although quite a few were willing, but many declined to be
interviewed. In total nineteen tax officials declined requests for interview. Mostly the reason of
decline was suggested to be busyness due the official work but the researcher believes that the
115
sensitivity of the topic of thesis also played its part behinds interview refusals. In addition, some
officials accepted the researcher's offer of an interview, but discussed the issue off-the-record
(meaning not tape-recorded). Due to the sensitivity of the information sought, some tax officials
were not willing to volunteer information about reasons and extent of corruption in FBR. Many
interview respondents declined to be interviewed on grounds that they do not have any valuable
information to share on the topic. Interestingly, many of the ones who declined for interviews
themselves, were enthusiastic in referring the researcher to another colleague of theirs who in
their view might have some pertinent information on the topic. Some were even able to make
reference calls and arrange meeting for interviews for the researcher, hence assuming a snow-
ball form of sampling strategy for selection of interview respondents at FBR. As a result of
exposure to this practice, later in the field work, the researcher often asked the willing
interviewees for references and calls to other colleagues who in their opinion might agree to be
interviewed. Many tax officials were hesitant to be interviewed owing to reservations of the
political implications of the subject of the interview. For this reason some respondents were very
guarded in their responses and only presented restrictive and official, rather than truthful
responses. This group of respondents were frank enough to tell the researcher that they only
agreed to be interviewed because of the recommendation of their colleague and were not
comfortable with the researcher taping the interview over fear of voices being recorded. In such
cases, the researcher abandoned taping the interviews, and only took notes. It is worthwhile to
highlight here that these ‘off-the-record’ interviews resulted in many interesting findings.
Negotiating access to the very senior tax officials proved particularly problematic because of busy
schedules on account of official and administrative responsibilities. Despite numerous efforts to
obtain interview appointments in the form of visits, letters and telephone calls to seek access to
highest level of tax officials, access was not given. For this reason interviews of tax officials of
116
very high level such as Members of the Board and Chairman FBR and high level officials in MoF
such as Secretary Finance and Minister of Finance could not be conducted on account of access
issues. On one occasion, the researcher was able to fix appointment for interview over phone
with one of the members of the board in FBR headquarters in Islamabad. Upon arrival from
Lahore to Islamabad for the interview, after brief interaction, the member rather than giving the
interview, directed the researcher to another high-ranking official in FBR for interview, for him
being more knowledgeable about ongoing tax administration reforms in FBR. This official who
was delegated with the responsibility of giving interview did not do so, rather promised to share a
study on corruption in FBR with the researcher over email. This promise was not kept up despite
repeated reminders over email.
In addition to tax officials at FBR, the researcher also conducted interviews with journalists, MoF
officials, taxpayers and businessmen/traders. In total 15 such interviews were conducted by the
researcher during March 2016 to June 2016. These interviews proved to be very beneficial as
they offered diverse opinion of multiple stakeholders towards tax reforms in Pakistan and its
effect on corruption. They helped to illuminate views of people other than tax officials relating to
reform. In detail two journalists, three MoF officials, five taxpayers and five businessmen/traders
were interviewed in the second stage interviews. The two journalists included Rehman Azhar
(Dunya News) and Imdad Bhatti (Jang News). The three MoF officials were gazetted officers and
occupied mid-level positions in Ministry of Finance in Islamabad, Pakistan. They were promised
anonymity in exchange for interview. The five taxpayers included friends of researcher and were
chosen because the researcher was certain that they actually paid taxes and were aware with tax
reforms in Pakistan. Their identities will also be kept anonymous. The five businessmen
represented prominent traders and were contacted for interview through a contact working at
Lahore Chambers of Commerce and Industry. They also promised to be interviewed in exchange
117
for anonymity. The second stage of interviews was conducted to address the criticism that only
interviewing tax officials might not fully represent the diverse opinion of multiple stakeholders to
tax reform in Pakistan and it was considered important to also include their understandings of tax
reform and its effect on corruption. All interviewees at the second stage, except of three
interviewees from MoF Islamabad, were based in Lahore, which helped the researcher with
access.
The preliminary contact with every interviewee was established by an official letter asking for their
support, in addition to highlighting a brief description of the study. This introductory letter was
also supported by other relevant documents including support letter provided by the supervisor,
registration certificate of the university in UK, identification card supporting evidence of affiliation
with the university in Pakistan for confirming academic status. This initial contact was followed-up
with telephone calls, e-mail prompts and personal visits to their offices to fix appointments for
interviews. A positive rapport with the interviewees was developed by being sensitive to the
occasion in terms of punctuality and politeness, ensuring the confidentiality of the interview such
that their names and designations will not be revealed, being empathetic by pairing active
listening with probing questions, and awareness of busy schedules of tax officials resulting in
different duration of interviews. The questions in Appendix 1 functioned as a guide for interview
and provided a framework. There were instances when the interviewer was not able to cover all
of the questions in the interview schedule, during the allocated time for the interview, on account
of workload and busy schedule of tax officials. In these instances, interviews were disrupted due
to unscheduled meetings of tax officials, but many times interviewer was able to wait for the tax
official to return and finish the interview. Due to the sensitive nature of the thesis topic, the
researcher promised complete anonymity to the interviewers. Hence, this thesis does not
disclose either the names or designations of interviewed tax officials and only uses coded names
118
in place of names/designations of interviewees (see Table 6.1). The interviews were only taped
when it was permitted and explicitly agreed by the respondent. Otherwise off-the-record un-taped
interviews were only noted by the researcher during and immediately after the discussions in the
form of detailed notes.
4.3.2 Document Analysis
In addition to the semi-structured interviews, the micro level of analysis was also supplemented
by the documentary analysis of secondary literature. Document collection was conducted in both
UK and Pakistan. The case study in the form of document analysis involved analysing tax
reforms in Pakistan over three years period ( 2011-2013). Since the researcher was mainly
based in UK for the duration of study, it was decided that newspaper analysis could provide
researcher with important and latest news related to reform progress and corruption. These
important news could then be verified from FBR for further information. The researcher had
regular internet access, which made it possible to access epapers of newspapers in Pakistan.
After discussion with the supervisor, two newspapers, ‘DAWN News (www.epaper.dawn.com)’
and ‘The News (www.e.thenews.com.pk)’ were selected for the analysis and analysed for three
years period 2011-2013. Both these newspapers were in English language and represent mostly
politically neutral instance. In other words, both these newspapers were not strongly aligned with
any political party because doing so could bias news coverage. After this the researcher began
with a very time intensive and disciplined routine of reading both newspapers every day.
Although initially this routine seemed to be very time-consuming but later this was compensated
with the quantity and quality of secondary literature generated for micro analysis. While
conducting micro analysis for sub-hypotheses, chapter six makes reference to these newspaper
sources when they are used. For example, the newspaper analysis enabled the researcher to
find out that FBR chairmen were very abruptly transferred after the reform along with the dates of
119
such transfers, the duration of every chairman, and possible reasons of transfer. This information
had important bearing for the micro analysis and is used in chapter six, section 6.3.1. Newspaper
analysis enabled the researcher to keep a very keen eye on regular issues relating to FBR in
Pakistan especially coverage on corruption cases, despite not being in Pakistan. Every interested
piece of news/article was printed by the researcher and kept in folders for future analysis. The
researcher maintained seven such folders, with each one for individual sub-hypotheses. At the
time of writing micro analysis, the researcher was able to refer to secondary literature in each
folder for analysing individual sub-hypotheses along with interview data. At the conclusion of
interviews, the notes were transcribed by the researcher and interview data was analysed as per
individual sub-hypotheses. The indicators for each sub-hypothesis were used to identify the
themes in the interview material generated. These identified themes served to highlight the most
relevant quotes from the interview material for each sub-hypotheses and they were reproduced
as such while writing up micro analysis for Pakistan. Many times, both literature from newspaper
and interview data supplemented each other in support of sub-hypotheses, thus serving the
purpose of cross-checking under triangulated methodology. The analysis over three years was
able to generate quite rich and detailed record of FBR performance against corruption as
reported in newspapers. This greatly supplemented the interview data for micro analysis. The
researcher was surprised with the quantity and depth of secondary literature generated though
newspaper analysis.
In addition to newspaper analysis, documentary evidence was also obtained from FBR
Headquarters; international donor organisations such as the World Bank, IMF, ADB; NGOs such
as Transparency International; ministerial and non-ministerial reports. Internet searches with the
help of key words were also conducted for micro level of analysis. Specifically, keywords of
Pakistan + tax reform + corruption were used to search for relevant documents on internet
120
regularly over study period. The documents obtained through these searches were analysed for
quality and only those documents coming from credible sources were utilized for case study on
Pakistan. A considerable amount of information was also gathered from the official website of
FBR. It often contained information about progress on cases against corruption. The data
collected from these secondary sources was used to supplement the semi-structured interviews
for the case study of tax administration reform in FBR Pakistan. Some of the issues discussed in
this thesis were based on reports of international donor organizations readily available from the
internet. Their acquisition proved quite straightforward. But same cannot be said about obtaining
internal documents and reports of FBR, such as cases of pending inquiry on account of
corruption charges against tax officials. The secrecy and lack of access to reports and
documents in FBR constrained the comprehensive micro level of analysis examining progress of
FBR towards corruption. Nonetheless, some of the very cooperative interviewees helped
immensely with the collection of documents from FBR. This included some documents which
were initially declined to be shared by the concerned department in FBR over questions about
their usage by the researcher. A couple of interview respondents also emailed the researcher
some documents which were not available straightforward. In a nutshell, the combination of
published work along with newspaper analysis provided very useful information and
supplemented interview data for micro study on Pakistan. A list of selected documents used for
micro analysis is presented next in table 4.3.

121
Table 4.3 List of Selected Secondary literature for Macro Analysis of Pakistan
1) ADB (2008) Implementing the Anti-Corruption Action Plan for Asia Pacific: Pakistan.
Manila: ADB
2) ADB (2013) Key Indicators Pakistan. Manila: ADB
3) ADB OECD (2006) Anti-Corruption Policies in Asia and the Pacific. Manila: ADB
4) ADB/OECD (2006) Combating Corruption in Asia Pacific. Manila: ADB
5) Cheema (2013) Representation without Taxation. Centre for Investigative reporting in
Pakistan.
6) House of Commons (2013) Aid Tax Pakistan. House of Commons London
7) IMF (2010) Letter of Intent. Washington DC: IMF
8) IMF (2010) Pakistan Fourth Review. Washington DC: IMF
9) IMF (2012) Article IV Consultation Pakistan. Washington DC: IMF
10) IMF( 2013) Article Iv Consultation And Request For An Extended Arrangement Under
The Extended Fund Facility. Washington DC: IMF
11) Martinez-Vezquez (2006) Pakistan federal tax assessment
12) Shah (2007) World Bank Performance Accountability and Combating Corruption
13) World Bank (2003) Integrated Safeguards data sheet tax. Washington DC: World Bank
14) World Bank (2004) Tax Administration Reform Project. Washington DC: World Bank
15) World Bank (2004) Tax Project Information Project. Washington DC: World Bank
16) World Bank (2009) Pakistan Tax Policy Report Full. Washington DC: World Bank
This chapter explained how methodology depends upon systematic review of previously
published studies, collection of documents, newspaper analysis and collection of primary
interview data with key respondents in order to collate and organise evidence and test validity of
the analytical framework. In order to document corruption in SARAs, and to specifically explore
why some SARAs have failed to effectively control opportunities for corruption, a detailed
mapping of SARA cases will be undertaken in chapter five for conceptually testing the Anti-
Corruption SARA framework for multiple SARA countries. Secondary literature will be employed
for conducting a systematic review of SARA cases at the macro level of analysis. This level of
analysis is labeled as macro to denote the broad level of analysis including many SARA
developing countries undertaken to develop a broader picture of SARA implementation against
corruption. To explore why tax administration reform failed to effectively control opportunities for
corruption in FBR Pakistan, Chapter six aims to conceptually test the Anti-Corruption SARA
122
framework for FBR, by conducting a detailed case study of tax administration reforms in
Pakistan. Two main sources of data will be employed for the case study at the micro level of
analysis: document analysis (secondary literature) and semi-structured interviews. In a nutshell,
the specific contribution of chapter four to the overall thesis argument lies in outlining the
methodology to be used for conducting analyses in chapters five and six. .
Next, a detailed mapping of SARA cases is undertaken for conceptually testing the Anti-
Corruption SARA framework for multiple SARA countries.

123
CHAPTER FIVE
MACRO LEVEL OF ANALYSIS: SYSTEMATIC REVIEW OF SARAs AGAINST
CONTROLLING CORRUPTION
5.0 INTRODUCTION
This chapter intends to apply the analytical framework to selected developing countries in terms
of SARAs’ effect on corruption. To inquire why some SARAs have failed to effectively control
opportunities for corruption, a detailed mapping of SARA cases is undertaken in this chapter for
conceptually testing the Anti-Corruption SARA framework. In other words, it aims to test the
framework by conceptualizing how individual SARA design components can be utilized to
separately control motivations and opportunities for corruption. By doing so, this chapter intends
to gauge the robustness and generalizability of the analytical framework by analysing different
SARA countries under a same lens. As highlighted before, this level of analysis is labeled as
macro to denote the broad level of analysis including many SARA developing countries
undertaken to develop a broader picture of SARA implementation against corruption.
Secondary literature was used to analyse the analytical framework for SARAs at the macro level
of analysis. A detailed discussion about how secondary literature was searched for and selected
has already been highlighted in chapter four, section 4.2. It highlighted how keyword searches
were used to search for and select each secondary literature source. It also highlighted which
secondary literature was finally selected and used for doing analysis for each sub-hypothesis.
Next, it is detailed how the analytical framework was applied to analyse and compare literature
sources. Each selected literature was reinterpreted as per sub-hypotheses in the analytical
framework. For example, in order to analyse a particular literature source ‘X’ for sub-hypothesis
124
1(a), it was analysed if the literature reported any increase in personnel autonomy in terms of
wages and rewards after SARA reform. If the literature did reported positive increase in wages
and rewards, then it was taken to control for motivations for corruption and supporting sub-
hypothesis 1(a). Similarly, in order to analyse a particular literature source ‘Y’ for sub-hypothesis
2(a), it was analysed if the literature reported ineffectiveness of oversight mechanisms due to
interference from MoF after SARA reform. If the literature did report this, then it was taken as a
failure to effectively control opportunities for corruption, thus supporting sub-hypothesis 2(a).
Similar analysis was conducted for all other sub-hypothesis as follows. Every document was
analysed for literature on personnel autonomy especially for indicators of wages and rewards
(sub-hypothesis 1a), probability of detection (sub-hypothesis 1b), enforcement of penalties (sub-
hypothesis 1c), and ethics in tax officials (sub-hypothesis 1d). Also, the documents were also
analysed for material on accountability especially for effective oversight mechanisms (sub-
hypothesis 2a), politicization of tax officials (sub-hypothesis 2b) and discretionary powers of tax
officials (sub-hypothesis 2c). Also analysis for sub-hypotheses was undertaken with an aim to
test research hypotheses 1 and 2, and the findings as per analysis of these sub-hypotheses will
mean to respond to both research hypothesis 1 and 2.
Also individual indicators developed for all sub-hypotheses were also utilized to analyze selected
published sources. Individual indicators relating to each sub-hypothesis have already been
identified and discussed in chapter three (See Sections 3.1.3, 3.1.4). As a next step, each
indicator was utilized to evaluate the selected secondary literature source. In essence, the
combination of all indicators was used for conducting systematic review for analysing personnel
autonomy and effective accountability for controlling motivations and opportunities for corruption
in SARAs through macro level of SARA analysis. Every literature source was interrogated as
above to discover whether, for each country specified, specific anti-corruption measures have or
125
have not been taken. As a next step detailed descriptive information in the form of key findings
and results was extracted and compared from all selected secondary literature sources and
presented in this chapter.
In spite of spread of SARAs in developing countries, there has been limited comparative
analytical work on their implementation against corruption. As a result there is dearth of
agreement and evidence on best practice in organizational design of SARAs. It is this gap this
chapter aims to contribute towards filling by analyzing analytical framework for SARAs. This
chapter aims to explain why certain SARA design components are more effective against
corruption than others, and will offer specific recommendations to improve SARAs’ capability
against corruption. There is a need to analyse all the localized disjointed SARA cases in
developing countries from a transnational angle as per analytical framework, hence the proposed
macro level of SARA analysis. This is to contribute towards filling a gap in literature for studies on
SARAs differentiating between motivations and opportunities for corruption.
One of the significance of macro level of SARA analysis in this chapter lies in the observation that
it aims to analyze secondary literature on SARAs under a new lens and for a different purpose
than originally envisaged. Originally, these literature studies were conducted to evaluate SARAs’
progress against improvement in revenue performance by improving autonomy and
accountability mechanisms. The macro analysis in this chapter aims to extend the use of this
literature studies, not originally done to evaluate SARAs progress against corruption, for
specifically evaluating balance of autonomy and accountability mechanisms under SARAs
reforms and its effect on controlling motivations and opportunities for corruption as per research
hypotheses. In addition, the macro analysis also includes literature studies (though limited in
number) which have tried to evaluate SARAs’ progress against corruption. The analysis in this
126
chapter aims to further develop these studies by segregating progress against corruption into
separately controlling motivations and opportunities for corruption.
The nature of research hypotheses developed in the analytical framework enabled them to be
analysed appropriately through a logical use of secondary literature. The nature of research
hypotheses is such that secondary literature, initially developed to analyze autonomy and
accountability and not corruption in SARAs, can be analyzed towards examining if motivations
and opportunities for corruption have been controlled. Specifically, with the help of sub-
hypotheses, the macro analysis also aims to segregate SARAs’ progress against corruption
towards separately controlling motivations and opportunities for corruption. So the contribution of
macro analysis lies in utilizing previous secondary literature on SARAs by expanding
interpretation of previous findings about autonomy and accountability towards controlling
motivations and opportunities for corruption in SARAs.
By analysing the experience of SARAs, this analysis aims to highlight pre-conditions necessary
for making this reform suitable to a developing country context. Taking the form of findings, this
chapter will help to identify factors which might be present in a specific developing country to be
considered a suitable candidate for SARA reform. The achievement of this objective should result
in identification of contextual factors, which could be analysed in advance to SARA reform,
making a developing country suitable or unsuitable candidate for this reform.
Table 5.1 lists all SARAs which have been established in developing countries so far.
127
Table 5.1: List of SARAs in Developing Countries (arranged in alphabetical order)
SARA Year
Name of SARA
Country Established
Argentina General Tax Directorate (GTD) 1988
Bolivia National Service of Internal Revenues (NSIR) 2001
Botswana Botswana Unified Revenue Service (BURS) 2004
Colombia Directorate of National Taxes (DNT) 1991
Ecuador Servicio de Rentas Internas (SRI) 1999
Ethiopia Ethiopian Revenues and Customs Authority (ERCA) 2008
Ghana National Revenue Secretariat (NRS) 1986
Guatemala Superintendencia de Administración Tributaria (SAT) 1999
Guyana Guyana Revenue Authority (GRA) 2000
Jamaica Tax Administration Jamaica (TAJ) 2011
Kenya Kenya Revenue Authority (KRA) 1996
Lesotho Lesotho Revenue Authority (LRA) 2003
Malawi Malawi Revenue Authority (MRA) 2000
Malaysia Inland Revenue Board of Malaysia (IRBM) 1996
Mexico Mexican Tax Administration Service (SAT) 1997
Mozambique Mozambique Revenue Authority (MRA) 2006
Peru Superintendencia Nacional de Administración Tributaria (SUNAT) 1991
Rwanda Rwanda Revenue Authority (RRA) 2000
Sierra Leone National Revenue Authority (NRA) 2002
Singapore Inland Revenue Authority of Singapore (IRAS) 1992
South Africa South African Revenue Service (SARS) 1997
Spain Agencia Estatal de Administracion Tributaria (AEAT) 1992
Tanzania Tanzania Revenue Authority (TRA) 1996
Uganda Uganda Revenue Authority (URA) 1992
Venezuela National Integrated Tax Administration Service (SENIAT) 1993
Zambia Zambia Revenue Authority (ZRA) 1994
Zimbabwe Zimbabwe Revenue Authority (ZIMRA) 2001
The rest of this chapter is organized as follows: Sections 5.1 and 5.2 will contain macro level of
SARA analysis, where individual sub-hypotheses will be analyzed for selected SARA countries
as per analytical framework. Section 5.3 will summarize the key findings arising out of macro
level of SARA analysis and concludes. This section will also integrate the findings of all sub-
hypotheses for the two research hypotheses of this study. The structure of Chapter 5 is shown in
Figure 5.1.
128
5.0
INTRODUCTION
5.1 5.2
PERSONNEL AUTONOMY CONTROLLING EFFECTIVE ACCOUNTABILITY CONTROLLING
MOTIVATIONS FOR CORRUPTION IN SARAs OPPORTUNITIES FOR CORRUPTION IN SARAs
5.1.1 5.2.1
Personnel Autonomy Controlling Motivations Effective Accountability Controlling
for Corruption by Increases in Wages and Opportunities for Corruption by Introducing
Rewards Effective Oversight Mechanisms
5.1.2 5.2.2
for Corruption by Increases in the Probability of Opportunities for Corruption by De-politicizing
Detection Tax Officials
5.1.3 5.2.3
for Corruption Increases in and Stricter Opportunities for Corruption by Reducing
Enforcement of Penalties for Corruption Discretionary Powers of Tax Officials
5.1.4 5.2.4
Personnel Autonomy Controlling Motivations Synthesis of Analysis for Research Hypothesis
for Corruption by Instilling Ethics in Tax 2
Officials
5.1.5
Synthesis of Analysis for Research Hypothesis
1
5.3
129
5.1 PERSONNEL AUTONOMY CONTROLLING MOTIVATIONS FOR CORRUPTION IN
SARAs
In this study, SARA design component of personnel autonomy is conceptualized to control
motivations for corruption by 1) increases in wages and rewards; 2) increases in the probability of
detection; 3) increases in and stricter enforcement of penalties for corruption; and 4) instilling
ethics in tax officials. Research hypothesis 1 proposed that by preferring personnel autonomy
over effective accountability, SARAs have been effective against motivations but not
opportunities for corruption. The macro level of SARA analysis in this section is aimed to analyse
research hypothesis 1. It comprises of four sub-hypotheses (1a, 1b, 1c, and 1d), with each one
referring to one specific preventive strategy to control motivations for corruption. First all four sub-
hypotheses will be analyzed and then these analyses will be integrated for the cumulative
analysis of research hypothesis 1 in sections below.
The next sections examine how SARAs used personnel autonomy to control motivations for
corruption. The overall proposition is that those SARAs making most progress in ensuring
personnel autonomy should also be the ones making greatest progress towards controlling
5.1.1 Personnel Autonomy Controlling Motivations for Corruption by Increases in Wages
and Rewards
Research hypothesis 1(a) proposed that by increases in wages and rewards, SARAs have been
effective against motivations but not opportunities for corruption.This section aims to identify and
analyze secondary literature on SARAs with particular focus on if and how personnel autonomy is
being utilized by SARAs to control motivations for corruption by increasing wages and rewards
for tax officials. The development of research hypothesis 1(a) in the analytical framework enabled
130
to propose that if SARAs have made progress in terms of personnel autonomy by resulting in
increases in wages and rewards, then it can also be taken to propose that motivations for
corruption have been controlled as an intended but unrecognized consequence. Table 5.2
illustrates an overview of secondary literature analysed to examine research hypothesis 1(a).
Table 5.2: Overview of Secondary Literature concerning Personnel Autonomy and

Increases in Wages and Rewards for SARAs
Selected Literature concerning Personnel Autonomy and Increases in Wages and Rewards
for SARAs
World Bank, 2001; 2004a Clark & Wood, 2001
Taliercio, 2004 Jenkins, 1994; 1995
Fjeldstad, 2002; 2003; 2005a Hall & Jenkins, 1995
Fjeldstad et al., 2003 Devas et al., 2001
Haltiwanger & Singh, 1999 Hlophe & Friedman, 2002
de Merode & Thomas, 1994 Durand et al., 1998
Chand & Moene, 1997; 1999 Therkildsen, 2004
Terkper, 1994 Fjeldstad & Tungodden, 2003
One of the objectives behind creation of Superintendencia Nacional de Administración Tributaria
(SUNAT) in Peru was to systematically root out corruption in tax administration by implementing
radical personnel reform (World Bank, 2001). The Peruvian case showed good progress towards
ensuring personnel autonomy by radically increasing salaries (Taliercio, 2004). Motivations for
corruption coming from lack of competitive compensation were tackled by a strong increase in
salaries of approximately 20 times. As a result the average salary of employees in the newly
created SARA in 1991 rose from US$ 50 per month to US$ 1000 per month (Haltiwanger &
Singh, 1999). In addition to increasing salaries to match employees’ experience, additional
constitutional provisions were also introduced to ensure that salaries at SUNAT would continue to
be competitive with the private sector. There is no denying that SUNAT’s massive rises in
salaries would not have come true without personnel autonomy, and this increase in salaries
helped SUNAT to curb motivations for corruption for its employees. This supposition was also
supported by public’s responses to perception survey in Peru carried out by Taliercio (2004). The
131
survey findings highlighted that public perception were extremely strong that SUNAT had
contributed towards controlling corruption in tax administration in Peru. About 85% of the
respondents agreed that corruption was substantially less than prior to SARA reform.
Ghana also represents one of the earlier SARA cases which utilized personnel autonomy to curb
motivations for corruption by increasing wages and giving bonuses if agreed revenue targets are
achieved. This was partly achieved by not re-hiring corrupt staff for the newly created National
Revenue Secretariat (NRS) in 1985 (de Merode & Thomas, 1994; Chand & Moene, 1997, 1999;
Terkper, 1994). The SARA in Ghana is a representative case of utilization of compensation-
based anti-corruption reforms. Performance linked bonuses were implemented within a year of its
inception to revert the downward trend in Tax-to-GDP ratio which declined to 4.8% in 1984 from
16% in 1976. Tax officials were rewarded with a bonus of 15% of annual salary if the annual
revenue target was achieved. These bonuses were paid year end to all those employees who
were rated positively by the Performance Review Report, and bonuses were financed through the
excess revenues collected over the agreed annual revenue target. These bonuses seem to work
as they helped to raise the tax-to-GDP ratio to a high 23.6% in 1993. The comparative advantage
of NRS employees to the rest of the civil service has eroded over the years due to across-the-
board raises in compensation for the whole civil service in Ghana. Notwithstanding erosion in
salaries over the years, Ghana showed considerable progress towards utilizing personnel
autonomy to curb motivations for corruption via increases in wages and rewards.
Kenya also showed good commitment towards personnel reform by establishing Kenyan
Revenue Authority (KRA) in 1996. KRA utilized personnel autonomy to introduce selective and
incremental increases in salaries. The management of the KRA introduced a new salary scale,
which was developed by adopting the salary scales of Kenya’s central and commercial banks as
132
benchmarks. The real term increase in salaries, based on the new scale, was applied to the top
management only. Across-the-board implementation of the new scale was prevented due to
funding restrictions. In lieu, the rest 98% of KRA employees were compensated with annual
raises over many years. Despite these incremental raises, there is continuing disparity between
improvement of wages for top management of KRA and rest of the employees. There is
widespread perception that increases in wages in KRA were not possible without personnel
autonomy and resulted in reduction, though not elimination, in corruption. Taliercio (2004)
showed a strong positive correlation between KRA’s overall performance (including control of
corruption) and its degree of autonomy (including personnel autonomy).
After its inception in 1997, the South African Revenue Service (SARS) utilized its restricted
personnel autonomy innovatively to increase wages and rewards. Market competitive salaries
were introduced in 1999, although the difference with the rest of the civil service was not huge.
SARS also utilized the tool of bonuses and rewards creatively to counteract the problem of
‘auditor poaching’ by the private sector, having experienced nearly 100% turnover of audit staff in
approximately three years. Personnel autonomy was utilized to introduce supplementary
premiums for employees with scarce skills at 10% of salary and performance contracts at further
20% of salary. This enabled SARS to retain its experienced auditors by offering a premium of
around 30% of salary. Immediately after its inception SARS also used a slightly controversial
performance reward system for three years. Staff was paid performance bonuses of around 46%
of annual salaries and functioned as a substitute of salary reform. These bonuses were based on
excess revenues collected by SARS over revenue target. Although these performance bonuses
were praised by SARS management for enabling improvement in revenue collection, they were
also criticized on account of compensating idlers as well as better performers and creating a
“prosecutorial mentality” of tax officials against taxpayers (Taliercio, 2004).

133
After its inception in 1992, the Uganda Revenue Authority (URA) increased its wages
dramatically to attract talent from the private sector and motivate existing employees to improve
performance and avoid corrupt behaviour (Clark & Wood, 2001; Fjeldstad et al., 2003). But in
case of Uganda, increased wages raised the opportunities for corruption in the appointment of
staff in URA. Uganda showed less progress in utilizing personnel autonomy to increase wages
and rewards with major harm done by interference from the MoF. The MoF was able to interfere
fully with the hiring and firing mechanisms in the newly established URA in the wake of increased
wages, although theoretically URA was supposed to be independent from executive in
recruitment choices. Therkildsen (2004) highlighted how several high level tax officials belonging
to the revenue department were simply ‘transferred’ to the URA and not ‘recruited’ by following
the laid down procedures. Uganda represents a case of SARAs where any increase in wages
and rewards, in the face of limited personnel autonomy, due to continuing political interference
from MoFs, simply results in increase in ills of patronage. Taliercio (2004) highlighted the
patronage problem in Uganda where tax officials continued reporting to work despite dismissal by
URA management. Also the increased wages did not keep up with inflation as they were
increased only once in a decade of 1991-2001.The real increase in wages for high level officials
in URA has declined over the years as MoF, though permitted, but refused to offer any further
funding for improvement in salary scales. Although the URA has shown some progress towards
ensuring personnel autonomy via increases in wages and rewards but these reforms are
undermined by interference from MoF. As a result although taxpayers perceived some progress
against corruption due to increases in wages, but these reforms have not dissuaded taxpayers
from continuing to perceive high level of corruption in URA (Taliercio, 2004).
Tanzania, which established Tanzania Revenue Authority (TRA) in 1996, represents a SARA
case which tried to control motivations for corruption by utilizing personnel autonomy to increase
134
wages (Clark & Wood, 2001). As a result there was drastic increase in wages, in some cases as
high as ten times of an average salary, to attract and keep skilled and honest staff. But Fjeldstad
(2003) showed how improved levels of remuneration co-existed with simultaneous high levels of
corruption after SARA reform. The reason of failure of increased wages to deter tax officials from
indulging in corruption was attributed to availability of opportunities for corruption coupled with
low portability of being caught. In other words, increased salaries were aimed at curbing
motivations, but not opportunities for corruption. The high probability of getting caught reduced in
TRA as corruption was found in its internal monitoring unit, which itself was designated with the
task of unearthing corruption. The internal investigation and monitoring unit (IIMU) of the TRA,
aimed at strengthening internal oversight mechanisms, by increasing the probability of corruption
detection, was accused of corruption itself. This resulted in dismissal of 24 employees (including
the head) from IIMU on account of corruption charges in 2000 (Fjeldstad, 2002). As a result
increased wages, in the absence of sufficient monitoring mechanisms, served as a bonus for tax
officials rather than a substitute, on top of bribes received through corruption. The experience of
TRA showed that increases in wages represent only one of the numerous factors affecting tax
official behavior towards corruption. Fjeldstad et al., (2003, p.69) argued that “since corruption to
some extent is more a question of lack of social stigma than low wages, high wages will not in
themselves keep people away from corrupt practices”. Tanzania represents a SARA case which
highlights that increase in wages will act as an add-on to bribes for tax officials in the face of
ineffective anti-corruption internal monitoring mechanisms. Hence, progress towards controlling
motivations for corruption in TRA through increases in wages was damaged by the weaknesses
in controlling opportunities for corruption due to ineffective internal monitoring mechanisms.
Venezuela, which introduced National Integrated Tax Administration Service (SENIAT) in 1993,
has not made much progress with personnel reform after its inception, and has not made any
135
considerable advances regarding utilizing increases in wages and rewards towards ensuring
personnel autonomy. Rather it focused more towards recruitment and retrenchment mechanisms
of personnel reform, which enabled SENIAT to reduce staff number and hiring experts from
private sector. Similarly to Venezuela, the progress of Mexico’s Mexican Tax Administration
Service (SAT), which was set up in 1997, towards ensuring personnel autonomy has been
appraised less than average (Taliercio, 2004). Like Venezuela, Mexico has not made any
considerable efforts towards utilizing increases in wages and rewards as a tool for ensuring
personnel autonomy. As a result taxpayers continued to perceive high levels of corruption in
Venezuela and Mexico after SARA reforms due to reversion of SARA to old ways (Venezuela)
and minor impact of reform on SARA (Mexico) (Taliercio, 2004).
Table 5.3 presents overview of macro level of SARA analysis with respect to the link between
personnel autonomy resulting in increases in wages and rewards for SARAs analysed above.
Table 5.3: Overview of Selected SARA Countries concerning Personnel Autonomy

resulting in Increases in Wages and Rewards
Personnel Autonomy and Increases in

SARA Country
Wages and Rewards*
Ghana +
South Africa +
Peru +
Tanzania +
Mexico -
Uganda +
Kenya +
Venezuela -
* + means personnel autonomy resulted in increases in wages and rewards in the SARA country.
- means personnel autonomy did not resulted in increases in wages and rewards in the SARA country.
This table illustrates that all analysed SARA cases, except Venezuela and Mexico, utilized
personnel autonomy to increase wages and rewards for tax officials. As per discussions in the
macro level of SARA analysis, the SARA cases showing good progress towards controlling
136
motivations for corruption by ensuring personnel autonomy in terms of increases in wages and
rewards include Peru, Kenya, Ghana, and South Africa, while SARA cases including Uganda,
Tanzania, Venezuela, and Mexico showed less progress towards controlling motivations for
corruption by inadequately utilizing personnel autonomy to increase wages and rewards for
SARAs. These SARA countries can be loosely categorized into three groups. All three of these
groups lend support to the basic argument behind sub-hypothesis 1(a) as per macro level of
SARA analysis. First, the progress against controlling motivations for corruption improved most
when personnel autonomy in terms of increases in wages and rewards was relatively high in
comparative terms (Peru, Kenya, Ghana and South Africa). These SARA cases made better use
of personnel autonomy to increase wages and rewards and were better able to control
motivations for corruption. Second, the progress against controlling motivations for corruption
improved least in cases where personnel autonomy in terms of increases in wages and rewards
was low (Venezuela, Mexico). These SARA cases made less than optimal use of personnel
autonomy to increase wages and rewards and were less effective in controlling motivations for
corruption. Third, the progress against controlling motivations for corruption varied, improving at
first and then declining over time, in cases where personnel autonomy in terms of increases in
wages and rewards decreased over time (Uganda, Tanzania). These SARA cases could not
sustainably increase wages and rewards and experienced decline in controlling motivations for
corruption over time. So for sub-hypothesis 1(a), the macro level of SARA analysis only provided
limited support for four out of eight SARAs analysed for increases in wages and rewards. Hence
sub-hypothesis 1(a) was only partially supported for four out of total eight SARA cases.
5.1.2 Personnel Autonomy Controlling Motivations for Corruption by Increases in the
Probability of Detection
137
Research hypothesis 1(b) proposed that by increases in the probability of detection, SARAs have
been effective against motivations but not opportunities for corruption.This section aims to
identify and analyze secondary literature on SARAs with particular focus on if and how personnel
autonomy is being utilized by SARAs to control motivations for corruption by increases in the
probability of detection.The development of research hypothesis 1(b) in the analytical framework
enabled to propose that if SARAs have made progress in terms of personnel autonomy by
resulting in increases in the probability of detection, then it can also be taken to propose that
motivations for corruption have been controlled as an intended but unrecognized consequence.
Table 5.4 illustrates an overview of secondary literature analysed to examine research
hypothesis 1(b).

Increases in the Probability of Detection for SARAs
Selected Literature concerning Personnel Autonomy and Increases in the Probability of

Detection for SARAs
Mann, 2004 Clark & Wood, 2001
Fjeldstad, 2002; 2003 Terkper, 1994
Martinez-Vazquez et al., 2004 World Bank, 2001
Ofosu-Amaah et al., 1999 Devas et al., 2001
Taliercio, 2004 Hlophe & Friedman, 2002
Fjeldstad et al., 2003 Durand et al., 1998
A powerful deterrent for tax officials to engage in corruption is the penalty received in case they
are caught. For this, the probability of detection represents an important factor towards motivating
tax officials not to engage in corruption. Increasing the probability of detecting corruption within
SARAs requires the introduction and strengthening of monitoring mechanisms to monitor the tax
officials. In Tanzania, the TRA took concrete steps towards controlling motivations for corruption
by increasing the probability of detection, but did not progressed well. Improvements in the
Prevention of Corruption Act in 1993 were largely aimed towards decreasing the motivations for
corruption by increasing the probabilities of detection via reinforcing the capability of TRA to
138
monitor staff. A new internal investigation and monitoring unit (IIMU) of the TRA, aimed at
strengthening internal oversight mechanisms, by increasing the probability of corruption
detection, was established. Despite this, the probability of getting caught reduced in TRA as
corruption was found in IIMU itself, although it was designated with the task of unearthing
corruption in the rest of the TRA. Right after its inception in 1996, many staff of the IIMU were
already taking bribes for not detecting corruption by 1997. This resulted in dismissal of 24
employees (including the head) from IIMU on account of corruption charges in 2000 (Fjeldstad,
2002; 2003; Mann, 2004). The TRA reforms were rated as successful in its first year of operation
with significant decline in corruption. This was partly owing to the reason that employees were
unable to gauge the probability of detection by the newly created IIMU due to unawareness of the
effectiveness of internal auditing mechanisms. After first year, though, progress against
corruption significantly declined (Osoro et al., 1999) as employees quickly learned that they could
collude with the members of the IIMU for corruption.
Legislation relating to possessing inexplicable wealth or property, in the form of mandatory
declaration of assets, can serve as an effective indirect indicator to increase the probability of
detection. SARA countries which have made progress by utilizing the force of law to enforce
declaration of assets by employees include Tanzania, Ghana, Uganda, Mozambique, and Malawi
(Martinez-Vazquez et al., 2004). Introduction of this legislation necessitates the investigation of
suspicious cases. But the implementation of any effective investigation can be fraught with
difficulty as it is commonplace for corrupt employees to transfer illegitimate assets to family
members and friends. To overcome this hurdle, Tanzania took an extra step of supporting the
investigation of assets of family members of employees under suspicion in the legislation (Ofosu-
Amaah et al., 1999). Tanzania represents a SARA case which tried to effectively increase the
probability of detection of corruption by establishing an internal monitoring and investigation unit

139
as well as legislating for monitoring dubious assets of employees, but this two-pronged strategy
crumbled to the ground in the face of corruption found in the monitoring unit itself.
In addition to Tanzania, other SARA cases trying to increase the probability of corruption
detection include Peru, Kenya, South Africa, and Guatemala. Like Tanzania, the progress of
Guatemala’s Superintendencia de Administración Tributaria (SAT) towards increasing the
probability of detection has been rated as less than satisfactory. The ineffectiveness of its internal
audit unit has been criticized to following a policy of “see no evil, hear no evil, speak no evil”
(Mann, 2004). For the period 2002- 2004, many employees of SAT accused of corruption
charges were simply let go by either dismissal from service or transferred to outer offices of SAT,
without prosecution. Peru, on the other hand, relied on legislation to effectively increase the
probability of corruption detection through a presidential executive decree in 1991. This decree
authorized the superintendent of SUNAT to probe the “indicators of immoral conduct” and firing
employees on account of “lack of probity” or “abuse of authority”. In particular, this decree
empowered the superintendent to get rid of employees who displayed discrepancy between
“exterior signs of wealth” and their tax declarations. The effectiveness of this legislation in
increasing the probability of corruption in Peru has been largely termed as effective (Taliercio,
2004).
In Africa, Kenya’s progress towards improving the probability of corruption detection has been
rated as exemplary. KRA incorporated numerous overlapping accountability mechanisms
including establishment of an internal audit division. This division was entrusted with the
responsibility of undertaking internal audits on a quarterly basis and reporting the results to the
Commission General of KRA. In addition, probability of detection also improved on account of
KRA’s code of conduct and establishment of a board disciplinary committee. These accountability
140
mechanisms have been evaluated to worked fairly well over the years and resulting in increase in
the professionalism of staff as well as reduction, though not elimination, of corruption in KRA.
Like Kenya, the SARS in South Africa also made good progress towards increasing the
probability of detection by creating an internal audit division in the year 2000. This division also
reports directly to the commissioner of SARS, in addition to having a direct link with the
chairperson of the advisory board of SARS. This division was charged with the responsibility of
undertaking internal audits on a quarterly basis and reporting the results to the Audit Committee
in SARS (Taliercio, 2004).
personnel autonomy resulting in increases in the probability of detection for SARAs analysed
above.

resulting in Increases in the Probability of Detection
Personnel Autonomy and Increases in the

SARA Country
Probability of Detection*
South Africa +
Peru +
Tanzania +
Guatemala +
Kenya +
* + means personnel autonomy resulted in increases in the probability of detection in the SARA country.
- means personnel autonomy did not resulted in increases in the probability of detection in the SARA country.
This table illustrates that all analysed SARA cases utilized personnel autonomy to increase the
probability of corruption detection. As per discussions in the macro level of SARA analysis, the
SARA cases showing good progress towards controlling motivations for corruption by ensuring
personnel autonomy in terms of increases in the probability of detection include South Africa,
Peru, and Kenya, while SARA cases including Tanzania and Guatemala showed less progress
towards controlling motivations for corruption by inadequately utilizing personnel autonomy to

141
increase the probability of corruption detection in SARAs. Both these groups lend support to the
basic argument behind research hypothesis 1(b) as per macro level of SARA analysis. First, the
progress against controlling motivations for corruption improved most when personnel autonomy
in terms of increases in the probability of detection was relatively high in comparative terms
(South Africa, Peru, and Kenya). These SARA cases made better use of personnel autonomy to
increase the probability of detection and were better able to control motivations for corruption.
Second, the progress against controlling motivations for corruption varied, improving at first and
then declining over time, in cases where personnel autonomy in terms of increases in the
probability of detection decreased over time (Tanzania and Guatemala). These SARA cases
could not sustainably increase the probability of detection and experienced decline in controlling
motivations for corruption over time. So for research hypothesis 1(b), the macro level of SARA
analysis only provided limited support for three out of five SARAs analysed for increases in the
probability of detection. Hence research hypothesis 1(b) was only partially supported for three out
of total five SARA cases.
5.1.3 Personnel Autonomy Controlling Motivations for Corruption Increases in and Stricter
Enforcement of Penalties for Corruption
Research hypothesis 1(c) proposed that by increases in and stricter enforcement of penalties for
corruption, SARAs have been effective against motivations but not opportunities for corruption.
This section aims to identify and analyze secondary literature on SARAs with particular focus on
if and how personnel autonomy is being utilized by SARAs to control motivations for corruption by
increases in and stricter enforcement of penalties for corruption. The development of research
hypothesis 1(c) in the analytical framework enabled to propose that if SARAs have made
progress in terms of personnel autonomy by resulting in increases in and stricter enforcement of
penalties for corruption, then it can also be taken to propose that motivations for corruption have
142
been controlled as an intended but unrecognized consequence. Table 5.6 illustrates an overview
of secondary literature analysed to examine research hypothesis 1(c).

Increases in and Stricter Enforcement of Penalties for Corruption in SARAs
Selected Literature concerning Personnel Autonomy and Increases in and Stricter

Enforcement of Penalties for Corruption in SARAs
Mann, 2004 Fjeldstad, 2005a
Kidd & Crandall, 2006 Fjeldstad & Moore, 2009
Martinez-Vazquez et al., 2004 Fjeldstad et al., 2003
World Bank, 1999; 2001 Devas et al., 2001
Taliercio, 2004 Delay et al., 1998
The imposition of high levels of penalties institutes an efficient deterrence to corruption. The logic
behind imposition of high levels of penalties for corruption rests on the need to combine positive
incentives with negative incentives. As Di Tella and Schargrodsky (2003, p. 3) posits “carrots and
sticks should be viewed as complementary tools in fighting corruption”. World Bank (1999) also
recommends adoption of this dual strategy in the anticorruption strategy for tax administrations.
Martinez-Vazquez et al., (2004) recommends imposition of strict penalties once corrupt activities
have been identified including monetary sanctions and/or job dismissals with a prospect of prison
sentences in order to reduce motivations for corruption in tax administrations.
In case of Uganda, the progress of URA towards controlling motivations for corruption via
increases in and stricter enforcement of penalties for corruption can be characterized by a case
of one step forward and two steps back. Adoption of SARA reform in Uganda failed to dissuade
taxpayers from continuing to perceive high levels of corruption, but some progress against
corruption was noticed on account of strict punishments for tax officials. Over a period of eight
years from 1993-2001, URA fired 257 employees on corruption charges (Taliercio, 2004). The
reform of revenue authority in Uganda resulted in dismissal of 14% of its staff on misconduct
143
charges (Fjeldstad et al., 2003). Rather than exemplifying the strengthening of the sanctioning
mechanisms, dismissal of such a high number of staff on corruption charges demonstrates the
ineffectiveness of the monitoring and sanctioning mechanisms in place in URA, enabling such a
high number of corrupt staff to continue working in the URA, until finally caught. Also the
effectiveness of dismissal from service as a penalty has been put to question in case of Uganda,
even though the obvious loss of wages and employment. Fjeldstad et al., (2003) highlighted a
troubling perception by several employees in the revenue authority that the span of their
employment at the URA consists of a very few number of years to make as much money as
possible, after which the inconvenience of dismissal is perceived to be quite limited once finally
caught. URA case highlights that despite a seemingly high number of dismissals on account of
corruption, enforcement of penalties has generally failed to deter tax officials from indulging in
corruption. On top of it, this dimension of personnel autonomy of URA, though generally proving
ineffective, has been disputed in court of law by criticizing personnel autonomy of URA which
enables it not to give a reason for dismissal of employees. The URA defends this policy stance
owing to complications of proving alleged corruption. In general, URA has made efforts to
increase the number and severity of penalties for corruption, which has resulted in some
reduction in corruption, with taxpayers continuing to perceive still high levels of corruption at the
URA.
In Tanzania, the objectives behind creation of TRA included autonomy for hiring and firing
mechanisms and increasing the probability of dismissal when caught for corruption, thus
increasing the opportunity cost of committing corruption (Martinez-Vazquez et al., 2004). Right
after its inception, in order to undertake a cleaning exercise, the TRA dismissed all of its previous
employees and let me re-apply for employment in the revenue authority by undergoing a rigorous
recruitment exercise. As a result more than one third of the employees who re-applied were
144
rejected on proof or suspicion of wrongdoing. Although, the reform of revenue authority in
Tanzania resulted in dismissal of a very high 35% of staff on account of misconduct charges
(Fjeldstad et al., 2003). And dismissal of such a high number of staff on corruption charges
demonstrated the ineffectiveness of the monitoring and sanctioning mechanisms in place before
the creation of TRA. But this cleaning exercise by TRA backfired in the wake of development of
corruption networks outside TRA. Martinez-Vazquez et al., (2004) suggested that the high
turnover in TRA may have led to formation of corruption networks external to TRA. The former
employees of the TRA, well versed with the knowledge as well as loopholes of the internal
organizational workings, were recruited in the private sectors as tax accountants and experts.
This increased the chances of development of a bigger corruption network spanning both within
and outside TRA, with chances of collusion between TRA staff and their former coworkers now in
the private sector. TRA’s efforts to increasing and strictly enforcing penalties for corruption
through dismissal of a large number of staff were neutralized by its inability to bring convictions
against the perpetrators of corruption. Hence it failed to bar them from gaining employment in the
private sector, greatly reducing their opportunity cost of dismissal from the TRA.
The case of Guatemala’s SAT is similar to Uganda in a sense that on paper it seems that over
the years it has moved quite forcefully against several high profile cases of corruption, and
seemingly tried to increasingly and strictly enforcing penalties for corruption. But further analysis
points towards the ineffectiveness of the monitoring and sanctioning mechanisms in place in
SAT, enabling such a high level of corruption being committed, until finally detected. Over the
years the former Vice President and Minister of Finance, who also chaired the board of directors
of SAT, have been imprisoned on corruption charges. SAT also moved forcefully against its
former superintendent. One of its former superintendents for period 2002-2004, who absconded
for five months, was finally caught and sentenced in 2004. He was accused of embezzling more
145
than 5 million US$. Several other co-accusers, including his family and SAT personnel were not
caught (Mann, 2004). The case of Guatemala highlights that despite efforts to increasingly detect
and penalize corruption in SAT, more effort is needed to try to prevent incidence of corruption of
such a high magnitude in the first place.
In Peru, the effectiveness of SUNAT’s personnel autonomy towards increasing and strictly
enforcing penalties for corruption has been rated to function well. This in part rested on the
introduction of a presidential executive decree in 1991. This decree authorized the
superintendent of SUNAT to dismiss employees on account of “lack of probity” or “abuse of
authority”. Although president Fujimori was largely praised for providing immense support for
SARA reform from the highest political level (World Bank, 2001), but towards the end of Fujimori
government agents from the National Intelligence Service (SIN) intruded into SUNAT (Taliercio,
2004). Starting in 2000, a former Deputy Superintendent of SUNAT, perceived to be very honest,
was given a second term at SUNAT as Superintendent, mainly to undertake the ‘cleaning’
exercise and getting rid of infiltrators coming from SIN. Although his term at SUNAT spanned for
only eight months and was eventually replaced with the arrival of a new government in 2001,
nonetheless he was able to get rid of almost 40 agents of the SIN who were working as SUNAT
employees. Also, the former head of SIN was also convicted and imprisoned for 15 years on
account of corruption charges, in addition to facing 47 additional charges. Over the next four
years till 2004, Peru was able to get convictions for more than 100 persons on account of
corruption charges in the Fujimori government, in addition to recovering more than 150 million
US$ from foreign banks (Mann, 2004). In view of these sanctions, Peru made headway towards
increasing and strictly enforcing penalties for corruption in Peru’s SARA to effectively curb

146
In South Africa, SARS has also tried to make significant progress towards increasingly and
strictly enforcing penalties for corruption. From 1998, SARS got rid of 173 personnel on account
of misconduct and corruption charges, which represented almost 1.5% of the total personnel. In
addition, SARS openly publicizes the total prison sentences attained against corruption as
deterrence for the rest of its personnel. It announced the figures of total 99 years of prison
sentences attained in 2000 and 121 years of prison sentences attained in 2001on account of tax
fraud (Taliercio, 2004). This resulted in significant reduction of tax officials as well taxpayers’
perception of prevalent corruption in SARS, mainly due to perceived improvement in quality and
integrity of the personnel that remain in SARS after reform. In case of Kenya as well, significant
progress has been noted in KRA towards increasingly and strictly enforcing penalties for
corruption, mainly on account of personnel autonomy resulting in independence of its
management to easily get rid of employees suspected of corruption, in addition to KRA’s code of
conduct and board disciplinary committee playing their part in increasing the integrity of staff.
personnel autonomy resulting in increases in and stricter enforcement of penalties for corruption
for SARAs analysed above.

resulting in Increases in and Stricter Enforcement of Penalties for Corruption
Personnel Autonomy and Increases in and

SARA Country
Stricter Enforcement of Penalties*
South Africa +
Peru +
Tanzania +
Guatemala +
Uganda +
Kenya +
* + means personnel autonomy resulted in increases in and stricter enforcement of penalties in the SARA country.
- means personnel autonomy did not resulted in increases in and stricter enforcement of penalties in the SARA
country.
147
This table illustrates that all analysed SARA cases utilized personnel autonomy to result in
increases in and stricter enforcement of penalties for corruption. As per discussions in the macro
level of SARA analysis, the SARA cases showing good progress towards controlling motivations
for corruption by ensuring personnel autonomy in terms of increases in and stricter enforcement
of penalties for corruption include South Africa, Peru, and Kenya, while SARA cases including
Tanzania, Uganda, and Guatemala showed less progress towards controlling motivations for
corruption by inadequately utilizing personnel autonomy to result in increases in and stricter
enforcement of penalties for corruption in SARAs. Both groups lend support to the basic
argument behind research hypothesis 1(c) as per macro level of SARA analysis. First, the
progress against controlling motivations for corruption improved most when personnel autonomy
in terms of increases in and stricter enforcement of penalties for corruption was relatively high in
comparative terms (South Africa, Peru, and Kenya). These SARA cases made better use of
personnel autonomy to result in increases in and stricter enforcement of penalties for corruption
and were better able to control motivations for corruption. Second, the progress against
controlling motivations for corruption varied, improving at first and then declining over time, in
cases where personnel autonomy in terms of increases in and stricter enforcement of penalties
decreased over time (Tanzania, Uganda, and Guatemala). These SARA cases could not
sustainably result in increases in and stricter enforcement of penalties for corruption and
experienced decline in controlling motivations for corruption over time. So for research
hypothesis 1(c), the macro level of SARA analysis only provided limited support for three out of
six SARAs analysed for increases in and stricter enforcement of penalties. Hence research
hypothesis 1(c) was only partially supported for three out of total six SARA cases.
148
5.1.4 Personnel Autonomy Controlling Motivations for Corruption by Instilling Ethics in
Tax Officials
Research hypothesis 1(d) proposes that by instilling ethics in tax officials, SARAs have been
effective against motivations but not opportunities for corruption. This section aims to identify and
analyze secondary literature on SARAs with particular focus on if and how personnel autonomy is
being utilized by SARAs to control motivations for corruption by instilling ethics in tax officials.
The development of research hypothesis 1(d) in the analytical framework enabled to propose that
if SARAs have made progress in terms of personnel autonomy by resulting in instilling ethics in
tax officials, then it can also be taken to propose that motivations for corruption have been
controlled as an intended but unrecognized consequence. Table 5.8 illustrates an overview of
secondary literature analysed to examine research hypothesis 1(d).

Instilling Ethics in Tax Officials in SARAs
Selected Literature concerning Personnel Autonomy and Instilling Ethics in Tax Officials in
SARAs
Shah, 2007 Fjeldstad et al., 2003
World Bank, 2001 Martinez-Vazquez et al., 2004
Silvani & Baer, 1997 Durand et al., 1998
Mayville, 2005 Devas et al., 2001
Taliercio, 2004 Clark & Wood, 2001
The importance of ethical behaviour and conduct by tax officials towards sustaining anti-
corruption SARA reforms cannot be overstated. One of the ways of dealing with corruption in
SARAs is by taking steps to strengthen the ethical and moral behavior of tax officials. One way of
doing it is by increasingly conducting ethics training aimed at sensitizing tax officials to different
forms of corruption, and how to avoid motivations for corruption. Second, adoption of a Code of
Conduct which elucidates expectations for ethical behavior by tax officials can be utilized as an
effective preventive tool for building ethics and checking corruption. The potential capability of
code of conduct towards checking corruption is one reason behind its adoption by many SARA
149
countries in 1990s. Their effectiveness against corruption depends upon a multitude of factors
including effective publicity by the government, exposure of tax officials to its contents, and
enforcement mechanisms in place for checking deviations from ethical standards. Adoption of
both ethics training and a code of conduct for tax officials have been considered essential
ingredients of an anti-corruption strategy for tax administrations (Martinez-Vazquez et al., 2004).
Tanzania took a number of concrete steps towards utilizing personnel autonomy to instil ethics in
tax officials. After its inception, the TRA undertook a cleaning exercise by dismissing all of its
previous employees and let them re-apply for employment in the revenue authority by undergoing
a rigorous recruitment exercise. Following a merit-based recruitment, it re-hired only those who
were not suspected to have engaged in corruption. TRA also established a code of conduct for
tax officials in 2000 which was largely based on sanctions and appeals procedures of the general
civil service and labor codes. The code of conduct entitled immediate dismissal of tax officials in
case of conviction by a court for fraud. This code of conduct, which was especially developed for
tax administration, also categorized offences and penalties into minor, serious, and very serious
offences. Likewise, the sequence of penalties for breach of conduct, in increasing order of
severity included written warning, reprimand, severe reprimand, fine, and finally summary
dismissal (Mayville, 2005). The code of conduct also specified the level and seniority of
administrative staff charged with taking disciplinary action against the suspected tax officials,
matched with the severity of offence. A detailed hierarchical grievance procedure was also in
place in case any tax official wished to report corruption by fellow employees. In 2003, TRA also
formulated a five member ethics committee which was chaired by the Deputy Commission
General and reported to the Commission General on a monthly basis. This committee was made
up of members based on the grade and seniority of the tax officials against whom penal action
was taking place (Fjeldstad et al., 2003). TRA’s efforts to increase ethics by clearly defining
150
corruption offenses and penalties via Prevention of Corruption Act (1993) and code of conduct
(2000) have been nullified to some extent due to corruption in its judicial system, resulting in
failure to ensure convictions for TRA (Martinez-Vazquez et al., 2004). Despite this, Tanzania
showed considerable efforts towards instilling ethics in tax officials.
Peru has also taken considerable steps toward instilling ethics in tax officials. Right after its
inception in 1991, SUNAT also undertook a purging exercise whereby all of its previous
employees were offered a choice of either resign or re-apply for employment in the revenue
authority by undergoing a rigorous merit-based recruitment exercise. Following a three-phased
recruitment exam, which included an assessment of moral judgment of tax officials, it re-hired
less than one third of the successful applicants (World Bank, 2001). In terms of ethics training,
the SUNAT stipulated a new condition of sixteen-hour course in ethics for all of its technical
personnel and auditors, as a way of controlling motivations for corruption (Silvani & Baer, 1997).
As for code of conduct, the SUNAT’s ethics code promulgated in 1996 described the
organizational arrangements of an ethics committee for dealing with corruption. The ethics
committee, which was an important part of SUNAT, comprised of the Commissioner of Revenue
(chair), the head of Tax Administration, the Director General of Administration, the Director of
Internal Audit, the head of Human Resources, the head of Legal Affairs, and the Director of
Organization. The Committee recommended approaches to deal with tax officials alleged for
illegal actions. In order to encourage ethical behavior, SUNAT also nominated and recognized
tax officials annually for performing their jobs with devotion and commitment to service, including
having supported, encouraged and demonstrated ethical behaviour as per norms of SUNAT
(Mayville, 2005). By taking up these measures, Peru made headway towards instilling ethics in
tax officials in SUNAT to effectively curb motivations for corruption.
Other SARA countries taking steps to improve the ethical behaviour of tax officials include Kenya
and Mexico. Mexico’s code of conduct, which was promulgated in 2000, has been used in
151
evaluating tax official’s performance and for determining promotions. Its code of ethics for the
fiscal career drew on the constitution of Mexico, the general civil service code, and tax and
customs legislation. In case of Mexico, although it had developed a code of conduct, but could
not manage to fully implement it. The code of conduct remained a sort of moral guide without
much effect on the behavior of tax officials. In Kenya, the KRA was recognized to have improved
the professionalism of tax officials and reducing the corruption in the revenue authority. KRA’s
code of conduct and board disciplinary committee were recognized to contribute towards
increasing the integrity of tax officials (Taliercio, 2004).
personnel autonomy resulting in instilling ethics in tax officials for SARAs analysed above.

resulting in Instilling Ethics in Tax Officials
Personnel Autonomy and Instilling Ethics

SARA Country
in Tax Officials*
Peru +
Tanzania +
Mexico +
Kenya +
* + means personnel autonomy resulted in instilling ethics in tax officials in the SARA country.
- means personnel autonomy did not resulted in instilling ethics in tax officials in the SARA country.
This table illustrates that all analysed SARA cases utilized personnel autonomy to instil ethics in
tax officials. As per discussions in the macro level of SARA analysis, all analyzed SARA cases
including Peru, Tanzania, and Kenya, except Mexico, showed good progress towards controlling
motivations for corruption by utilizing personnel autonomy in terms of instilling ethics in tax
officials. Mexico showed less progress towards controlling motivations for corruption by
inadequately utilizing personnel autonomy to instill ethics in tax officials in SARAs. Both groups
lend support to the basic argument behind research hypothesis 1(d) as per macro level of SARA
152
analysis. First, the progress against controlling motivations for corruption improved most when
personnel autonomy in terms of instilling ethics in tax officials was relatively high in comparative
terms (Peru, Tanzania, and Kenya). These SARA cases made better use of personnel autonomy
to instill ethics in tax officials and were better able to control motivations for corruption. Second,
the progress against controlling motivations for corruption varied, improving at first and then
declining over time, in case where personnel autonomy in terms of instilling ethics in tax officials
decreased over time (Mexico). This SARA case could not sustainably instill ethics in tax officials
and experienced decline in controlling motivations for corruption over time. In case of Tanzania,
the macro level of SARA analysis for research hypothesis 1(a), 1(b) and 1(c) in sections 5.1.1,
5.1.2 and 5.1.3 highlighted that it made less than optimal use of personnel autonomy to control
motivations for corruption by unsustainably increasing wages and rewards for tax officials,
unsustainably increasing the probability of detection and ineffectiveness in increasingly and
strictly enforcing penalties for corruption, but showed better progress in case of instilling ethics in
tax officials. So, Tanzania has shown some notable progress towards utilizing personnel
autonomy to instill ethics in tax officials, but not the other three preventive strategies to control
motivations for corruption. So for research hypothesis 1(d), the macro level of SARA analysis
only provided partial support for three out of four SARAs analysed for instilling ethics in tax
officials. Hence research hypothesis 1(d) was only partially supported for three out of total four
SARA cases.
5.1.5 Synthesis of Analysis for Research Hypothesis 1
As highlighted at the start of this section, the macro level of SARA analysis was aimed to analyse
research hypotheses 1 and 2. Research hypothesis 1 proposed that by preferring personnel
autonomy over effective accountability, SARAs have been effective against motivations but not
opportunities for corruption. Research hypothesis 1 comprised of four further sub-hypotheses

153
(1a, 1b, 1c, and 1d), which have already been examined in sections above. As it was proposed at
the start of section 5.1, analysis of these four sub-hypotheses will be undertaken with an aim to
test research hypotheses 1, and any answers or findings as per analysis of these sub-
hypotheses will mean to respond to research hypothesis 1. Consequently this section aims to
synthesize all four sub-analyses discussed in sections above for the cumulative analysis of
research hypothesis 1.
Table 5.10 illustrates the accumulation of all four sub-analyses discussed in sections 5.1.1, 5.1.2,
5.1.3 and 5.1.4 of the macro level of SARA analysis with respect to the link between personnel
autonomy and all four preventive strategies to control motivations for corruption.
Table 5.10: Overview of Selected SARA Countries concerning Personnel Autonomy and all
four preventive strategies to control Motivations for Corruption
Research Hypothesis 1
Personnel Autonomy
Increases in Increases in
SARA Country Increases in Instilling
the and Stricter
Wages and Ethics in Tax
Probability of Enforcement
Rewards Officials
Detection of Penalties
Ghana + n.d n.d n.d
South Africa + + + n.d
Peru + + + +
Tanzania + + + +
Mexico - n.d n.d +
Guatemala n.d + + n.d
Uganda + n.d + n.d
Kenya + + + +
Venezuela - n.d n.d n.d
Note: n.d means no data available.
+ means personnel autonomy resulted in increase in the specific preventive strategy to control motivations for
corruption, as per macro level of SARA analysis.
- means personnel autonomy did not resulted in increase in the specific preventive strategy to control motivations for
corruption, as per macro level of SARA analysis.
154
This table presents a snapshot of selected SARA countries towards utilizing different preventive
strategies to control motivations for corruption. This table also summarizes the findings in terms
of individual sub-hypotheses. As evident from the table, almost one third cells (13/36) display
non-availability of comparative data, which constraints comprehensive analysis for some SARA
countries in terms of all four preventive strategies in support of research hypothesis 1. In terms of
individual sub-hypotheses, research hypothesis 1 was provided limited support by sub-
hypothesis 1(a) {support for 4/8 SARAs}, limited support by sub-hypothesis 1(b) {support for 3/5
SARAs}, limited support by sub-hypothesis 1(c) {support for 3/6 SARAs}, and also limited support
by sub-hypothesis 1(d) {support for 3/4 SARAs}. Hence, it is pertinent to conclude that as per
individual sub-hypotheses, research hypothesis 1 was only provided limited support. In sum, this
section was aimed at analyzing the overall proposition contending that those SARAs making
most progress towards ensuring personnel autonomy should also be the ones making greatest
progress towards controlling motivations for corruption. This supposition was only provided
limited support above.
5.2 EFFECTIVE ACCOUNTABILITY CONTROLLING OPPORTUNITIES FOR CORRUPTION IN
SARAs
In this study, SARA design component of effective accountability is conceptualized to control
opportunities for corruption by 1) introduction of oversight mechanisms; 2) de-politicization of tax
officials; and 3) reduction of discretionary powers of tax officials. Research hypothesis 2
proposed that SARAs have been ineffective in controlling opportunities for corruption due to
ineffective accountability by MoF. The macro level of SARA analysis in this section is aimed to
analyse research hypothesis 2. It comprises of three sub-hypotheses (2a, 2b, and 2c), with each
one referring to one specific preventive strategy to control opportunities for corruption. First all
155
three sub-hypotheses will be analyzed and then these analyses will be integrated for the
cumulative analysis of research hypothesis 2 in sections below.
The next sections examine how SARAs used effective accountability to control opportunities for
corruption. The overall proposition is that those SARAs which had ineffective accountability due
to interference form MoFs would also be ineffective in controlling opportunities for corruption.
5.2.1 Effective Accountability Controlling Opportunities for Corruption by Introducing
Effective Oversight Mechanisms
Research hypothesis 2(a) proposed that SARAs have been ineffective in controlling opportunities
for corruption due to ineffective accountability by MoF by not introducing effective oversight
mechanisms. This section aims to identify and analyze secondary literature on SARAs with
particular focus on if and how effective accountability is being utilized by SARAs to control
opportunities for corruption by introduction of oversight mechanisms. The development of
research hypothesis 2(a) in the analytical framework enabled to propose that if SARAs have
made progress in terms of ensuring effective accountability by introducing effective oversight
mechanisms, then it can also be taken to propose that opportunities for corruption have been
secondary literature analysed to examine research hypothesis 2(a).
Table 5.11: Overview of Secondary Literature concerning Effective Accountability and

Introduction of Oversight Mechanisms for SARAs
Selected Literature concerning Effective Accountability and Introduction of Oversight

Mechanisms for SARAs
Clark & Wood, 2001 Hall & Jenkins, 1995
Delay et al., 1998 Kidd & Crandall, 2006
Fjeldstad, 2002, 2003, 2005a, 2005b Martinez-Vazquez et al., 2004
Fjeldstad & Rakner, 2003 Taliercio, 2004, 2004a
Haldenwang, 2010 Terkper, 1999
156
Matinez-Vazquez et al., (2004) highlighted that SARA reform can strengthen the internal
monitoring mechanisms of revenue authorities when properly implemented. Kenya represented
a SARA case where a strong MoF consistently tried to meddle with its accountability relationships
with other oversight bodies, and undermined effective oversight as a consequence. This was
especially the case with its governing board of directors, all of which members were appointed, or
could be dismissed, by the minister of finance. In KRA case, the design of accountability
mechanisms empowered the minister of finance immensely. The board eventually reflected the
favourites of the minister owing to the power to constitute the board to his liking. Since the
minister had complete freedom in shaping board membership, he appeared to use it generously.
This severely impacted the security of tenure of board members, including the chairperson and
Commission General. During the six year period of 1995-2001 four chairmen changed hats at the
KRA with an average tenure of around 1.6 years (Taliercio, 2004). KRA’s four Commission
Generals (mostly bureaucrats), who were also appointed directly by the minister of finance,
experienced short lived average tenure of around 1.8 years during the same period. The board
members of KRA also experienced a less severe form of turnover rate with an average tenure of
2.8 years against a minimum term of three years. There were instances where the board
members have been dismissed all together over differences with the government. This data
indicating a high turnover rate of KRA’s board members highlighted the instability of its
autonomy, and interference and encroachment of MoF in its affairs. It also highlighted how MoF
undermined the effective oversight mechanisms of KRA by undermining its accountability
relationships with other oversight bodies i.e., the revenue board in this case (Hall & Jenkins,
1995; Fjeldstad & Moore, 2009; Terkper, 1999).
Peru represented a SARA case which used superintendents in place of governing board of
directors. Although the superintendent model bestows a huge amount of authority in one person’s
157
hands, in comparison to board of directors, which might undermine effective accountability. But
in case of Peru the most damage to SUNAT’s autonomy and accountability occurred due to
interference from MoF. Although, Peru started out as a SARA case which was initially
characterized by a restrained MoF control and its indirect influence in SUNAT’s operations.
Legislation provided that the superintendent of SUNAT can be selected and dismissed by the
president only, based on the recommendation of minister of finance , who provided this with the
agreement of the council of ministers. In practice, this legislation entailed that when the president
was strong and supportive of SARA reform, the minister of finance could not interfere with the
naming of superintendent. The first two superintendents of SUNAT were appointed by the
president directly without any input from the minister of finance. This resulted in decrease in
corruption as well as improvement in the accountability mechanisms of Peru at that time
(Taliercio, 2004, 2004a). However, when the support of the president diminished over time, due
to political reasons, the authority of minister in the appointment decision increased
proportionately, owing to advising powers of the minister in the appointment procedure of
superintendent by law. This subjected SUNAT to the whim of minister of finance and weakened
its autonomy over time. The minister of finance gained effective control over the selection and
dismissal of superintendent of SUNAT, which has effectively transferred a lot of autonomy from
SUNAT to MoF. In addition to the appointment of superintendent, MoFs is also charged with the
authority of approving the budget of SUNAT, and carry out periodic assessments of SUNAT. This
loss of autonomy of SUNAT ultimately resulted in deterioration in revenue performance for Peru
over time. Peru represented a SARA case which took off well, but experienced decline in its
autonomy over time due interference from MoF (Martinez-Vazquez et al., 2004; Fjeldstad, 2005b;
Haldenwang, 2010; Kidd & Crandall, 2006).

158
Uganda represented a SARA case which was directly under the influence of minister of finance.
The dominance of MoF of URA’s board was characterized by the situation where the minister of
finance selected five out of total seven members of the board, in addition to selecting the
chairperson and Commission General. The URA act authorized the minister of finance to dismiss
any member of the board for reasons including malfeasance, incapacitation, or “for any other
sufficient cause”, thus severely undermining their security of tenure (Taliercio, 2004). In addition,
the minister of finance could direct the board for compliance with his directives pertaining to how
to perform their duties, effectively making the URA board an instrument of MoF. This resulted in
URA boards composed of members having political ties within the ministry, such that these
boards were more political than technical. The Commission Generals of URA, who were
appointed and could be removed directly by the minister of finance, proved to be mere agents of
the minister. As a result the relationship between the Commission Generals and minister proved
to be more active, than the relationship between the commission generals and the board. Such
high dominance of MoF over URA’s board raised questions about its purpose and value-addition
to the accountability mechanisms. In view of these arrangements, the effect of SARA reform on
revenue performance proved to be unstable over time, only rising initially, in view of
improvements in organizational workings, but deteriorating afterwards (Clark & Wood, 2001). The
deterioration in revenue improvement and corruption were blamed to the ineffectiveness of the
accountability mechanisms, due to interference of MoF in URA’s board (Delay et al., 1998;
Fjeldstad, 2005a, 2005b; Fjeldstad & Rakner, 2003; Jenkins & Khadka, 2000; Kidd & Crandall,
2006).
Tanzania represented a SARA case engaged in tug-of-war with MoF over control and autonomy.
SARA reform translated into loss of considerable autonomy and diminished patronage benefits
for its MoF, which in turn became strong resistance against successful implementation of SARA
159
reform, and was eventually able to over-turn the reform process, and win back autonomy and
patronage benefits (Clark & Wood, 2001; Talierco 2004). This practice in TRA was quite similar
to the case of URA discussed above, in terms of institutional framework including the legislation.
The effectiveness of the accountability relationship between the principal (minister) and agents
(board of directors) declined over time, despite efforts by the TRA by making available standard
reports of authority’s performance against agreed revenue collection targets to the MoF.
Fjeldstad (2003) showed how SARA reform, conceptually designed to curb corruption, did not
fared well when empirically dealing with corruption in TRA in Tanzania, and showed that
improved levels of remuneration co-existed with high levels of corruption, in the fact of
ineffectiveness accountability mechanisms. The deterioration in corruption was blamed to the
ineffectiveness of the accountability mechanisms, due to interference of MoF in TRA’s operations
(Fjeldstad, 2002; Kidd & Crandall, 2006).
effective accountability resulting in introduction of oversight mechanisms for SARAs analysed
above.
Table 5.12: Overview of Selected SARA Countries concerning Effective Accountability

resulting in Introduction of Oversight Mechanisms
Effective Accountability and Introduction of

SARA Country
Oversight Mechanisms*
Peru -
Tanzania -
Uganda -
Kenya -
*+ means SARA design component of effective accountability was successful to introduce effective oversight
mechanisms in the SARA country, with no interference from MoF.
- means SARA design component of effective accountability was unsuccessful to introduce effective oversight
mechanisms in the SARA country, due to interference from MoF.
160
This table illustrates that all analysed SARA cases experienced interference from MoFs
undermining introduction of effective oversight mechanisms. As per discussions in the macro
level of SARA analysis, all analyzed cases including Peru, Tanzania, Uganda, and Kenya
showed less than optimal progress towards controlling opportunities for corruption, due to
interference from MoFs, undermining SARA design component of effective accountability to
introduce effective oversight mechanisms for SARAs. All analyzed SARA cases lend support to
the basic argument behind research hypothesis 2(a) as per macro level of SARA analysis. These
SARA cases made less than optimal use of effective accountability to introduce effective
oversight mechanisms and were less effective in controlling opportunities for corruption. For
research hypothesis 2(a), the macro level of SARA analysis provided considerable support for
four out of four SARAs examining how presence of MoFs undermined effective oversight
mechanisms of SARAs. Hence, research hypothesis 2(a) was considerably supported for four out
of total four SARA cases.
5.2.2 Effective Accountability Controlling Opportunities for Corruption by De-politicizing
Tax Officials
Research hypothesis 2(b) proposed that SARAs have been ineffective in controlling opportunities
for corruption due to ineffective accountability by MoF by not de-politicizing tax officials. This
section aims to identify and analyze secondary literature on SARAs with particular focus on if and
how effective accountability is being utilized by SARAs to control opportunities for corruption by
de-politicization of tax officials. The development of research hypothesis 2(b) in the analytical
framework enabled to propose that if SARAs have made progress in terms of ensuring effective
accountability by de-politicization of tax officials, then it can also be taken to propose that
opportunities for corruption have been controlled as an intended but unrecognized consequence.
161
Table 5.13 illustrates an overview of secondary literature analysed to examine research
hypothesis 2(b).
Table 5.13: Overview of Secondary Literature concerning Effective Accountability and De-
politicization of Tax Officials for SARAs
Selected Literature concerning Effective Accountability and De-politicization of Tax

Officials for SARAs
Durand & Thorp, 1998 Martinez-Vazquez et al., 2004
Fjeldstad & Moore, 2009 Osoro et al., 1999
Fjeldstad et al., 2003 Taliercio, 2004
Jenkins & Khadka, 2000 Terkper, 1999
Jenkins, 1994 Therkildsen, 2004
Kidd & Crandall, 2006 World Bank, 2001
Mann, 2004
Political interference by the MoFs has proved to be a big hurdle in the successful implementation
of SARAs. As Talbot (1994) highlighted that managers’ ability to refrain from political interference
in operations has drastically declined owing to old habit of micro-management. Taliercio (2004)
demonstrated that when taxpayers perceived the tax agency to be free of any political influence,
then they also perceived the tax administration as more fair. Also, when taxpayers perceived the
personnel of the tax agency to be more free of political influence, then they also perceived the tax
agency to be more fair towards taxpayers. In other words, taxpayers perceived autonomous tax
agencies as fairer and less politicized in comparison to traditional line agencies closely
connected to MoFs. The politicization and patronage problem is especially severe for tax
administrations owing to large-scale patronage opportunities available. These political
appointments result in selection of wrong person for the right job, and high turnover and
corruption.
The URA represented a SARA case which provides a good example of how continuing
interference from MoF led to inadequate progress in controlling political patronage. The URA
162
struggled to reach even minimum level of autonomy from MoF. The URA was supposed to
translate into loss of considerable autonomy and diminished patronage benefits for its MoF,
which in turn became strong resistance against URA’s successful implementation, and was able
to effectively over-turn the reform process and win back autonomy and patronage benefits (Clark
& Wood, 2001; Taliercio, 2004). In practice, the URA was not granted the autonomy as stipulated
in the legal framework, which resulting in continuing interference by MoF, in the form of one-sided
decisions affecting operations in URA. The MoF continued to exercise undue powers to meddle
with the funding of URA each year, despite the legal provision stipulating stable funding. In
addition, the politicians in MoF, who were empowered to approve the budget of URA on an
annual basis, used this autonomy to further their political interests by politicizing high level
officials in URA. Political interference also resulted in large number of political appointments in
URA, which not only sabotaged merit-based appointments, but also made it difficult to get rid of
corrupt but politically-aligned employees. Managers reported problems of politicization where
employees dismissed on misconduct continued to report to work by relying on the power of their
strong political connections. In case of URA, it seems that an improvement in compensation
mechanism, in the face of continuing political interference, simply resulted in increase in
patronage problem. It also highlighted how MoF undermined de-politicization of tax officials in
URA by not enabling enough autonomy to exercise effective accountability (Hall & Jenkins, 1995;
Fjeldstad et al., 2003; Fjeldstad & Moore, 2009; Martinez-Vazquez et al., 2004; Terkper, 1999;
Therkildsen, 2004).
Peru represented one of the very few SARA cases which were able to work without political
interference, albeit only initially, with deterioration in de-politicization experienced over time.
SUNAT’s resistance to political interference in appointments, through a merit-based recruitment
drive, was largely due to political support from the highest level i.e., the president. President
163
Fujimori showed pronounced interest in SUNAT’s reform for quickly improving revenue
performance. Owing to commitment shown towards SARA reform from the highest level of
government, SUNAT demonstrated the value of merit-based recruitment reforms towards
reducing politicization of tax officials. After its inception, more than 97% of its employees were
selected according to a merit-based recruitment policy. As a result, Peru emerged, at least for a
limited time, as a SARA largely free from patronage pressures. Starting in 1995, the MoF started
meddling with SUNAT’s affairs, in the wake of diminishing presidential support. It stripped
SUNAT of its influence over tax policy, and patronage pressures led to penetration of political
appointees, mainly agents of the National Intelligence Service into SUNAT. The decline in
governmental support also impacted progress against corruption due to declined effectiveness of
judicial system and police support, negatively affecting the institutional support needed for any
concrete progress against corruption. In Peru’s case, deterioration in political support not only
resulted in increased politicization of tax officials, but also impacted the revenue performance
negatively. Hence, SUNAT’s tax-to-GDP ratio followed a downward trend after 1997. In case of
Peru, although the reform took off well, but increasing interference from MoF gave way to
politicization of tax officials over the long run (Jenkins, 1994; Durand & Thorp, 1998; Fjeldstad &
Moore, 2009; Mann, 2004; World Bank, 2001).
In Tanzania, the TRA was created with the objective of de-politicization of tax officials by
following a merit-based recruitment mechanism. As the opportunities for patronage and misuse of
merit based recruitment system become higher with the involvement of MoF in the selection of
head of SARA. So, the president directly appointed the CEO of the TRA, as well as constitution
of a governing board for the supervision of TRA’s management. The effectiveness of the merit-
based recruitment system was undermined in case of TRA due to wrongdoings in the application
process. Although all of its employees had to go through a recruitment cleaning exercise, which
164
got rid of almost 1200 employees mostly for misconduct, nonetheless senior managers were still
able to hire their favorites by twisting the application process (Osoro et al., 1999; Fjeldstad et al.,
2003; Mann, 2004; Martinez-Vazquez et al., 2004). In case of Kenya as well, although in theory
the management of the KRA was entrusted with autonomy to effectively fend off patronage
pressures. But minister of finance, in this case as well, misused his capability to appoint
governing board members by hiring his past colleagues for these positions. In this case as well,
MoF undermined de-politicization of governing board in KRA by cherry-picking its board
members based on favouritism. In case of both Tanzania and Kenya, SARAs experienced
corruption-led form of micromanagement by MoFs. Because politicians did not wanted to let go of
autonomy to SARAs, as it also entailed letting go of possible opportunities for patronage and
corruption. Conversely, deeply entrenched corruption in MoFs crawled back into SARAs in the
guise of political appointees by politicians (Delay et al., 1998; Jenkins & Khadka, 2000; Kidd &
Crandall, 2006).
effective accountability resulting in de-politicization of tax officials for SARAs analysed above.

resulting in De-politicization of Tax Officials
Effective Accountability and De-

SARA Country
politicization of Tax Officials*
Peru -
Tanzania -
Uganda -
Kenya -
*+ means SARA design component of effective accountability was successful to de-politicize tax officials in the
SARA country, with no interference from MoF.
- means SARA design component of effective accountability was unsuccessful to de-politicize tax officials in the
SARA country, due to interference from MoF.
165
undermining de-politicization of tax officials. As per discussions in the macro level of SARA
analysis, all analyzed cases including Peru, Tanzania, Uganda, and Kenya showed less than
optimal progress towards controlling opportunities for corruption, due to interference from MoFs,
undermining SARA design component of effective accountability to de-politicize tax officials for
SARAs. All analyzed SARA cases lend support to the basic argument behind research
hypothesis 2(b) as per macro level of SARA analysis. These SARA cases made less than optimal
use of effective accountability to de-politicize tax officials and were less effective in controlling
opportunities for corruption. It is worthwhile to highlight that the findings obtained for research
hypothesis 2(b) are very consistent with the findings reached for research hypothesis 2(a). This is
partly due to inclusion of same SARA cases for both sub-hypotheses. For research hypothesis
2(b), the macro level of SARA analysis provided considerable support for four out of four SARAs
examining how presence of MoFs led to continuation of politicization of tax officials in SARAs.
Hence, research hypothesis 2(b) was considerably supported for four out of total four SARA
cases.
5.2.3 Effective Accountability Controlling Opportunities for Corruption by Reducing
Discretionary Powers of Tax Officials
Research hypothesis 2(c) proposed that SARAs have been ineffective in controlling opportunities
for corruption due to ineffective accountability by MoF by not reducing discretionary powers of tax
officials. This section aims to identify and analyze secondary literature on SARAs with particular
focus on if and how effective accountability is being utilized by SARAs to control opportunities for
corruption by reduction of discretionary powers of tax officials. The development of research
hypothesis 2(c) in the analytical framework enabled to propose that if SARAs have made
progress in terms of ensuring effective accountability by reduction of discretionary powers of tax

166
officials, then it can also be taken to propose that opportunities for corruption have been
secondary literature analysed to examine research hypothesis 2(c).
Table 5.15: Overview of Secondary Literature concerning Effective Accountability and

Reduction of Discretionary Powers of Tax Officials for SARAs
Selected Literature concerning Effective Accountability and Reduction of Discretionary

Powers of Tax Officials for SARAs
Delay et al., 1998 McCarten, 2006
Fjedstad & Moore, 2009 Taliercio, 2004, 2004a
Fjeldstad & Rakner, 2003 Terkper, 1999
Fjeldstad, 2005b World Bank, 1999
Kidd & Crandall, 2006
Tanzania represented a SARA case where the ineffectiveness of its accountability mechanisms
to control opportunities for corruption, led to increase in discretionary and monopoly powers
available to tax officials. As highlighted before, in order to purge corrupt employees, TRA
underwent a recruitment cleaning exercise, which resulted in laying-off of almost one third of its
total employees. In spite of this, the TRA failed to reduce discretionary powers available to tax
officials and break existing corruption networks within the revenue authority. The recruitment
managers required bribes to be paid by the employees to have their applications supported by
the management. In addition to bribes received, this also enabled the managers to find
employees who were willing to collude in corruption networks. This situation also highlighted that
behind a façade of a merit-based recruitment drive, potential employees continued to perceive
TRA as a lucrative place for corruption and were willing to bribe for a place in it. For TRA, the
efforts to improve the integrity of its employees by depoliticizing the appointment procedure
through a merit-based recruitment exercise were undermined by the ineffective accountability
and discretionary powers available to recruiting managers. In case of Tanzania, the TRA
167
remained ineffective in ensuring meritocracy or de-politicization of the recruitment procedure on
account of discretionary powers available to recruitment managers. This scenario can also be
utilized to explain why Tanzania was not able to effectively control opportunities for corruption
even after TRA reform. TRA represented a case where increased autonomy for hiring managers
was not balanced with increased effective accountability mechanisms. Rather autonomy levels
granted to recruiting managers were greater than as could be effectively controlled with available
accountability. This occurrence led to conversion of increased autonomy under TRA into
discretionary and monopoly powers available to recruiting managers in the face of ineffective
accountability mechanisms, leading to increases in opportunities for corruption (Delay et al.,
1998; Fjeldstad, 2005b; Fjedstad & Moore, 2009; Fjeldstad & Rakner, 2003; Terkper, 1999;
World Bank, 1999).
The Mexican SARA was representative of complete dominance by MoF of SAT’s board, leading
to no value addition by the board to the effectiveness of accountability arrangements of SAT by
oversight bodies other than MoF. The board of directors of SAT comprised of total six members,
including the minister of finance, and the president of the SAT, and both of these nominating two
members each from the MoF and SAT. In essence, all six members were either for the MoF or
the SAT, and there were no private sector members. The legislation placed the board completely
under the control of minister of finance by providing the minister with full control over the board,
although the legislation also bestowed the board with important decision making powers. In
practice, the board did not play any significant role towards the management or policy making,
owing to the composition of the board. The composition of the board did not enabled any value
addition to the accountability mechanisms, since all members of the board were already part of
top management of SAT and MoF anyway. Owing to MoF’s complete control over SATA’s board,
the board’s presence seemed unnecessary. The SAT’s board was generally perceived to be
168
inessential and ineffective, it met very rarely over time (all members meet-up in other formal
meetings anyway) and its decision making powers were openly encroached by the minister of
finance without any official consent of the board. Taliercio (2004, 2004a) also highlighted how
interference by the MoF in SAT’s autonomy negatively impacted revenue performance, with
decline in Tax-to-GDP ratio over time. In view of MoF’s complete control of the SAT’s board, it
raised questions about any value added to the effectiveness of the accountability mechanisms of
SAT (Kidd & Crandall, 2006; McCarten, 2006).
In Venezuela, the SENIAT employed a superintendent model of governance. The minister of
finance was bestowed with complete authority of selecting and dismissing both the
superintendent and adjunct superintendent of SENIAT. This mechanism very clearly
subordinated the SARA to the goodwill of the minister of finance. This scenario resulted in
minister of finance usually getting his way by the superintendent of SENIAT, and on several
occasions superintendents were simply removed by the minister over disagreements, depicting
poor security of tenure. Consequently, the average term of superintendents in SENIAT was less
than a year (10.5 months) over six years period of 1994-2000. In 1999, the MoF almost took over
SENIAT by not appointing any superintendent over a considerable period of time, and exercised
direct control over the revenue authority. Within few years of the reform, MoF was able to
circumvent the autonomy of SENIAT by fully controlling its superintendents, and in the wake of
insufficient checks over minister’s control over SARA, the MoF emerged as the main overseer of
SENIAT in practice. In its early years though, the SENIAT was bestowed with the goodwill of a
pro-reform minister of finance, whose support enabled the launch of SENIAT reform. This
support quickly disappeared with the change of minister of finance within next two years.
Taliercio (2004, 2004a) also highlighted how interference by the MoF in SENIAT’s autonomy
negatively impacted revenue performance, with decline in Tax-to-GDP ratio over time. The
169
SENIAT was representative of complete dominance by MoF of its superintendents, leading to no
value addition by the superintendent to the effectiveness of accountability arrangements of
SENIAT by oversight bodies other than MoF (Kidd & Crandall, 2006).
effective accountability resulting in reduction of discretionary powers of tax officials for SARAs
analysed above.

resulting in Reduction of Discretionary Powers of Tax Officials
Effective Accountability and Reduction of

SARA Country
Discretionary Powers of Tax Officials*
Tanzania -
Mexico -
Venezuela -
*+ means SARA design component of effective accountability was successful to reduce discretionary powers of tax
officials in the SARA country, with no interference from MoF.
- means SARA design component of effective accountability was unsuccessful to reduce discretionary powers of tax
officials in the SARA country, due to interference from MoF.
undermining reduction of discretionary powers of tax officials. As per discussions in the macro
level of SARA analysis, all analyzed cases including Tanzania, Mexico and Venezuela showed
less than optimal progress towards controlling opportunities for corruption, due to interference
from MoFs, undermining SARA design component of effective accountability to reduce
discretionary powers of tax officials for SARAs. All analyzed SARA cases lend support to the
basic argument behind research hypothesis 2(c) as per macro level of SARA analysis. These
SARA cases made less than optimal use of effective accountability to reduce discretionary
powers of tax officials and were less effective in controlling opportunities for corruption. It was
found that these SARA countries, which were not able to effectively control opportunities for
170
corruption, were represented by an instance where autonomy levels granted to tax officials after
reforms were greater than as could be effectively controlled by existing accountability levels. In
this case, this un-checked autonomy concept transformed into a combination of monopoly plus
discretionary powers available to tax officials, as not controlled by effective accountability
mechanisms.For research hypothesis 2(c), the macro level of SARA analysis provided
considerable support for three out of three SARAs examining how presence of MoFs resulted in
increases in discretionary powers available to tax officials in SARAs. Hence, research hypothesis
2(c) was considerably supported for three out of total three SARA cases.
5.2.4 Synthesis of Analysis for Research Hypothesis 2
As highlighted at the start of this section, the macro level of SARA analysis was aimed to analyse
research hypotheses 1 and 2. Research hypothesis 2 proposed that SARAs have been
ineffective in controlling opportunities for corruption due to ineffective accountability by MoF.
Research hypothesis 2 comprised of three further sub-hypotheses (2a, 2b and 2c), which have
already been examined in sections above. As it was proposed at the start of section 5.1, analysis
of these three sub-hypotheses will be undertaken with an aim to test research hypotheses 2, and
any answers or findings as per analysis of these sub-hypotheses will mean to respond to
research hypothesis 2. Consequently this section aims to synthesize all three sub-analyses
discussed in sections above for the cumulative analysis of research hypothesis 2.
Table 5.17 illustrates the accumulation of all three sub-analyses discussed in sections 5.2.1,
5.2.2 and 5.2.3 of the macro level of SARA analysis with respect to the link between effective
accountability and all three preventive strategies to control opportunities for corruption.
171
Table 5.17: Overview of Selected SARA Countries concerning Effective Accountability and
all three preventive strategies to control opportunities for Corruption
Effective Accountability
SARA Country Reduction of
Introduction of
De-politicization Discretionary
Oversight
of Tax Officials Powers of Tax
Mechanisms
Officials
Peru - - n.d
Tanzania - - -
Mexico n.d n.d -
Uganda - - n.d
Kenya - - n.d
Venezuela n.d n.d -
+ means SARA design component of effective accountability resulted in increase in the specific preventive strategy
to control opportunities for corruption, with no interference from MoF, as per macro level of SARA analysis in the
sub-proposition analyses.
- means SARA design component of effective accountability did not resulted in increase in the specific preventive
strategy to control opportunities for corruption, due to interference from MoF, as per macro level of SARA analysis in
the sub-proposition analyses.
strategies to control opportunities for corruption. This table also summarizes the findings in terms
of individual sub-hypotheses. As evident from the table, almost one third of cells (7/18) display
non-availability of comparative data, which constraints comprehensive analysis for some SARA
countries in terms of all three preventive strategies in support of research hypothesis 2. In terms
of individual sub-hypotheses, research hypothesis 2 was provided considerable support by sub-
hypothesis 2(a) {support for 4/4 SARAs}, considerable support by sub-hypothesis 2(b) {support
for 4/4 SARAs}, and also considerable support by sub-hypothesis 2(c) {support for 3/3 SARAs}.
Hence, it is pertinent to conclude that as per individual sub-hypotheses, research hypothesis 2
was provided considerable support.
This section was aimed at analyzing the overall proposition that those SARAs which had
ineffective accountability due to interference form MoFs would also be ineffective in controlling
172
opportunities for corruption. This supposition was considerably supported above. The cumulative
finding for research hypothesis 2 has demonstrated that SARA design component of effective
accountability, which is adopted to enhance the accountability of the reform model, has proved to
be ineffective due to presence of MoF in the accountability equation of SARAs, and this in turn
has led to increases in opportunities for corruption. This section questioned why SARAs have
been unable to adopt effective accountability? It was found that presence of MoFs in the
accountability arrangements have rendered SARAs ineffective in controlling opportunities for
corruption. Continued interference from MoFs was found to undermine not only the effective
oversight mechanisms of SARAs (sub-hypothesis 2a), but also found to lead to continuation of
politicization of tax officials (sub-hypothesis 2b), and resulting in increases in discretionary
powers available to tax officials (sub-hypothesis 2c). This finding suggesting how SARAs’
accountability mechanisms were undermined due to interference by MoFs, point towards a case
for not keeping SARAs directly accountable to MoFs, in the presence of other effective oversight
bodies including Revenue Boards/Superintendents and Parliament etc.
It is very important to highlight here that the macro analysis of SARAs represents only a
systematic review coming up with tentative findings and cannot claim the rigour of primary data
studies. Since the researcher lacked resources of time and money to carry out in-depth case
studies of selected SARA cases in pursuit of primary data, the researcher was left with no other
choice but to rely on secondary literature. Similarly the findings obtained for the hypothesized
relationships in the macro analysis are also tentative and only suggest that relationship might
exist. This also points towards future research in this area where primary data might be collected
to analyze the validity of analytical framework for SARAs. Nonetheless, the macro analysis has
helped to build some sort of picture to understand how SARAs’ implementation against corruption
173
fared so far in some developing countries. Doing so also sets the stage for micro level of analysis
where the tax administration reforms for Pakistan are examined in detail.
As highlighted at the start of this chapter, the macro level of SARA analysis was aimed to analyse
research hypothesis 1 and 2. Consequently this section aims to integrate all sub-analyses
discussed in sections above for the cumulative analysis of research hypothesis 1 and 2. Figure
5.2 presents the graphical representation of overall findings of macro level of SARA analysis for
research hypothesis 1 and 2.

174
Figure 5.2 Graphical Representation of Findings of Macro Level of SARA Analysis
Increases in Wages
and Rewards
Increases in the 4/8

3/5
Personnel RH 1(c) Increases in and
Motivations for
Autonomy Research Hypothesis 1 Stricter Enforcement of 3/6
Corruption
Penalties for Corruption
3/4
Instilling Ethics in Tax
Officials
Reduction in
Corruption
Effective Introduction of
RH 2(a)
Accountability Research Hypothesis 2 Oversight Mechanisms
4/4 Opportunities for
Corruption
De-politicization of Tax 4/4
Officials
3/3
Reduction of
Discretionary Powers of
Tax Officials
RH = Research Hypothesis
175
As illustrated in figure, in terms of individual sub-hypotheses, research hypothesis 1 was provided
limited support by sub-hypothesis 1(a) {support for 4/8 SARAs}, limited support by sub-
hypothesis 1(b) {support for 3/5 SARAs}, limited support by sub-hypothesis 1(c) {support for 3/6
SARAs}, and also limited support by sub-hypothesis 1(d) {support for 3/4 SARAs}. Hence, it is
pertinent to conclude that as per individual sub-hypotheses, research hypothesis 1 was only
provided limited support. In terms of individual sub-hypotheses, research hypothesis 2 was
provided considerable support by sub-hypothesis 2(a) {support for 4/4 SARAs}, considerable
support by sub-hypothesis 2(b) {support for 4/4 SARAs}, and also considerable support by sub-
hypothesis 2(c) {support for 3/3 SARAs}. Hence, it is pertinent to conclude that as per individual
sub-hypotheses, research hypothesis 2 was provided considerable support.
Next, Table 5.18 illustrates the accumulation of all sub-analyses discussed in sections 5.1.5 and
5.2.4 of the macro level of SARA analysis for research hypotheses 1 and 2.
176
Table 5.18: Overview of Selected SARA Countries accumulating sub-analyses of the macro level of SARA analysis for research hypotheses 1
and 2
SARA Country Research Hypothesis 1 Research Hypothesis 2
Personnel Autonomy Effective Accountability
RH 1(a) RH 1(b) RH 1(c) RH 1(d) RH 2(a) RH 2(b) RH 2(c)
Increases in Increases in the Increases in and Instilling Ethics Introduction of De-politicization of Reduction of Discretionary
Wages and Probability of Stricter in Tax Officials Oversight Tax Officials Powers of Tax Officials
Rewards Detection Enforcement of Mechanisms
Penalties
Ghana + n.d n.d n.d n.d n.d n.d
South Africa + + + n.d n.d n.d n.d
Peru + + + + - - n.d
Tanzania + + + + - - -
Mexico - n.d n.d + n.d n.d -
Guatemala n.d + + n.d n.d n.d n.d
Uganda + n.d + n.d - - n.d
Kenya + + + + - - n.d
Venezuela - n.d n.d n.d n.d n.d -
+ means personnel autonomy resulted in increase in the specific preventive strategy to control motivations for + means SARA design component of effective accountability resulted in increase in the
corruption, as per macro level of SARA analysis in the sub-proposition analyses. specific preventive strategy to control opportunities for corruption, with no interference from
- means personnel autonomy did not resulted in increase in the specific preventive strategy to control motivations for MoF, as per macro level of SARA analysis in the sub-proposition analyses.
corruption, as per macro level of SARA analysis in the sub-proposition analyses. - means SARA design component of effective accountability did not resulted in increase in
the specific preventive strategy to control opportunities for corruption, due to interference
from MoF, as per macro level of SARA analysis in the sub-proposition analyses.

177
strategies to ensure personnel autonomy and effective accountability for controlling motivations
and opportunities for corruption. To explore why some SARAs have failed to effectively control
opportunities for corruption, a detailed mapping of SARA cases, labelled as the macro level of
SARA analysis, was undertaken in this chapter. This chapter applied the analytical framework to
selected developing countries in terms of SARAs’ effect on corruption, and gauged the
robustness and generalizability of the analytical framework by analysing different SARA cases
under a same lens. In a nutshell, the specific contribution of chapter 5 to the overall thesis
argument lies in aiming to answer the question set out above by conceptually testing the Anti-
Corruption SARA framework by analyzing individual SARA design components in their separate
contributions towards controlling motivations and opportunities for corruption. This was to
contribute towards filling a gap in literature for studies on SARAs differentiating between
motivations and opportunities for corruption. It is concluded that SARAs preferred adoption of
personnel autonomy over ensuring effective accountability for controlling corruption. In other
words, SARAs preferred adoption of autonomy-enhancing design components over
accountability-enhancing design components for controlling corruption. This imbalance led to
some progress in controlling motivations for corruption, but failing to effectively control
opportunities for corruption. This chapter also questioned why SARAs have been unable to adopt
effective accountability? It was found that presence of MoFs in the accountability arrangements
rendered SARAs ineffective in controlling opportunities for corruption. Continued interference
from MoFs was found to undermine not only the effective oversight mechanisms of SARAs, but
also lead to continuation of politicization of tax officials, and resulting in increases in discretionary
powers available to tax officials.

178
In sum, by conducting the macro level of SARA analysis, progress was made towards a two part
answer to the main question of the thesis. To explore why some SARAs have failed to effectively
control opportunities for corruption, it was found that SARAs have made partial progress to
control corruption by focusing more on controlling motivations for corruption (through autonomy-
enhancing design components) and lesser focus on controlling opportunities for corruption
(through accountability-enhancing design components). Continued interference from MoFs was
found to undermine not only effective accountability for SARAs, but also undermining control of
opportunities for corruption. These findings are significant as they not only highlighted ‘why’
opportunities for corruption were not controlled by SARAs due to imbalance in adoption of
personnel autonomy over effective accountability. But also showed ‘how’ this imbalance rose
due to continued interference from MoFs. The macro level of SARA analysis was also aimed to
offer propositions about why certain SARA design components are more effective against
corruption than others, and to offer specific recommendations to improve SARAs’ effectiveness
against corruption. The macro level of SARA analysis was also aimed at improving, revising and
re-forming the existing SARA reform model, to make it more robust against corruption, by
incorporating changes in existing SARA model as identified by analysis. It is pertinent to conclude
here that balancing both autonomy and accountability-enhancing SARA design components is
highly warranted to effectively control both motivations and opportunities for corruption. Also,
findings suggesting how SARAs’ accountability mechanisms were undermined due to
interference by MoFs, leading to ineffectiveness in controlling opportunities for corruption, point
towards a case for not keeping SARAs directly accountable to MoFs, in the presence of other
effective oversight bodies including Revenue Boards/Superintendents and Parliament etc.
To explore why some SARAs have failed to effectively control opportunities for corruption, a two-
staged analysis of SARAs was proposed to be conducted in chapters five and six of the thesis.
179
Next, in chapter six of the thesis, a detailed case study of tax administration reforms in Pakistan
will be conducted as per analytical framework developed in this study. Two main sources of data
will be employed for the case study at the micro level of analysis: document analysis (secondary
literature) and semi-structured interviews.

180
CHAPTER SIX
MICRO LEVEL OF ANALYSIS: CASE STUDY OF TAX ADMINISTRATION REFORMS IN
PAKISTAN
6.0 INTRODUCTION
In the previous chapter, a detailed mapping of SARA cases, labelled as the macro level of SARA
analysis was undertaken for conceptually testing the Anti-Corruption SARA framework for
multiple SARA countries. This analysis helped to conclude that some SARAs have made partial
progress to control corruption by focusing more on controlling motivations for corruption (through
autonomy-enhancing design components) and lesser focus on controlling opportunities for
corruption (through accountability-enhancing design components). Taking the form of findings,
chapter seven helped to identify factors, which could be analyzed in advance to SARA reform,
making a developing country suitable or unsuitable candidate for SARA reform. Such as
continued interference from MoFs was found to undermine not only effective accountability for
SARAs, but also undermining control of opportunities for corruption.
Building onto the discussions raised in chapter five, this chapter aims to conceptually test the
Anti-Corruption SARA framework for FBR, by conducting a detailed case study of tax
administration reforms in Pakistan. In spite of increasing proliferation of SARAs in developing
countries for controlling corruption in tax administrations, the researcher has not come across
any significant research on the practical side of designing and implementing SARA reform
against controlling corruption. It is this gap that this research aims to contribute towards filling by
doing a practical work of designing an Anti-Corruption SARA framework (chapter 3) and applying
it to Pakistan in this chapter.

181
This level of analysis is labeled as micro level of analysis, as this proposed analysis will enable to
examine in detail the current state of affairs and potential and suitability of SARA reform for
Pakistan. While the macro level of analysis was aimed at questioning why some SARAs have
failed to effectively control opportunities for corruption. The micro level of analysis aims to
question why tax administration reform failed to effectively control opportunities for corruption in
FBR Pakistan. This chapter intends to apply the Anti-Corruption SARA framework to FBR to
gauge the effect of tax administration reform on corruption, and will offer specific
recommendations to improve progress against corruption.
Two main sources of data were employed for the case study at the micro level of analysis:
document analysis (secondary literature) and semi-structured interviews. The triangulated
methodology collects two types of data (documentary and interview) for cross-checking. Hence,
the micro level of analysis for FBR was aimed at theorizing what is going wrong with practice by
approaching both new and old evidence (semi-structured interviews and secondary literature) in
a new way (i.e., analytical framework). The information obtained from the interviewees through
semi-structured interviews was analysed to show interviewees' views about why tax
administration reform failed to effectively control opportunities for corruption in FBR Pakistan. The
views of significant actors were also solicited to explore potential and suitability of SARA reform
for Pakistan. In particular these interviews were aimed to gauge if opportunities for corruption
have not been controlled in FBR due to imbalance in adoption of personnel autonomy over
effective accountability, and if this imbalance arisen due to continued interference from MoF in
FBR. In total the researcher conducted 31 semi-structured interviews with 16 tax officials, 2
journalists, 3 MoF officials, 5 taxpayers and 5 businessmen/traders. The two journalists included
Rehman Azhar (Dunya News) and Imdad Bhatti (Jang News). The three MoF officials were
gazetted officers and occupied mid-level positions in Ministry of Finance in Islamabad, Pakistan.
182
They were promised anonymity in exchange for interview. The five taxpayers included friends of
researcher and were chosen because the researcher was certain that they actually paid taxes
and were aware with tax reforms in Pakistan. Their identities will also be kept anonymous. The
five businessmen represented prominent traders and were contacted for interview through a
contact working at Lahore Chambers of Commerce and Industry. Detailed discussion about the
process of selection and other details for interviews has already been discussed in chapter four
(Section 4.3.1). Thus, this chapter reports the results of semi-structured interviews which were
used to acquire the views of 31 respondents. In addition to semi-structured interviews, the micro
level of analysis was also supplemented by the documentary analysis of secondary literature. In
particular newspaper analysis over a period of three years (2011 to 2013) and collection of other
relevant documents from FBR Headquarters; international donor organizations such as the World
Bank, IMF, ADB; NGOs such as Transparency International; ministerial and non-ministerial
reports was undertaken. Further details about the process of newspaper analysis and collection
of other documents have already been discussed in chapter four (Section 4.3.2)
In order to analyze the validity of the framework for FBR, the micro level of analysis in this
chapter will also be examined in terms of sub-hypotheses, with an aim to test both research
hypothesis 1 and 2, and any answers or findings as per analysis of these sub-hypotheses will
mean to respond to research hypothesis 1 and 2. Further, by examining individual sub-
hypotheses, this analysis also aims to segregate FBR’s progress against corruption towards
separately controlling motivations and opportunities for corruption. In particular, this case study
aims to confirm if the findings reached for SARA countries in the macro analysis are also
applicable to FBR. The case study will try to analyze if opportunities for corruption have not been
controlled in FBR due to imbalance in adoption of personnel autonomy over effective
accountability, and if this imbalance arisen due to continued interference from MoF in FBR as
183
well. Such analysis will help to conclude if there is potential and suitability of adopting Anti-
Corruption SARA framework for FBR to control corruption in tax administration of Pakistan.
The rest of this chapter is organized as follows: Section 6.1 will detail the history and status of tax
administration reform in Pakistan. This section will specifically examine the question if tax
administration reform has failed to effectively control opportunities for corruption in FBR Pakistan.
Sections 6.2 and 6.3 will contain micro level of analysis, where the individual sub-hypotheses will
be tested for FBR Pakistan. Section 6.4 will summarize the key findings arising out of micro level
analysis and concludes. This section will also integrate the findings of all sub-hypotheses for two
main research hypotheses of this study.
The structure of Chapter 6 is shown in Figure 6.0.

184
6.0
INTRODUCTION
6.1
HISTORY AND STATUS OF TAX
ADMINISTRATION REFORMS IN PAKISTAN
6.2 6.3
PERSONNEL AUTONOMY CONTROLLING INEFFECTIVENESS OF ACCOUNTABILITY TO
MOTIVATIONS FOR CORRUPTION IN FBR CONTROL OPPORTUNITIES FOR CORRUPTION
IN FBR
6.2.1 6.3.1
Ineffectiveness of Accountability to Control
Personnel Autonomy Controlling Motivations
Opportunities for Corruption in FBR by Failing
for Corruption in FBR by Increasing Wages
and Rewards to Introduce Effective Oversight Mechanisms
6.2.2 6.3.2
Personnel Autonomy Controlling Motivations Ineffectiveness of Accountability to Control
for Corruption in FBR by Increasing the Opportunities for Corruption in FBR by Failing
Probability of Detection to De-politicize Tax Officials
6.2.3 6.3.3
Personnel Autonomy Controlling Motivations Ineffectiveness of Accountability to Control
for Corruption in FBR by Increases in and Opportunities for Corruption in FBR by Failing
Stricter Enforcement of Penalties for to Reduce Discretionary Powers of Tax
Corruption Officials
6.2.4 6.3.4
Personnel Autonomy Controlling Motivations Integration of Sub-Analyses for Research
for Corruption in FBR by Instilling Ethics in Tax Hypothesis 2
Officials
6.2.5
Integration of Sub-Analyses for Research
Hypothesis 1
6.4
185
6.1 HISTORY AND STATUS OF TAX ADMINISTRATION REFORMS IN PAKISTAN
As per World Justice Project (WJP) Pakistan stands as the seventh most corrupt nation in the
world out of total 97 countries analysed for year 2012 (The News, 2012a). An example of the
scale of system-wide corruption in tax system of Pakistan can be illustrated by the case of
Osama-bin-Laden. Even the most wanted terrorist of the world could not escape corruption in
taxation in Pakistan as he had to bribe Rs. 50,000 to the local revenue official (locally known as
Patwari) to obtain permission to build the compound in Pakistan. As dully noted in his diary, only
after taking the bribe, Osama was allowed to build a three-floor compound, complete with 14 feet
high outer wall along with iron fencing. Osama was not only aware of the bribing practice, but
also gave explicit approval to bribe the Patwari (The News, 2012b). The Chief of National
Accountability Bureau (NAB), Pakistan’s central anti-corruption agency, made headlines in 2012
by projecting an eye watering daily figure of corruption in Pakistan of around Rs. 13 billion (or
almost 5 trillion a year). Out of this total corruption, a figure of Rs. 7 billion per day (Rs. 2500 to
3000 billion per year) was given on account of tax evasion (Raza, 2012). Adding to these figures,
the Chief added:
“The nexus between the legislature and executive which aggravated corruption in the
80s and 90s has now become monstrous. Corruption in Pakistan is no longer a party-
centric or incident-centric phenomenon, but is now an attitude across the board. A flood
of corruption is flowing in which there are fish and crocodiles, but rather than catching
them, we have to tap the flow of corruption so that the fish and crocodiles die a natural
death when the flow of corruption stops” (The News, 2012c, p.1).
These statements did not sit well with the government of Pakistan at that time, and a body was
formed to probe these claims. The Chief was ultimately removed in the next few months.
In Pakistan no one has been impeached for personal income tax evasion in the last 25 years,
says chairman FBR (The News, 2012d). In case of Pakistan the problem of tax evasion starts at
186
the top. The ones who make revenue policies, run the government, and collect taxes have not
been demonstrating exemplary behavior. In 2011, only 90 out of total 341 members of the
national assembly of Pakistan filed tax returns, i.e., only 33% of parliamentarians, and 38% of
cabinet ministers complied with their tax responsibility. Almost 70% of Pakistan’s MPs did not file
a tax return in 2011, including the President Asif Ali Zardari (Ali, 2012; The News, 2012a). In
2008, 61% of the lawmakers in Pakistan did not paid anything on account of taxes. In the same
year, the prime minister, all his cabinet members, including the minister of finance did not paid
any income tax (Cheema, 2012a). Upon media furor, rather than discharging their tax
responsibility and paying taxes, politicians openly criticized FBR of unlawfully releasing data.
FBR, on the other hand, rather than taking any legal action against tax crooks, guaranteed full
action against the persons involved in releasing data. A former chairman of FBR described the
Pakistan taxation system as skewed, whereby the poor subsidize the rich:
“This is a system of the elite, by the elite, and for the elite” (Ali, 2012, p. 1).
Pakistan is characterized by a country where the political parties do not file tax returns as an
across-the-board practice, so much so that FBR has never issued notices to any party (Bukhari &
Haq, 2013). The situation is so bad that many leading cricketers don’t file tax returns, taking the
plea that they get their income after tax is deducted at source, although they are legally bound to
file tax return even if the income is deducted for tax at source, for declaring any other sources of
income (Khan, 2012b).
Keeping in view the dismal state of affairs projected above, Pakistan has made numerous
attempts to reform its tax machinery over years since its independence. Before independence,
the Central Board of Revenue (CBR) was formed on 1 st April 1924 via enactment of Central
Board of Revenue Act, 1924. In 1944, a bigger revenue division was established underneath
187
MoF. After independence in 1947, this organization continued to operate till 31 st August 1960,
when CBR was reorganized as an attached department of the MoF, on the recommendations of
the administrative re-organization committee. By 1974, the post of chairman CBR was
established to further rationalize the organization and functions of CBR. This post was equivalent
in status to an ex-officio additional secretary. Consequently, this post replaced the secretary
finance who exercised responsibilities as ex-officio chairman of the CBR. Over the next thirty
years, CBR faced further reorganizations. The status of CBR as a revenue division under MoF
was returned on 22nd October, 1991, to eliminate barriers in the exercise of powers by the
secretary, and for effective creation and implementation of fiscal policy actions. However, the
CBR relapsed to the pre-1991 situation, when the revenue division was abolished in January
1995. Another summersault came on 1st December, 1998, when the revenue division was again
restored. With the enactment of FBR Act 2007, the CBR was replaced with FBR in July 2007. In
terms of autonomy, the status of FBR lies in between a revenue authority and a centralized
government department working under close supervision of the ministry. FBR is neither a SARA
nor a conventional government department. As per World Bank (2004b) case of converting CBR
into a SARA for Pakistan, i.e., creation of a ‘Pakistan Revenue Authority’ was considered at the
start of reform exercise. But ultimately a decision was taken in favour of an organizational
arrangement of FBR which would remain under governmental control, but with significant
autonomy in terms of recruitment, compensation, and promotions etc. The decision to go with this
‘contradictory’ or ‘confused’ form of organizational arrangement, i.e., under governmental control
but autonomous, was attributed to following international best practices.
Although Pakistan has made numerous attempts to reform its inefficient and corrupt tax
administration, but a prominent reform effort came in the form of the Tax Administration Reform
Project (TARP) in 2005. This project was mutually funded by the World Bank ($100 million),
188
Department for International Development (DFID) ($23 million), and Government of Pakistan
(GoP) with a total financing of $135million. This reform was designed to implement the findings of
a stakeholders’ feedback exercise conducted by the CBR in 1999, and a perception gathering
survey conducted by the ‘Syed Shahid Hussain Committee’ in 2001. Both these surveys
confirmed that taxpayers perceived CBR as very unfair in its professional dealings, and very
corrupt. One of underlying reasons behind entrenched corruption was blamed on below-
subsistence level wages for tax officials. Some of the major objectives of TARP included
improving organisational efficiency and effectiveness of revenue administration; facilitating and
promoting voluntary compliance by taxpayers; increasing tax to GDP ratio; increasing
transparency and integrity; broadening the tax base; and strengthening audit and enforcement
procedures. The major thrust of the TARP reform project was laid on re-organization of the CBR
on functional lines, automation, and emphasis on human resource management (FBR, 2008).
After initiation of TARP program, how did Pakistan do with the reform? In its official evaluation,
the World Bank labelled the TARP program outcome as ‘unsatisfactory’ or failed project. The tax-
to-GDP ratio at the launch of TARP stood at 10.1% in 2005, but dropped to 9.3% in 2011 at the
project completion, which was not only very poor in Pakistan history, but also in comparison to
the world. In 2011, Pakistan had the second lowest tax-to-GDP ratio among 154 countries (ADB,
2013). According to FTO, 1% decrease in tax-to-GDP ratio cost the government of Pakistan a
massive $100 billion (approx.) in terms of tax collection. So, while FBR was busy with TARP to
improve tax collection, it was haemorrhaging $100 billion of taxes side by side. A House of
Commons Committee on International Development, while evaluating TARP in Pakistan noted
that although it was right to support the reform program by DFID, but it was poorly supervised
both by the World Bank and DFID. TARP was evaluated as “non-performing” by the World Bank
in 2008, with no one detecting that audit had been stopped within a year into reform. Work on
189
functional administration, which was the basic objective of TARP, had not even begun until 2009.
The committee concluded that TARP also failed due to very late interventions by the donors, not
willing to criticize the government in office. The major criticism on TARP was only brought
forward in 2008 when Musharraf government had been weakened (House of Commons, 2013).
The critics of TARP have labelled the reform exercise just a name change from CBR to FBR, with
the organization retaining its out-of-date character and dysfunctional corrupt culture, adding only
coatings of modernity over an outdated structure (Shah, 2011).
In order to salvage poor performance under TARP, another Tax Reform program under
International Monetary Fund (IMF) was initiated with the change of government in Pakistan in
2008. This was proposed by the then President Zardari to the ‘Friends of Pakistan’ conference in
New York in 2009, thus becoming the base of another IMF program of $11.3 billion minus any
conditions. The major thrust was proposed towards creation of arms-length tax administration
and elimination of Statutory Rules and Orders (SRO) powers. This program also collapsed in
2010, amid dispute between IMF and government of Pakistan for not undertaking promised
radical tax reforms. This program was evaluated as half-hearted, and failed due to strong
resistance from administration and tax officials of FBR alike, refusing to let go of rent-seeking
opportunities. The major failure was attributed to resistance by those in powers in Pakistan,
corrupt politicians and tax administrators alike, unwilling to abandon opportunities for corruption.
Another controversy came when Pakistan tried its hand at personnel reform by merging different
services in FBR into an integrated Inland Revenue Service (IRS) under the TARP project. As a
result a new occupational group was created by integrating income tax, sales tax, and federal
excise to administer domestic taxes. Previously, federal excise was administrated by the customs
group, who was not happy with the move, and complained against the World Bank for
constitutional violations. This move resulted in a formal dispute between the government of
190
Pakistan and the World Bank, with both parties blaming each other for violations. The customs
association held that by insisting Pakistan to create IRS, the World Bank not only desecrated its
own policies, but also distorted Pakistan’s constitution to their detriment. This resulted in World
Bank dissociating itself from the creation of IRS, taking the position that creation of IRS was
never part of World Bank reform programme. This left Pakistan into a predicament of how to
move ahead with the merger of different taxation groups. Pakistan maintained that the World
Bank was not only informed, but rather pressed on Pakistan towards creation of integrated IRS.
FBR Chairman said:
“For the World Bank to initially press for the FBR to undergo these reforms and make
them as benchmarks in GoP’s negotiations with IMF, and to now take adverse notice of
successful implementation by FBR/GoP, is totally inappropriate and undesirable” (Kiani,
2010, p.1).
Pakistan maintained that the establishment division issued notices of creation of IRS in 2009 with
the approval of the then prime minister, and on the advice of the World Bank and IMF. Another
setback towards creation of IRS came from Pakistan’s central recruitment agency, the Federal
Public Service Commission (FPSC) which refused to recognize IRS as an approved service
group. The FPSC maintained that it was never consulted by the government who singlehandedly
approved the creation of new group through the establishment division via office memorandum.
FPSC clarified that it was mandatory for the government to seek advice from the commission,
and the matter eventually went to court for violation of FPSC rules. This instance again
highlighted the half-hearted attempt to reform tax administration in Pakistan, lacking fair
treatment of all FBR staff affected by proposed reorganization.
After these failures in TARP implementation to reform tax administration, what is the response of
the government of Pakistan? Ironically, preparations are underway to bid for second phase of
TARP from the World Bank, after TARP-I ended miserably in 2011. The government of Pakistan
191
is currently busy in consultations for finalizing the roadmap for TARP-II with possible funding from
the World Bank, to complete the unfinished agenda which it failed to do so first time (The News,
2013b). The total cost of this adventure would be a credit of more than $300 million to be re-paid
back from the (very few) honest taxpayers’ money in Pakistan. It is also worthwhile to mention
here that the FBR is yet to examine in detail the factors which led to the failure of TARP-I, though
it is excitedly preparing for the unfinished reform business through TARP-II.
Pakistan’s poor progress and non-commitment towards tax reform has not gone un-noticed by
the international donor community, which has openly criticized Pakistan for the failures against
reform. The calls for making aid conditional to tax reform are emerging from several corners.
Former US Secretary of State, Hillary Clinton openly criticized Pakistan’s elite. She observed:
“The very well-off in Pakistan do not pay their fair share for the services that are needed,
in health and education primarily” (Iqbal, 2010, p. 14).
She pressed on the rich elite in Pakistan to pay more tax for self-sufficiency, and to lessen
Pakistan’s reliance on foreign aid. She expressed that Pakistan should not become “perpetually
dependent on financial dole” and noted that it is wrong to assume that American taxpayers could
forever pay for Pakistan’s economic and social development.
Similarly, the International Development Committee in the UK asked the British government to
hold back additional aid to Pakistan, unless Pakistan takes necessary steps to collect taxes from
its rich citizens. The committee recommended that the increase in British aid to Pakistan (from
£267 million to £446 million) should be put off, and made conditional to Pakistan collecting more
taxes from the rich and combating corruption effectively. The report also pressed on UK ministers
towards ensuring that aid was concentrated on anti-corruption efforts (Mustafa, 2013). Chairman
192
of the Commons International Development Committee, Sir Malcolm Bruce commented that
Pakistan’s approach towards aid has been:
“We’ll take your money and do what we please”.
He further added:
“We cannot expect people in the UK to pay taxes to improve education and health in
Pakistan if the Pakistani elite do not pay meaningful amounts of income tax” (The News,
2013c, p.1).
Another member of the committee, Fiona O’Donnell commented on distrust of Pakistani citizens
towards government of Pakistan by quoting common people saying:
“I do not pay my taxes because the government is so corrupt that it does not do any
good, so I would rather engage in private philanthropy” (The News, 2013d, p.1).
This reaction of the international donor community not only highlights their response on the poor
reform efforts of tax administration in Pakistan, but also highlights how significant it is to examine
the reason behind failure of tax reform efforts in Pakistan in this chapter. This introductory section
of the chapter, in addition to providing the history and context of tax reform efforts in Pakistan,
was also able to lend a positive answer to the two-part question raised at the start of this chapter
asking ‘If’ and ‘Why’ tax administration reform failed to effectively control opportunities for
corruption in FBR Pakistan. Yes the tax administration reform has failed to effectively control
opportunities for corruption in FBR in Pakistan. Next, the following sections aim to further explore
the reasons behind ‘Why’ tax administration reform failed to effectively control opportunities for
corruption in FBR Pakistan and what should be done about it. In addition, the potential and
suitability of adopting Anti-Corruption SARA framework for Pakistan to control corruption in tax
administration will be analyzed.

193
Before moving onto next sections, tables 6.1, 6.2, and 6.3 as well as figures 6.1, 6.2, and 6.3
illustrates an overview of revenue performance (in terms of Tax-to-GDP ratio) along with
progress against corruption (in terms of CPI rating) for Pakistan over years and in comparison to
its regional countries, in addition to making comparisons with SARA countries of the macro level
of analysis.
Table 6.1: Overview of Pakistan’s Progress towards Revenue Improvement and Corruption
Control from 2001-2012
Corruption
Tax Revenue
Year Perception Index
(% of GDP)*
(CPI) score**
2001 10.5 % 23
2002 10.7 % 26
2003 11.4 % 25
2004 10.8 % 21
2005 10.1 % 21
2006 9.8 % 22
2007 9.6 % 24
2008 9.9 % 25
2009 9.1 % 24
2010 10.1 % 23
2011 9.3 % 25
2012 10.2 % 27
*Source: ADB (2013)
**Source: Transparency International. Available online http://cpi.transparency.org
‘The Corruption Perceptions Index ranks countries based on how corrupt their public sector is perceived to be. A
country’s score indicates the perceived level of public sector corruption on a scale of 0 - 100, where 0 means that a
country is perceived as highly corrupt and 100 means it is perceived as very clean’.
194
Figure 6.1: Graphical Representation of Pakistan’s Progress towards Revenue

Improvement and Corruption Control from 2001-2012
40
35
30
25
20 CPI Rating
Tax Revenue
15
10
0
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
As illustrated in table 6.1 along with figure 6.1, Pakistan’s Tax-to-GDP ratio, which is quite low by
international standards, has stagnated over the last twelve years and not shown any
considerable improvement over this period, despite efforts to reform. In addition, its CPI rating,
which again is very low, consistently placed Pakistan as one of most corrupt countries of the
world over the period analyzed.
Control in Comparison to Regional Counterparts
Corruption
Tax Revenue (% of
Country Perception Index
GDP) 2012*
(CPI) 2012**
Malaysia 16.2 % 49
Thailand 15.3 % 37
Nepal 13.8 % 27
Philippines 12.9 % 34
Indonesia 11.3 % 32
Sri Lanka 11.1 % 40
Bangladesh 10.5 % 26
Pakistan 10.2 % 27
India 7.4 % 36
*Source: ADB (2013)
**Source: Transparency International. Available online http://cpi.transparency.org
195

Improvement and Corruption Control in Comparison to Regional Counterparts
Moreover, as illustrated in table 6.2 along with figure 6.2, Pakistan’s current tax revenue in
comparison to other regional developing countries is one of the lowest in the group, with only
India having the lower tax-to-GDP ratio than Pakistan. This highlights poor progress made by
Pakistan to improve its revenue performance over time, in comparison to other developing
countries with similar contexts. In addition, its CPI rating is also one of the lowest in the group,
with only Bangladesh having lower CPI rating than Pakistan, placing Pakistan as one of most
corrupt countries in the group.

196
Control in Comparison to SARA Countries in Macro level of Analysis
Corruption
Tax Revenue (% of
SARA Country Perception Index
GDP) 2012*
(CPI) 2012**
Mexico 29.7 % 34
South Africa 26.9 % 43
Venezuela 25.0 % 19
Ghana 20.8 % 45
Kenya 18.4 % 27
Peru 18.0 % 38
Uganda 12.6 % 29
Tanzania 12.0 % 35
Guatemala 11.9 % 33
Pakistan 10.2 % 27
*Source: Heritage Foundation. Available online http://www.heritage.org/index/ranking
**Transparency International. Available online http://cpi.transparency.org

Improvement and Corruption Control in Comparison to SARA Countries in Macro level of
Analysis
Lastly, table 6.3 along with figure 6.3 illustrates Pakistan’ progress towards revenue improvement
and progress against corruption in comparison to SARA developing countries included in the
197
macro level of analysis in chapter five. As evident, Pakistan has the lowest tax-to-GDP ratio in
comparison to the group of better as well as poorly performing SARA countries. In addition,
Pakistan’s CPI rating is also one of the lowest in the group, with only Venezuela having lower CPI
rating than Pakistan. This highlights poor progress made by Pakistan to improve its revenue
performance and progress against corruption in comparison to developing countries having
undertaken SARA reform to varied levels of success.
The tables and figures above helped to demonstrate that Pakistan’s progress towards
improvement in revenue collection as well as controlling corruption has not improved over time,
or in comparison to its regional developing countries, or in contrast to selected SARA developing
countries. This makes it even more pertinent to undertake the proposed micro level of analysis in
this chapter to examine why tax administration reform failed to effectively control opportunities for
corruption in FBR Pakistan and what should be done about it.
6.2 PERSONNEL AUTONOMY CONTROLLING MOTIVATIONS FOR CORRUPTION IN FBR
In order to analyze the potential and suitability of adopting Anti-Corruption SARA framework for
FBR Pakistan to control corruption in tax administration, the micro level of analysis in this section
is aimed to analyze research hypothesis 1 along with its four sub-hypotheses. It is further
highlighted here that research hypothesis 1 along with its sub-hypotheses have been adapted to
examine tax administration reform in FBR Pakistan. This adaptation is undertaken by replacing
the term ‘SARA’ in these hypotheses with ‘FBR’. Hence, in this chapter it is proposed that by
preferring personnel autonomy over effective accountability, FBR have been effective against
motivations but not opportunities for corruption. Research hypothesis 1 comprises of four sub-
hypotheses (1a, 1b, 1c, and 1d), with each one referring to one specific preventive strategy to
control motivations for corruption. SARA design component of personnel autonomy is

198
conceptualized to control motivations for corruption by 1) increases in wages and rewards; 2)
increases in the probability of detection; 3) increases in and stricter enforcement of penalties for
corruption; and 4) instilling ethics in tax officials. First all four sub-hypotheses will be analyzed
and then these analyses will be integrated for the cumulative analysis of research hypothesis 1 in
sections below.
The next sections examine how FBR used personnel autonomy to control motivations for
corruption by utilizing both primary (semi-structured interviews) as well as secondary literature.
The overall proposition is that if FBR has made progress towards ensuring personnel autonomy,
then it can also be taken to propose that progress towards controlling motivations for corruption
has been made.
6.2.1 Personnel Autonomy Controlling Motivations for Corruption in FBR by Increasing
Wages and Rewards
Research hypothesis 1(a) has been adapted for FBR and proposes that by increases in wages
and rewards, FBR have been effective against motivations but not opportunities for corruption.
Personnel autonomy is operationalized to control motivations for corruption in FBR by affecting
implementation of preventive strategy of increases in wages for tax officials and utilization of
performance-linked bonuses. This section aims to analyze evidence on tax administration
reforms in FBR Pakistan with particular focus on if and how personnel autonomy is being utilized
by FBR to control motivations for corruption by increasing wages and rewards for tax officials.
One of underlying reasons behind entrenched corruption in FBR pre-reform was blamed on
below-subsistence level wages for tax officials. For this reason, one of the major focuses of
TARP reform project in Pakistan was to improve the Human Resource Management (HRM) at
199
FBR (FBR, 2008). As a result, major focus of HRM strategy was laid on improved induction,
promotion and remuneration packages for tax officials in FBR. Starting from the FBR
headquarters, reform in wages for 150 tax officials was introduced in the form of 100% special
allowance in December, 2007. These tax officials were hired via a competitive procedure of
internal job postings (World Bank, 2004b). As per this increase, an allowance equivalent to 100%
of basic pay of tax officials was given, hence doubling the total salary per month. As a result of
this reform, these FBR employees saw their monthly wages doubled in comparison to the rest of
the civil service. This tactic of increased wages, initially only rolled out for the staff at FBR
headquarters, proved very popular among the staff, and was subsequently extended to all other
staff of FBR in other wings and field offices (FBR, 2008). In terms of bonuses and rewards, in
2012 a special rewards scheme for tax officials working in the customs department was
introduced (Appendix 2). As per this scheme, the customs officials are remunerated with cash
rewards equivalent to about 20% of the tax evasion detected. This reward scheme has also
proved popular among tax officials, although some tax officials (interviewees) were of the view
that this scheme was about to be rolled back on ground of misuse by some officials. No official
confirmation about the withdrawal of this reward scheme was available at the FBR headquarter.
One of the major objectives behind increased wages and rewards for tax officials was aimed to
curb their motivations for corruption, arising due to non- provision of living wages. But was this
strategy effective? Not as per majority of tax officials interviewed. When asked (Appendix 1) if the
increase in wages and rewards have been sufficient enough/effective to deter tax officials from
committing corruption in FBR? The majority answers were emphatic “No!”. As Tax Official 1
elaborated:
‘You have no idea what is the amount of money they are making from corruption, no
amount of increase in salary……I repeat no amount of increase in salary can be enough
to deter them not to do it’.
200
Another Tax Official 2 added on the similar line:
‘How can you expect to raise the salary by Rs. 30,000 per month, and expect them to
forgo bribes more than ten times of this amount? It’s just not workable. Unless you catch
the most corrupt and punish them, maybe then officials will start to see the
consequences and be happy with their increased wages, otherwise no chance’.
Another Tax Official 3 had a much stronger opinion:
‘I take this increase as a slap on my face for being honest and doing my job properly’.
Upon inquiring what reform effort would make the situation better for the honest tax officials, he
replied:
‘First stop them making millions every month in front of my eyes, they know they are
doing it, I know they are doing it, they know I know they are doing it, everybody higher up
knows they are doing it, but nobody can stop it. How can I be content with my double
salary with the knowledge that the person sitting in the next office, equal in rank and
everything, is making millions, while I am struggling to make ends meet with few more
thousands’.
When asked if the wages and reward structure at FBR enough to retain employees with the right
skills? One of the Tax Official 4 had an interesting take. He said:
‘If you are honest and intends to stay honest, then your best bet is to use FBR as a
training ground, get foreign qualifications in the name of capacity building, come back,
serve for few years, and venture into private sector as a tax consultant. I know many
people who have done this, and are earning much more than what they were getting at
FBR. If you are corrupt, this advice does not apply to you. No money in the private sector
could match with what you could make here if you are not honest’.
In addition to interviewing tax officials for micro level of analysis, this case study also interviewed
multiple other stakeholders of FBR reform who expressed similar concerns when quizzed about
role of increased wages and rewards towards controlling corruption at FBR. Notably, MoF Official
1 said:
201
‘See we did everything we could in the book to deter them from being corrupt, we
doubled salaries to give them more money because money is what pushes them to do
corruption, isn’t it? but I guess they wanted lots more of it. Of course we couldn’t
increase salaries beyond a certain acceptable limit. We doubled salaries but even it was
not enough. Guess they were making lots more of money from corruption!’
Similar sentiments were expressed by the other two interviewees at MoF, who were also
disappointed that even doubling the salaries did work very well. MoF Official 2 said:
‘I think it did deter some of the officers who were not making lots of corruption money
and maybe had guilt in their hearts for doing it. See even our religion speaks so openly
against corruption. But yes the big fish had no impact what-so-ever’.
One of the businessman in the second stage of interviews had a personal experience regarding
the rewards for tax officials for uncovering tax evasion. Businessman 1 said:
‘They introduced these rewards for catching evasion, you know what it did to us. Tax
official in my area openly threatened me to cooperate with him otherwise he is bound to
make money by rewards by falsely declaring my evasion and will make it impossible for
me to continue with my business. I had to hire a legal tax consultant to get out of this
situation and had to pay him high fees. My friends thought it would have been easier if I
had simply bribed the officer in the first place. That’s how business is done in Pakistan.
But you already must know this’.
In the similar line, Taxpayer 1 also contributed by saying:
Now when I look back, I think it’s just a joke in the name of reform. They doubled
salaries, gave rewards but simultaneously also let them take bribes. Wasn’t it a win-win?
They should have tightened ways to catch the corrupt ones and punish them too. Have
you ever heard that any tax officer been convicted of corruption. Even if caught, they get
free and have all the illegal money to go abroad and enjoy life. Not a bad tradeoff!’
Such sentiments about ineffectiveness of increased wages and rewards towards curbing
motivations for corruption were consistent with most interview respondents. The main problem
seems to be that in the presence of widespread opportunities for corruption and ineffective
oversight mechanisms, the increase in wages and rewards sort of backfired for FBR, even
resulting in increased cynicism in tax officials.

202
The above micro level of analysis was aimed at analyzing research hypothesis 1(a) for FBR. This
analysis illustrated that although FBR did utilized personnel autonomy to increase wages and
rewards for tax officials, but was not able to effectively control motivations for corruption. Hence,
FBR did not lend support to the basic argument behind research hypothesis 1(a) as per micro
level of analysis, as the progress against controlling motivations for corruption improved least in
FBR, although personnel autonomy in terms of increases in wages and rewards was high. In
FBR, personnel autonomy proved ineffective in controlling motivations for corruption by
increasing wages and rewards. Hence, research hypothesis 1(a) was not supported for FBR.
6.2.2 Personnel Autonomy Controlling Motivations for Corruption in FBR by Increasing the
Research hypothesis 1(b) has been adapted for FBR and proposes that by increases in the
probability of detection, FBR have been effective against motivations but not opportunities for
corruption. Personnel autonomy is operationalized to control motivations for corruption in FBR by
affecting implementation of preventive strategy of increases in the probability of detection with the
indicators being increases in the quality and frequency of internal audits and probes by
establishing internal audit and anti-corruption divisions; increases in the number of supervisory
personnel assigned to internal audit and anti-corruption divisions; and statutory condition of
declaration of assets by all employees. This section aims to analyze evidence on tax
administration reforms in FBR Pakistan with particular focus on if and how personnel autonomy is
being utilized by FBR to control motivations for corruption by increases in the probability of
detection.
203
For increasing the quality and frequency of internal audits and probes by establishing internal
audit and anti-corruption divisions, the FBR constituted a ‘Directorate General Intelligence &
Investigation’ as part of tax administration reform. This Directorate was not a newly established
entity, and existed in other forms before reform. The origin of this Directorate dates back to 1957,
when it was only entrusted with the task of customs intelligence and investigation. By 1974, the
Directorate was also entrusted with intelligence and investigation of central excise. In 1995, the
Directorate was further entrusted with audit of sales tax fraud as well, hence becoming
Directorate General (Intelligence and Investigation) Customs, Sales Tax and Central Excise. As a
result of TARP implementation in 2005, the Directorate General (Intelligence and Investigation)
Customs, Sales Tax & Federal Excise was re-labelled as Directorate General Intelligence &
Investigation of FBR with the job of vigilance and integrity management of both direct and indirect
taxes. By 2010, this directorate was also nominated as one of central agencies for investigations
into money laundering under the AML Act 2010. One of the basic objectives behind
establishment of this Directorate was to investigate accusations against FBR officials and
detecting corrupt practices by corrupt officials (World Bank, 2004b). As mentioned in its charter of
functions by the FBR, one of the main functions of this Directorate includes:
“To look into and investigate cases of corruption and mal-practices of the revenue collecting
agencies working under the FBR received by the Directorate General through public, press or
any other source and to propose appropriate corrective or punitive action” (FBR, 2013).
As for increasing the number of supervisory personnel assigned to internal audit and anti-
corruption divisions, the FBR has taken concrete steps in this regard by establishing seven
regional directorates of the Directorate General Intelligence & Investigation based at FBR
headquarters across Pakistan. As depicted in figure 6.4 below, regional directorates with
significant number of personnel have been established in all major cities and areas of Pakistan.
204
Figure 6.4: Overview of Directorate General Intelligence & Investigation at FBR

Headquarters in Islamabad, along with regional Directorates
Source: FBR (2013)
Has the establishment of these Directorates improved the probability of corruption detection at
FBR, or is it just a name change? When asked if the increase in the quality and frequency of
internal audits and probes by establishing Directorate General Intelligence & Investigation along
with allied regional Directorates at FBR has deterred tax official from committing corruption?
Tax Official 2 opined:
‘You do know this is not an entirely new function? It has been conducted in some form
right from the start, even before CBR and FBR. Yes they have got bigger and better
offices now, but the mindset is same. Before, people under investigation used to bribe
the old office. The only change is it’s a new office now and the bribe is higher as well.
That’s all!’
Another Tax Official 1 added:
‘The officers who are corrupt have got connections in the directorates. They are all
colleagues after all. While they are at it, they are certain in their mind that even if they get
caught or somebody complains, they have got somebody in the directorate who will save
them. The taxpayers have no choice but to pay the bribe when asked. They know they
can complain, but it will be of no use, just a waste of time, rather they might be implicated
205
for a bigger tax evasion. This has happened in the past and in the end the taxpayer had
to bribe a much bigger amount than initially demanded’.
When asked if the increase in the number of supervisory personnel assigned to Directorate
General Intelligence & Investigation along with allied regional Directorates at FBR has deterred
tax official from committing corruption? Tax Official 4 reflected on the lack of autonomy exercised
by Directorate staff:
‘Anybody can be posted and transferred from the directorates at any time to other
functions of FBR. They know if they are very tough to a certain colleague, that person
might create problems for them in the future, and if they are really well-connected at the
right places, they will simply get them transferred. The officials who are working for long
time in these intelligence directorates are the one who have learned the tricks of trade.
They only do investigations of cases their superiors want them to, and use their
discretion to go soft on colleagues whenever they can, to get favors from them in the
future’.
Another Tax Official 5 elaborated on the discretionary powers exercised by the supervisory
personnel in these Directorates:
‘They are supposed to check everyone for tax evasion and corruption, but who is
supposed to check them? They operate as they have complete immunity from any action
what so ever. I am sure the ones at the top are also involved. Otherwise how could they?
Surely people can complain against them for asking for bribes etc. But so far in my
career I have never heard of any dismissal of intelligence directorate staff for committing
corruption by not detecting corruption’.
As for progress towards establishment of statutory condition of declaration of assets by all
employees, FBR has taken important steps. Before reform, under the Government Servants
Conduct Rules, 1964, all civil servants, including FBR employees, were already required to
disclose information about their assets and liabilities on a yearly basis. The screening of civil
servants’ assets and liabilities was aimed towards identifying any unjustified wealth as an
indicator of corrupt behaviour. After reform, FBR publicizes that the asset declaration system has
been revamped, with an automated system now scrutinizing the declarations. For improving the
capability of monitoring and ensuring accuracy of asset declarations, the system has been
206
overhauled by introducing computerized monitoring mechanism. For this reason, all data had
been transferred from the old to a new system, along with adoption of new software since 2008
(ADB, 2008). In addition to system overhaul, Pakistan has also made efforts towards improving
the quality and quantity of information sought via the Asset Declaration form. As can be seen in
Appendix 3, the form is very detailed seeking information not only about income (including
income of spouse and children), but also details of utility bills (including mobile bills of all family
members), details of international travels of all family members, details of educational institutions
attended by children (local as well as foreign), membership of private clubs, details and sources
of gifts received, details of moveable as well immoveable assets including details of any assets
disposed-off during the year, details of all investments, and details of bank accounts (Appendix
3). It is also worthwhile to mention here that FBR also tied the increase in wages for tax officials
with mandatory declaration of assets. At the start of reform, only those tax officials saw their
wages increased who had submitted all pending asset declaration forms for previous years of
service. Anecdotal data suggested that after the initial push, the situation deteriorated and if any
tax official did not complied with their responsibility of submitting completed asset declaration
form for any particular year, then they were only furnished with reminders to do so, but continued
to enjoy increased wages.
When asked if they have noticed any effectiveness of the reformed system of statutory condition
of declaration of assets by FBR towards deterring tax officials from committing corruption? Most
of the respondents responded negatively. One Tax Official 6 reflected on the ineffectiveness of
these forms by adding:
‘I submit my asset declaration form every year, most of us do. But we know nobody
would bother to have a look at them. They are collected and filed in our records. That’s
all. Some time ago they used to catch dust. Not anymore, as they are computerized now.
There only purpose seems to be that if any FBR tax official is caught up in a big inquiry in
any court etc. then, the Directorate can very efficiently furnish the copies of these forms,
207
and on the instruction of court start looking into these forms. Only after the crime! Why
not before? By having these up-to-date forms completed and filed, the Directorate people
think their job is done. I can write anything in the form with certainty that if I never get into
any trouble; nobody will know what I declared. The directorate people only inquire these
declarations when they are asked to do so either by higher-ups at FBR, or FTO, or
courts’.
There has been a high profile corruption case of pilferage of containers entering Pakistan under
the transit trade pact with Afghanistan. As per FTO’s report into the issue, the national exchequer
suffered a loss in tune of Rs. 37 billion over four year’s period (DAWN, 2011a). Reflecting on this
issue of massive corruption in relation to declaration of assets, Tax Official 6 elaborated:
‘This case of massive tax corruption by customs officials is in the court. Do you know
what has been the response of FBR on this matter so far? They have suspended a
number of tax officials alleged to be involved in this scam. They have also promised the
Supreme Court that while these officials are suspended, FBR will conduct a detailed
inquiry into their asset declarations forms and will duly submit a report to the court. What
do they expect to find in the declarations? I hope they are not hoping to find Rs. 37
Billion worth of assets in the declarations, because they’ll be disappointed. If I am
corrupt, the one thing I will make sure is to keep my asset declarations current and clean.
They always are! I don’t think they have ever found a corrupt person so naïve to stash
everything in his name and then also declare it, because that’s the only way they can
catch him’.
In addition to interviewing tax officials for micro level of analysis, this case study also interviewed
multiple other stakeholders of FBR reform who expressed similar concerns when quizzed about
role of increases in the probability of detection towards controlling corruption at FBR.
Notably, Taxpayer 2 said:
‘What is the point of these asset declaration forms? We all know that corrupt ones are
the smarter ones. They don’t even keep their money and assets through corruption in the
names of their family members now. They have special friends for these purposes who
get favors in return. They keep bribe money in bank accounts of friends and there is no
way one can relate them with each other as they are not family. It’s a very common
practice these days’.
208
Also Rehman Azhar from Dunya News had interesting take on this issue:
‘Not very long ago, the corrupt ones would hide assets either in the names of family
members or very close friends. Some even would get caught for doing so. But now times
have changed. Look at what happened in Panama leaks about Pakistan. More than 400
Pakistanis had off-shore companies with no trail of how money got out of country. Many
of these people are government officials. It tells us they have got innovative in hiding
their corrupt money. We won’t find any off-shores in Panama decelerated in asset
declaration forms’.
MoF Official 3 contributed by highlighting the formal position and saying:
‘Well all of us have to do it. It is the requirement by the establishment division of Pakistan
that all government officials declare assets annually. As for their effectiveness against
deterring corruption, it is at times useful and directs towards money possibly made from
corrupt sources’.
Imdad Bhatti from Jang News had an interesting take about asset declaration forms:
I think it is only the very silly ones who sometimes are caught for declaring un-
proportionate level of income and then get caught. It’s very rare. The first condition to do
corruption is that you have to be very smart in your head, after all you have to beat the
system in doing so. It’s not an easy task. They first understand system fully and then
come up with ways to distort them. The corrupt ones are definitely not usually stupid, if
are, then get caught quickly’.
These views about ineffectiveness of increases in the probability of detection towards curbing
seems to be that in the presence of patronage networks and absence of effective checks on
discretionary powers exercised by the intelligence Directorate staff, majority of interview
respondents remained pessimist about any value addition by dedicated intelligence Directorates.
The tool of declaration of assets by tax officials also seems to be utilized only after corruption had
been done, and not as an effective preventive tool against corruption.
The above micro level of analysis was aimed at analyzing research hypothesis 1(b) for FBR. This
analysis illustrated that although FBR did utilized personnel autonomy to increase the probability
of detection, but was not able to effectively control motivations for corruption. Hence, FBR did not
lend support to the basic argument behind research hypothesis 1(b) as per micro level of
209
analysis, as the progress against controlling motivations for corruption improved least in FBR,
although personnel autonomy in terms of increases in the probability of detection was high. In
FBR, personnel autonomy proved ineffective in controlling motivations for corruption by
increasing the probability of detection. Hence, research hypothesis 1(b) was not supported by
FBR.
6.2.3 Personnel Autonomy Controlling Motivations for Corruption in FBR by Increases in
and Stricter Enforcement of Penalties for Corruption
Research hypothesis 1(c) has been adapted for FBR and proposes that by increases in and
stricter enforcement of penalties for corruption, FBR have been effective against motivations but
not opportunities for corruption. Personnel autonomy is operationalized to control motivations for
corruption in FBR by affecting implementation of preventive strategy of increases in and stricter
enforcement of penalties for corruption with the indicators being increases in imposition of high
levels of penalties including monetary sanctions, job dismissals and prison sentences. This
section aims to analyze evidence on tax administration reforms in FBR Pakistan with particular
focus on if and how personnel autonomy is being utilized by FBR to control motivations for
corruption by increases in and stricter enforcement of penalties for corruption.
For increasingly imposing high levels of penalties including monetary sanctions, job dismissals
and prison sentences, the FBR has entrusted this responsibility to Member Administration of the
board. Part of his/her job description is to ‘deal with disciplinary/litigation cases of FBR
employees’ (FBR, 2013). In practice, in terms of increasingly imposing high levels of penalties,
the situation on the ground seems dismal. FBR seems less than motivated for utilizing high level
of penalties as a deterrence tool against corruption. There have been several instances when
high level officials in FBR have pushed rules to save their colleagues. In one such instance, a
210
senior level FBR tax collector was acquitted for corruption committed in 2005 in tune of Rs. 140
million refund case. FBR simply declared the alleged tax collector innocent by moving a summary
from the Member Administration to Finance Secretary in MoF, who moved it to the Prime
Minister. The Prime Minister agreed with the summary and the alleged tax collector was simply
reinstated as Collector Customs in 2010. In this case, despite the hue and cry by the Public
Accounts Committee (PAC), the maximum penalty faced by the tax collector was a suspension
from service for a limited time (Ghumman, 2010). This incident, along with many others, did not
do much to improve the morale of honest tax officials working down the line; and highlighted the
existence of powerful network of high level tax officials going to any lengths to save each other.
In addition to the ineffectiveness and non-commitment of FBR towards serious utilization of
penalties for corruption, there have been several instances of battles between the courts, FTO
office and FBR over jurisdiction for sanctioning penalties. In one such highly publicized case fight
broke out between the FTO and FBR over conviction of an additional commissioner income tax
on charges of arbitrary assessment of tax and bribe received from two taxpayers. The alleged
commissioner was sentenced to three months’ imprisonment and fine of Rs. 50,000 for contempt
of court by the FTO on account of not presenting himself before the FTO proceedings, in addition
to transferring him from his post. In response to this ‘first of its kind sentencing’ several tax
officials in FBR went on strike against the decision, maintaining that the FTO had no authority to
pass such judgment. In response to the strike, chairman FBR assured all tax officials that the
alleged official will be provided full legal support, in the form of a very expensive legal counsel, by
FBR. In addition, FBR made sure that the alleged official would not spend even a night in jail, by
duly transferring him to a hospital, immediately after sentencing, on ground of health issues. After
these assurances the strike was called off at FBR (Cheema, 2012b). Subsequently, the sentence
awarded to the additional commissioner by the FTO was suspended by the Supreme Court on
211
ground of no jurisdiction of FTO to award such sentence (DAWN, 2013). This incident highlighted
the clan mentality of tax officials for collectively going against penalties etc. along with going to
extreme lengths to save each other. The classical case of ‘you scratch my back now, and I’ll
scratch yours later’.
Nonetheless, there have been instances, albeit very few, of penalties for FBR officials including
monetary sanctions, job dismissals and prison sentences. In majority of these cases, the
judgments were passed by the accountability courts against FBR tax officials, and only due to
overwhelmingly strong evidences against the accused, FBR could not save them (as per majority
of interview respondents). It is a very rare practice that FBR initiates proceeding for administering
penalties against its own employees internally from within the organization. In one case, the
commissioner income tax of Sargodha city was convicted of corruption in tune of Rs.180 million.
He was sentenced to one year imprisonment along with fine of Rs.10 million, in addition to
seizure of all assets and properties by the accountability court (Jang, 2011). In another case, the
accountability court convicted a customs superintendent, posted at Islamabad international
airport, for corruption. He was sentenced to two year imprisonment along with fine of Rs.17.8
million. He was charged for owning assets worth Rs.17.8 million which were inconsistent with his
declared sources of income. He was indicted of making these assets through corruption (DAWN,
2010). In still another case, two customs officials along with police officials were booked for
corruption in tune of Rs. 7.239 million. These regional level tax officials were charged with
receiving bribes of Rs. 2 million from smugglers for tampering with the quantity of goods seized
during a raid. They were sentenced to seven years imprisonment, along with fine of Rs. 2 million
each (DAWN, 2011b).

212
The FBR has tried its hand at job dismissals for purging the tax administration from patronage
networks of political appointees. In 2000, under the Removal from Service Ordinance (2000) FBR
tried to purge a significant number of suspected tax officials by suspending them from the
service. But this exercise failed to get rid of any significant number of suspended officials with
majority joining back their jobs without facing any penalties, on account of political connections
(Shah, 2011). Similarly, Hors (2001) elaborated corruption purges of customs officials in Pakistan
as strategies that failed to took off mainly due to lack of political will and inconsistency in
recruitment policies. The FTO has time and again stressed the need to punish offenders to create
deterrence in the system against motivations for corruption, and to build ability to prosecute. On
one occasion he said:
“High punishments are available in the law but no one is punished, send 20-25 top tax
evaders every year behind the bars” (The News, 2013a, p.1).
Chairman FBR on the other, while trying to make a case of a controversial amnesty scheme,
confessed that the FBR’s strategy to catch tax evaders has been hindered because of corruption,
where no action against the evaders has been taken by tax officials due to bribes received from
tax evaders. He admitted dismissing a number of tax officials on account of corruption, but
added:
“I can’t fire everyone in the tax collection system” (The News, 2012d, p. 4).
This statement quite clearly portrays not only the extent of corruption at FBR, but also the
incapability of penalties for getting rid of corruption from FBR, even by the chairman himself.
One of the major objectives behind increased imposition of high level of penalties for tax officials
if they commit corruption was aimed to curb their motivations for corruption, by deterring them
through a possibility of severe penalties. Did the interview respondents notice any change and
213
effectiveness in imposition of high level of penalties for corruption after reform? Was FBR active
in pursuing high level of penalties for corruption? When asked if the increases in imposition of
high levels of penalties including monetary sanctions, job dismissals and prison sentences in
FBR has deterred tax official from committing corruption? Tax Official 7 responded:
‘It is very rare here that someone not only gets caught, but also receives any penalties
for corruption. They always do their homework. They always weight the benefits against
the costs. They know all the tricks of the trade on how to avoid any penalties. In most
cases, they just get suspended from their jobs for a while, till they find the way to get
back. You can’t consider suspension from service any penalty’.
Another Tax Official 8 had a different take:
‘I know someone who uses to boast openly that he always had a target of making at
least Rs. 10 million from every posting in different districts. When quizzed what happens
if he’ll get caught, he responded. So what? The maximum penalty might be removal from
service, jail and fine. Although all these three were highly improbable, but it’s worth it
considering how much he was making at that moment. For him the benefits outweighed
the cost and the maximum penalty was not a deterrence’.
Tax Official 8 added:
‘If someone does get caught, that usually means they were on the solo flight and were
not keeping their superiors happy. They can avoid the penalty by making it up to the
superiors and come back without any charge. Everyone has some sort of inquiry going
on, but how many get penalties? Very few!’
Tax Official 9 added on the same line:
‘If you have been caught and penalized, that simply means you did not had any good
connections at the top, or you irritated someone there. It is always possible to avoid
penalties if you know how to? Almost everyone does’.
In relation to the interactions between the FBR, the FTO, and accountability courts for
administering penalties, Tax Official 9 elaborated:

214
‘The speed of justice is very slow. If someone does gets caught and they are suspended
from service till inquiry in FBR, then all they need to do is to find someone willing to
collude in judiciary. As a result there suspension is suspended by the courts over some
legal loophole, and the result is a time intensive litigation proceedings between FBR and
the court. There have been instances where cases remained in courts for decades. They
get the judicial proceedings delayed either to buy time or get someone favorable in the
court. Justice delayed is justice denied’.
Discussing the non-severity of monetary sanctions, job dismissals and prison sentences, Tax
Official 10 explained:
‘The maximum monetary sanction any tax official can get is the total amount of
corruption proved. As a result they are fined the total amount corruption, and their total
assets are seized for realizing the sanction. If there assets are less than the total
sanctioned amount, they are sentenced to maybe a year more to prison. How can one
expect that the corrupt officer would have kept everything in their own name? They just
don’t do it. Not even family. It’s either stashed in friends’ names or the money is simply
laundered to foreign bank accounts. The approach is to survive the sentence for few
years and then go abroad with the family and enjoy life. They had probably already made
enough money to feed seven generations’.
Tax Official 10 further added:
‘If you examine most high profile cases of corruption involving billion of rupees in
Pakistan, you will find that the accused simply entered a plea bargain with the
government, returned the proved amount of corrupted money and that’s it. They were
free to go. They were the most corrupt. Corruption was proved. They ran away, they
were caught. Money was taken off and they were let go. How is this justice? What if they
had made billions more, which I am sure in most cases they have, hidden somewhere
else. Surely a corrupt act involving billion of rupees was not there first time. What
message does this send to the honest ones?’
MoF Official 2 had a slightly different take favoring FBR:
‘You see get rid of corrupt ones is not only the job of the government. The Police and
judicial system also gets involved. We cannot also be blamed for the inefficiencies in
other systems. We are trying our best t minimize corruption’.
Taxpayer 2 opined:
‘Corruption is everywhere but I hardly hear any news in media about them getting caught
and punished. Corruption is everywhere because penalties are not strong enough’.
Businessman 4 also responded:

215
‘I think this is also related to the problems with police and judiciary. Corruption is
everywhere. We cannot eliminate corruption in governments departments before killing it
for police and judiciary. Otherwise they find corrupt ones there and escape punishments’.
As highlighted above, the ineffectiveness of imposition of high level of penalties towards curbing
seems to be that in the presence of patronage networks in FBR and beyond, majority of interview
respondents remained pessimist about effectiveness and fair applicability of penalties on corrupt
tax officials to deter them from indulging in malpractices. The inapplicability, delay, and non-
severity of penalties for corruption was seemingly failing to deter the corrupt ones, but also
sending a demoralizing signal to the honest tax officials. The nexus between the accountability
courts, FBR, and FTO also seemed marred with conflicts over encroachment of each other’s
jurisdictions at certain times, not helping with effective administration of penalties for corruption
against tax officials.
The above micro level of analysis was aimed at analyzing research hypothesis 1(c) for FBR. This
analysis illustrated that although FBR did utilized personnel autonomy to result in increases in
and stricter enforcement of penalties for corruption for tax officials, but was not able to effectively
control motivations for corruption. Hence, FBR did not lend support to the basic argument behind
research hypothesis 1(c) as per micro level of analysis, as the progress against controlling
motivations for corruption improved least in FBR, although personnel autonomy in terms of
increases in and stricter enforcement of penalties was high. In FBR, personnel autonomy proved
ineffective in controlling motivations for corruption by increases in and stricter enforcement of
penalties. Hence, research hypothesis 1(c) was not supported by FBR.

216
6.2.4 Personnel Autonomy Controlling Motivations for Corruption in FBR by Instilling
Ethics in Tax Officials
Research hypothesis 1(d) has been adapted for FBR and proposes that by instilling ethics in tax
officials, FBR have been effective against motivations but not opportunities for corruption.
Personnel autonomy is operationalized to control motivations for corruption in FBR by affecting
implementation of preventive strategy of instilling ethics in tax officials with the indicators being
increases in ethics training and adoption of code of conduct. This section aims to analyze
evidence on tax administration reforms in FBR Pakistan with particular focus on if and how
personnel autonomy is being utilized by FBR to control motivations for corruption by instilling
ethics in tax officials.
FBR adopted a code of conduct and published it in 2004. It was prepared by the member HRM
with help from member Administration and consultants, notwithstanding limited input either by tax
officials or external stakeholders of FBR. This code of conduct included the vision, mission, and
values of FBR.
The vision statement states:
“To be a modern, progressive, effective, autonomous, and credible organization for

optimizing revenue by providing quality service and promoting compliance with tax and
related laws”.
The mission statement states:
“To enhance the capability of the tax system to collect due taxes through application of
modern techniques, providing taxpayer assistance and creating a motivated, satisfied
and dedicated and professional workforce”.
Values include:
217
“Integrity, Professionalism, Teamwork, Courtesy, Fairness, Transparency, and

Responsiveness”. (FBR, 2013).
This code was specifically established to educate and emphasize the code of conduct
requirements for tax officials. Before, the conduct of tax officials was already regulated by the
Government Servants Conduct Rules, 1964. It was perceived that the tax officials have neither
read nor adopted the rules and regulations under the Conduct Rules, 1964. To deal with this,
FBR developed its own code of conduct in an easy and simple format, and in the form of a small
pocket-sized booklet for easy referencing. It was emphasized in the booklet that the code of
conduct by FBR has not replaced the Government Servants Conduct Rules, 1964, rather it has
only adapted these rules to FBR context (Mayville, 2005). All FBR tax officials are obligated to
read the contents of the code of conduct, and affix a signature on a statement page, declaring
that they have not only understood the contents, but will also abide by them. This signed page
from the booklet is duly transferred to the personal files of all tax officials.
The code of conduct details standard responses to situations that might lead to unethical
conduct. Examples include not receiving gifts/favors from taxpayers, not borrowing money from
taxpayers, to avoid conflict of interest between official duties and personal interests. Expected
behavioral norms pertaining to dealing with taxpayers’ complaints, revelation of official identity,
conflict of interest, and off-duty conduct is also outlined in the code. This code also offers
guidelines pertaining to conduct in the work environment including hygiene, proper clothing,
behavior and attitude, and timekeeping. Personal conduct is also discussed pertaining to
receiving gifts, lending and borrowing, employment in the private sector, usage of government
housings, political actions, voicing political opinions etc. The code of conduct also includes the
definitions of misconduct and types of penalties applicable as per general civil service rules. The
tax officials are encouraged to involve senior authorities when facing an ethical dilemma. In
218
assessment, on paper the code of conduct established by the FBR incorporates all essential
features of modern codes, thus illustrating best practice. For increasingly conducting ethics
training towards adoption of code of conduct to curb motivations for corruption in tax officials in
FBR, the progress is limited. Although FBR publicizes that it conducts ethics training of tax
officials not only at inductions, but also many times during their careers, incorporated as essential
ingredients of mandatory training exercises compulsory for promotions. These ethics trainings
are conducted by the anti-corruption agency (NAB) staff (ADB, 2006).
But what did the interview respondents think about the effectiveness of code of conduct and
ethics training in practice in FBR towards controlling motivations for corruption. When asked if the
adoption of reformed code of conduct by FBR has deterred tax officials from committing
corruption? Tax Official 10 replied:
‘I am aware that the code exists; I also read it in excitement when I was given a copy at
my induction training. It exists but serves no useful purpose other than for seniors to
boast that they have developed a code themselves. If it was any useful, would there be
so much corruption?’
Another tax official 11 had a thoughtful instance on the code of conduct:
‘There are three categories of officers in FBR. One category is extremely corrupt. They
don’t care about the code. Second category includes the honest ones, come what may.
They also don’t need the code as much. The third much bigger category includes
majority of tax officers in the middle, who can’t be classed as either corrupt or honest as
they are still learning about the system. The code might be most beneficial for this
category, to convince them to follow the right path. In practice I see the code rarely
applied or talked about. I don’t think it does much to influence the thinking of those who
are in the middle and prone to corruption’.
When discussing the contents of the code of conduct, tax official 12 elaborated:
‘It’s quite well written and explains all the definitions well. It’s a sort of manual about why
not to do corruption with all details of penalties etc. But it’s definitely no deterrence for
the corrupt ones, as they do everything explained in the code including taking gifts and
219
bribes etc. I think if they combine this with severe penalties for corruption, then officers
might start taking it seriously’.
In addition to the code of conduct at FBR, interview respondents had similar opinions about
effectiveness of ethics training towards curbing motivations for corruption. When asked if the tax
officials are less likely/less motivated to commit corruption in FBR as a result of increased ethics
trainings? Tax Official 12 explained:
‘I don’t think there are any special ethics trainings per se. Sure the generalized trainings
have some ethics components in them, but that’s all. Nothing new! They mostly repeat
what we already know, like definitions etc. and what to? and how to? about corruption. I
don’t come out of these trainings knowing something new about corruption’.
Tax Official 11 also contributed on the similar line:
‘I think FBR is obligated to conduct these trainings with ethics components, so that the
seniors can tell the donors in the reform meeting for getting more funds that see we have
done so much to educate officers about ethics and corruption. But they don’t do much to
catch them or to penalize them. Surely I need to know what is corruption etc. but it
shouldn’t end there. They should do more than only establishing codes and trainings for
ethics’.
Tax Official 12 put it well:
‘These trainings have done nothing to deter me from committing corruption if I want to.
Yes they have made me more aware about corruption and its different forms. Trainings
are more about awareness than deterrence against corruption’.
Referring to his experience in his last training exercise, Tax Official 13 explained:
‘By the end of training session it became a big discussion group with everyone
contributing with stories of corruption from their regions. What was the end result? I
came out of training more convinced that nothing could be done about corruption at FBR,
and I am really foolish for not being corrupt. That’s what everyone suggested anyway. So
much for motivation!’
Taxpayer 4 opined:
220
‘Well there goes another way of wasting taxpayer money on arranging ethics trainings for
tax officials. What’s the point of these trainings? If they were any good, would we see so
much corruption everywhere in Pakistan’.
Taxpayer 1 also said:
‘I think what would be more effective than these ethics trainings would be to conduct
regular name and shame events where they highlight the recent cases of officers getting
caught and punished. But wait, first they’ll need to catch them, which they aren’t!’
MoF official had a different take on this issue:
‘Well it’s a good thing that we have these trainings and code. If we wouldn’t conduct
them, people like you would criticize us for not doing so. Yes now the criticism is on their
effectiveness against corruption. But at least give the government some credit for having
such training exercises. At least we are on the right path’.
These views about ineffectiveness of efforts to instill ethics in tax officials towards curbing
seems to be that in the absence of backing-up of efforts to instill ethics in tax officials through
code of conduct with increased probability of getting caught and penalized, most interview
respondents viewed establishment of reformed code of conduct by FBR as a mere adherence to
procedures exercise. In addition, most interview respondents did not notice any special emphasis
towards increased ethics trainings in FBR, rather they were perceived as compulsory
components of generalized trainings. These trainings did not do much to deter against corruption,
rather they were more aimed towards creating awareness about corruption.
The above micro level of analysis was aimed at analyzing research hypothesis 1(d) for FBR. This
analysis illustrated that although FBR did utilized personnel autonomy to instill ethics in tax
officials, but was not able to effectively control motivations for corruption. Hence, FBR did not
lend support to the basic argument behind research hypothesis 1(d) as per micro level of
analysis, as the progress against controlling motivations for corruption improved least in FBR,
although personnel autonomy in terms of instilling ethics in tax officials was high. In FBR,
221
personnel autonomy proved ineffective in controlling motivations for corruption by instilling ethics
in tax officials. Hence, research hypothesis 1(d) was not supported by FBR.
6.2.5 Integration of Sub-Analyses for Research Hypothesis 1
As highlighted at the start of this section, the micro level of analysis was aimed to analyse
research hypothesis 1. This research hypothesis comprised of four further sub-hypotheses (1a,
1b, 1c, and 1d), which have already been examined in sections above. As it was proposed at the
start of section 6.2, analysis of these four sub-hypotheses will be undertaken with an aim to test
main research hypothesis 1, and any answers or findings as per analysis of these sub-
hypotheses will mean to respond to research hypothesis 1. Consequently this section aims to
integrate all four sub-analyses discussed in sections above for the cumulative analysis of
research hypothesis 1. The research hypothesis 1 was adapted for FBR and proposed that by
preferring personnel autonomy over effective accountability, FBR have been effective against
motivations but not opportunities for corruption. Table 6.4 illustrates the accumulation of all four
sub-analyses discussed in sections 6.2.1, 6.2.2, 6.2.3 and 6.2.4 of the micro level of analysis with
respect to the link between personnel autonomy and all four preventive strategies to control
motivations for corruption, in addition to providing latest CPI rating for Pakistan.
Table 6.4: Overview of FBR concerning Personnel Autonomy and all 4 preventive
strategies to control Motivations for Corruption
Personnel Autonomy
Corruption
Increases in Increases in
Country Increases in Instilling Perception Index
the and Stricter
Wages and Ethics in Tax (CPI) 2012*
Probability of Enforcement
Rewards Officials
Detection of Penalties
Pakistan - - - - 27
+ means personnel autonomy resulted in increase in the specific preventive strategy, which led to effectively
controlling motivations for corruption, as per micro level of analysis in the sub-hypotheses analyses.
- means personnel autonomy resulted in increase in the specific preventive strategy, which did not led to effectively
controlling motivations for corruption, as per micro level of analysis in the sub-hypotheses analyses.
*Source: Transparency International. Available online http://cpi.transparency.org
222
This table presents a snapshot of progress of FBR towards utilizing different preventive strategies
to control motivations for corruption by ensuring personnel autonomy. As depicted by the -
symbols for all four preventive strategies in table 6.4, FBR did not provided support for research
hypothesis 1 by showing that adoption of personnel autonomy (in terms of all four preventive
strategies) did not resulted in FBR being effective in controlling motivations for corruption. In
other words, FBR did not made an optimal use of personnel autonomy in terms of all four
preventive strategies to control motivations for corruption. Table 6.4 also illustrates the findings in
terms of individual sub-hypotheses. It highlights that in terms of all sub-propositions (1a, 1b, 1c,
and 1d) the micro level of analysis did not provided support for FBR analyzed for all four
preventive strategies (signified by - symbols). The discussions in the micro level of analyses in
the previous sections showed how FBR made less than optimal use of personnel autonomy to
control motivations for corruption by increasing wages and rewards, increases in the probability
of detection, increases in and stricter enforcement of penalties for corruption, and instilling ethics
in tax officials. This contention for FBR was also supported by Pakistan’s latest CPI rating, which
is one of the lowest among its regional developing countries and selected SARA developing
countries, showing Pakistan not doing well to control corruption over time and placing it as one of
most corrupt countries among the groups. How could this finding be explained for FBR? It is
suggested that FBR represents a case which on paper made efforts towards ensuring personnel
autonomy in terms of all four preventive strategies to control motivations for corruption. But in
practice its progress against controlling motivations for corruption has not improved owing to
improper implementation of any real personnel autonomy. It was highlighted that FBR remained
ineffective towards imparting any real and effective personnel autonomy in terms of all four
preventive strategies to control motivations for corruption.

223
In sum, this section was aimed at analyzing the overall proposition contending that if FBR has
made progress towards ensuring personnel autonomy, then it can also be taken to propose that
progress towards controlling motivations for corruption has been made. FBR did not supported
this contention by showing how adoption of personnel autonomy (in terms of all four preventive
strategies) did not resulted in FBR being effective in controlling motivations for corruption. Hence,
it is pertinent to conclude that research hypothesis 1 was not supported in case of FBR.
6.3 INEFFECTIVENESS OF ACCOUNTABILITY TO CONTROL OPPORTUNITIES FOR
CORRUPTION IN FBR
In order to analyze the potential and suitability of adopting Anti-Corruption SARA framework for
FBR Pakistan to control corruption in tax administration, the micro level of analysis in this section
is aimed to analyze research hypothesis 2 along with its three sub-hypotheses. It is further
highlighted here that research hypothesis 2 along with its sub-hypotheses have been adapted to
examine tax administration reform in FBR Pakistan. This adaptation is undertaken by replacing
the term ‘SARA’ in these hypotheses with ‘FBR’. Hence, in this chapter it is proposed that FBR
has been ineffective in controlling opportunities for corruption due to ineffective accountability by
MoF. Research hypothesis 2 comprises of three sub-hypotheses (2a, 2b, and 2c), with each one
referring to one specific preventive strategy to control opportunities for corruption. Effective
accountability is conceptualized to control opportunities for corruption by 1) introduction of
oversight mechanisms; 2) de-politicization of tax officials and 3) reduction of discretionary powers
of tax officials. First all three sub-hypotheses will be analyzed and then these analyses will be
integrated for the cumulative analysis of research hypothesis 2 in sections below.
The next sections examine how FBR used effective accountability to control opportunities for
corruption by utilizing both primary (semi-structured interviews) as well as secondary literature.

224
The overall proposition is that if FBR has ineffective accountability relationships due to
interference form MoF, then it would also be ineffective in controlling opportunities for corruption.
6.3.1 Ineffectiveness of Accountability to Control Opportunities for Corruption in FBR by
Failing to Introduce Effective Oversight Mechanisms
Research hypothesis 2(a) has been adapted for FBR and proposes that FBR has been
ineffective in controlling opportunities for corruption due to ineffective accountability by MoF by
not introducing effective oversight mechanisms. Effective accountability is operationalized to
control opportunities for corruption in FBR by affecting implementation of preventive strategy of
introduction of oversight mechanisms with the indicators being effective accountability
relationships with all supervising bodies including Ministry of Finance, Revenue
Boards/Superintendents (whichever applicable), and Parliament etc. This section aims to analyze
evidence on tax administration reforms in FBR Pakistan with particular focus on if and how
effective accountability is being utilized by FBR to control opportunities for corruption by
introduction of oversight mechanisms.
To introduce effective oversight mechanisms for tax officials, after reform the FBR was
restructured on the model of a revenue board, in contrast to a superintendent or CEO model, and
consists of total 13 members. These members include: 1) Member (Facilitation & Taxpayers
Education; 2) Member (Inland Revenue – Policy); 3) Member (Enforcement & Withholding Tax);
4) Member (Administration); 5) Member (Inland Revenue - Operations); 6) Member FBR (Port
Folio not yet assigned); 7) Member (Taxpayers Audit); 8) Member (HRM); 9) Member
(Accounting); 10) Member (Customs); 11) Member (Legal); 12) Member (Strategic Planning,
Reforms & Statistics); 13) Member IT (Information Technology) (FBR, 2013) . It is worth
highlighting here that at the time of writing, two members of the board occupied two portfolios
225
each, namely Member (Taxpayers Audit) also had the additional charge of HRM. While Member
(Customs) had an additional charge for portfolio of Strategic Planning, Reforms & Statistics. This
double occupancy by two members, in the presence of many eligible senior tax officials in FBR
raised issue of merit in the appointment of members of the board. It is particularly pertinent since
the two portfolios each assigned to these members were not even related to each other, and
separately demanded full attention by the members. Another issue worth highlighting here
involves appointment of members of the board from the open market. FBR (2008) publicized that
after reform, FBR hired professional and skilled members from the open market by following a
fair recruitment policy. The board composition, as it stands, is made up of all members belonging
to the civil service cadre of Pakistan. There does not seem to be any representation of members
from the private sector as well, such as members from professional and reputable trade
organizations, such as Chambers of Commerce and Industry etc. In essence, the board
composition is fairly closed in terms of representing views and interests of the private sector. The
organizational structure of FBR is illustrated in figure 6.5 below:

226
Figure 6.5: Organogram of FBR
Source: FBR (2013)
As per Federal Board of Revenue Act, 2007 (Appendix 4) the board is stipulated to be consisting
of at least seven members to be appointed by the Federal Government. The federal government
also appoints the chairman FBR with no stipulated details of terms and conditions in the act, and
these are generally determined by the government on case per case basis. The chairman FBR,
in addition to being the executive head of the board, also occupies the position of Secretary
Revenue Division. The chairman FBR in turn appoints the secretary of the board. The board is
entrusted with the responsibility of handling internal transfers and postings of employees of FBR;
to grant incentives and rewards to employees including board members; to take suitable actions
to fight corruption within FBR, and to safeguard integrity of employees through periodic checks;
227
to undertake inquiries involving tax evasion, fraud, money laundering etc.; to continually monitor
performance and conduct of employees; and to assess integrity of employees for promotions and
placements, among several other functions (FBR Act, 2007). Other responsibilities of the board
include formulation of annual budget of the FBR to be submitted to finance division for approval,
in addition to preparing an annual report of progress to be presented to the prime minister,
national assembly and senate. The FBR in turn is overseen by Cabinet Committee for Federal
Revenue. The federal government is entrusted with the power of constituting the cabinet
committee and to nominate its members. This committee is headed by the Minister of Finance
along with the Secretary Ministry of Finance included as the ex officio member of the Committee.
On paper it seems as if MoF has adopted a hands-off approach towards FBR, but in practice the
situation on the ground is quite different. Take for example the case of appointment of chairman
FBR. The FBR Act, 2007 stipulates that the federal government appoints chairman FBR. In
practice, the MoF in Pakistan exercises significant control over appointment of chairman FBR, so
much so that it is not possible that chairman FBR could be appointed against the wishes of
minister of finance. The establishment division prepares a summary with generally three
prospective names for post of chairman FBR, which is forwarded to the prime minister for
approval. In practice, the prime minister does not take decision of appointing chairman FBR
without the consent and advice of minister of finance.
More or less the post of chairman FBR is a political one, with practically no security of tenure in
Pakistan. It is very commonplace that the chairman FBR is replaced with every change of
government in Pakistan, as every new government is generally keen to appoint their own
favorites on this important post. In its 66 years of existence, so far Pakistan has seen 41 heads of
its central tax agency (Appendix 5). This high figure demonstrates a very poor security of tenure
228
for this important post. The average tenure of different chairmen over 66 years period has been
exactly 1.5 years (18 months), which is quite low. Importantly, since its inception in 2005 after tax
administration reform, FBR has seen a total of nine different chairmen changing hats at the helm
of affairs. In other words, in its eight years of existence, FBR has already seen nine different
heads, with an average tenure of less than a year.
What is the effect of this high turnover rate of chairman FBR in practice? World Bank (2009)
noted that one of reasons TARP did not quite got-off the ground was due to frequent changes of
FBR leadership at the top during the reform exercise. Khan (2012a) highlighted that during the
TARP implementation period, four chairmen and eight project directors were changed. More
recently, three project directors were changed in just one year, despite that they were actively
involved with preparing for the next TARP-II. This high turnover rate for chairman FBR has not
gone un-noticed by the traders and industrialists in Pakistan, who are deeply affected by these
changes, and have time and again call for end of political victimization. The All Pakistan Traders
Association repeatedly called for making FBR an independent institution which is free from undue
influence of the present government (The News, 2012e). The General Secretary of the
Association stressed on the government to make constitutional amendment to rules regarding the
appointment of FBR with a stipulated tenure of at least four years to aid continuation of revenue
policies.
There have also been instances where present governments have hired their favorites as
chairman FBR without following the laid down procedures, rules and merit. In one such case, the
matter went to court and the chairman was subsequently removed from the post on account of
non-adherence to at least 10 essential requirements of the appointment procedure, including
failing to follow a competitive process by not advertising the post, and no examination of the
229
candidature by the selection committee. It was argued against the present incumbent that the
post of chairman FBR and Secretary Revenue Division was mainly a civil service cadre position,
too crucial to be open for private sector persons. It was also criticized that the appointment was
made by the prime minister on a summary forwarded by the establishment division containing
just one name, although the standard practice was a choice among three names. The counsel
defending chairman surprised everyone in the court by trying to defend the case on the ground
that there is no specific criterion or set procedure for the appointment of chairman FBR. The
appointment was termed illegal by the court and the chairman, who was from the private sector,
was removed from the post (Asad, 2013). It is also worth highlighting here that the removal of
chairman was swiftly followed by the removal of one of his close aides, working not only as senior
member of the board, but also the official spokesperson of FBR. His contract with FBR was also
terminated, thus also demonstrating poor security of tenure of board members as well. While
commenting on the nexus between the bureaucrats and politicians in Pakistan resulting in mega
cases of corruption, the chief of Pakistan’s central anti-corruption agency NAB commented:
“There are only two pillars, and not three, of the state now in the country - one consists of
legislators and executives and the other is judiciary. The legislators appoint the
executives of their choice and this is happening at the federal as well as provincial levels”
(Raza, 2012, p. 1).
But what did the interview respondents think about the steps taken by FBR to introduce effective
oversight mechanisms for tax officials? When asked if FBR failed to introduce effective oversight
mechanisms for tax officials due to ineffective accountability due to interference form MoF? Most
of the interview respondents agreed. Tax Official 14 explained:
‘It is just not possible that the Prime Minister would appoint the chairman FBR without the
agreement of finance minister. Otherwise if something undesirable would happen in FBR
in the future, then minister would just raise hands and say ‘told you’. Who would prime
minister blame? The minister of finance runs the whole finance team and FBR is a very
important component’.
230
Tax Official 16 elaborated in a different dimension:
‘Just look at the history of chairmen FBR, how many do you find from either the customs
or IRS groups. Almost non-existent! Most of chairman FBR are appointed from the
District Management Group (DMG) cadre of the civil service. Now again have a look at
how many former secretaries of MoF have been from DMG group. Too many! It all
explains, isn’t it? It all comes down to clan mentality in DMG’.
He added:
‘The board is accountable to Cabinet Committee for Federal Revenue. Who heads this
committee? It’s the minister of finance, how can chairman FBR go against his wishes?’
Also, Businessman 5 responded:
‘The post of chairman is very important for us. They bring with them a certain mindset
and we have to adjust our business practices accordingly. But they are abruptly
changed. This is not good for our business as the new chairman comes and immediately
nullifies previous decisions. There should be security of tenure for chairmen, at least five
years I would say’.
Another Businessman 2 said:
‘The chairmen have to follow whatever the politician ssays and the minister of finance is
very powerful anyways. I always try to find links to establish rapport with the minister.
The chairmen come and go frequently’.
These views about ineffectiveness of oversight mechanisms of FBR towards curbing
opportunities for corruption were consistent with most interview respondents. The main problem
seems to be that in the absence of any real autonomy granted to chairman FBR to introduce
effective oversight mechanisms along with poor security of tenure, MoF continued to interfere
with the accountability relationships of FBR along with other oversight bodies such as Cabinet
Committee for Federal Revenue. In practice, this committee was supposed to replace the control
by MoF, but due to continued interference from the ministry, the committee did not seem to help
with curbing opportunities for corruption, especially due to spillage of corruption networks from
231
MoF to FBR. In practice, the MoF emerged as the main overseer of FBR, despite projecting a
more hands-off approach on paper.
The above micro level of analysis was aimed at analyzing research hypothesis 2(a) for FBR. This
analysis illustrated that FBR made less than optimal use of effective accountability to introduce
effective oversight mechanisms, due to interference from MoF, and was less effective in
controlling opportunities for corruption. Hence, FBR lend support to the basic argument behind
research hypothesis 2(a) as per micro level of analysis. In FBR, effective accountability proved
ineffective in controlling opportunities for corruption by failing to introduce effective oversight
mechanism for tax officials. Therefore, research hypothesis 2(a) was supported by FBR.
Failing to De-politicize Tax Officials
Research hypothesis 2(b) has been adapted for FBR and proposes that FBR has been
ineffective in controlling opportunities for corruption due to ineffective accountability by MoF by
not de-politicizing tax officials. Effective accountability is operationalized to control opportunities
for corruption in FBR by affecting implementation of preventive strategy of de-politicization of tax
officials with the indicators being effective accountability relationships with all supervising bodies
especially Revenue Boards/Superintendents (whichever applicable), and Parliament etc., in
addition to Ministry of Finance. This section aims to analyze evidence on tax administration
reforms in FBR Pakistan with particular focus on if and how effective accountability is being
utilized by FBR to control opportunities for corruption by de-politicizing tax officials.
Politicization of civil servants is very common in Pakistan, with every change of government
signaling across-the-board major shuffle at all levels. Especially the civil servants at the top are
232
changed by the new governments to bring in their own favorites and team. FBR, being the central
tax collection agency also bears the brunt of these political appointments. As already highlighted
in the section above, the post of chairman FBR is a political one and practically has no security of
tenure. In addition to the high turnover rate of chairman FBR, every such change also brings a lot
of shuffling down the line in FBR as well. In practice, many new incumbents in the office follow
their arrival with a lot of changes in FBR, in the form of transfers and postings to bring in their
own teams. In one such instance, one newly inducted chairman FBR transferred 16 top officials
within a month of assuming charge. This shuffle included members of the board and senior
officials in tax collectorates and units (Khan, 2011). This was followed by another 100 senior level
tax officials shuffled across FBR within the next month at key collectorates and large taxpayer
units. According to tax officials these large-scale shuffling in the mid fiscal year significantly
hampered efforts to achieve the set revenue target for the year, and to plug loopholes and
unearth tax evasion, on account of time taken by transferred tax officials to make adjustments
with the new setups.
The change of the last government in Pakistan also saw a major re-shuffling of bureaucracy,
more severe at the federal level. As a result all major heads of institutions, including FBR, were
changed to bring in their own favorites (The News, 2013e). This situation makes sense about
why federal government has chosen to stipulate appointment of chairmen FBR with no details of
terms and conditions in the Federal Board of Revenue Act, 2007, which are determined by the
government on case per case basis. These clauses of the act unquestionably aid towards
changing chairmen FBR at the whim of the politicians as and when they deem fit, hence making
this post very political one.

233
In order to de-politicize tax officials the FBR has tried its hand at job dismissals for purging tax
administration from patronage networks of political appointees. In 2000, under the Removal from
Service Ordinance (2000) FBR tried to purge a significant number of suspected tax officials by
suspending them from the service. But this exercise failed to get rid of any significant number of
suspended officials with majority joining back their jobs without facing any penalties, on account
of political connections (Shah, 2011). He also noted that operational level tax officials in FBR,
especially in customs, are very often inducted merely on the basis of patronage networks based
on familial or ethnic grounds, which enable them to work with complete immunity from any
adverse action. Accusing FBR of promoting institutionalized corruption, the FTO noted:
“FBR promotes and posts corrupt officers at key posts deepening the perception of
institutionalized corruption in the department” (Rana, 2013b, p. 3).
But what did the interview respondents think about the steps taken by FBR to de-politicize tax
officials? When asked if FBR failed to de-politicize tax officials due to ineffective accountability
due to interference form MoF? Most of the interview respondents agreed. Tax Official 14
explained:
‘I will not say that everyone uses political connections to get good postings, because
everyone does not have good political connections. But those who do, they are the ones
who are always posted at best posts, money-making wise, and they are the ones who
always get out-of-turn promotions. You just need to know the right person in the ministry’.
While commenting on the across-the-board transfers with the change of government, Tax Official
13 noted:
‘Yes most of us are transferred, but not the powerful ones. If they are happy where they
are, and they have some friends at the ministry, then they’ll remain at the post. They only
move if they have somewhere better to go’.
234
Tax Official 15, while commenting about the politicization of the post of chairman FBR added:
‘All the chairmen have to make sure that they keep the minister of finance to their sides.
It is essential if they want to remain in the post. There have been few cases when the
finance minister simply got the chairman removed without any notice. But it’s very rare.
When you do get to such a high place in the system, you are well aware how to play in it.
Otherwise you would not have got there in the first place. In most cases chairman and
minister are friends or good colleagues. They know each other for years and that’s why
they are there!’
Tax Official 16 added:
‘It’s almost impossible to depoliticize tax officials. How can you de-politicize them? They
are here because they are politicized, and they work as if it’s their birth right to be in
FBR. They believe nobody can touch them. Mostly it’s right. How can you take any
disciplinary action against such politically well-connected employees?’
While recalling a personal experience, tax official 12 detailed:
‘One of my colleagues has recently been transferred three times in three months. Why?
Because he helped to recover Rs. 4 billion worth of taxes from the textile lobby. The
lobby tried to bribe him but he did not agree. Their next step was to bribe someone in
ministry to get him transferred. It’s as simple, like following steps. This would not have
happened to him if he was not honest or if he was politically aligned’.
Also, Rehman Azhar from Dunya News commented:
‘Due to my job nature, I see these connections every day. All prominent politicians have
favourite officers. If the politician assumes office, in no time they’ll bring their blue-eyed
boys to serve them even if they are not the best people for the job. Recently, I have seen
the emergence of these ‘celebrity’ bureaucrats, who are the favourites of the ruling
politicians, occupy major positions and would appear all the time in news and TV to
openly support the actions of their political masters. Gone are the days of impartial
bureaucrats, its partiality what keeps them at prized postings’.
Such sentiments about failure to de-politicize tax officials in FBR towards negatively affecting
control of opportunities for corruption were consistent with most interview respondents. The main
problem seems to be that in the absence of any real autonomy granted to FBR to de-politicize tax
officials, MoF continued to interfere with the accountability relationships of FBR with other
235
oversight bodies, not helping with curbing opportunities for corruption, especially due to spillage
of corruption networks from MoF to FBR.
The above micro level of analysis was aimed at analyzing research hypothesis 2(b) for FBR. This
analysis illustrated that FBR made less than optimal use of effective accountability to de-politicize
tax officials, due to interference from MoF, and was less effective in controlling opportunities for
corruption. Hence, FBR lend support to the basic argument behind research hypothesis 2(b) as
per micro level of analysis. In FBR, effective accountability proved ineffective in controlling
opportunities for corruption by failing to de-politicize tax officials. Therefore, research hypothesis
2(b) was supported by FBR.
Failing to Reduce Discretionary Powers of Tax Officials
Research hypothesis 2(c) has been adapted for FBR and proposes that FBR has been ineffective
in controlling opportunities for corruption due to ineffective accountability by MoF by not reducing
discretionary powers of tax officials.Effective accountability is operationalized to control
opportunities for corruption in FBR by affecting implementation of preventive strategy of reduction
of discretionary powers of tax officials with the indicators being balancing effective accountability
relationships between all supervising bodies especially between Ministry of Finance and other
supervising bodies including Revenue Boards/Superintendents (whichever applicable) and
Parliament etc.This section aims to analyze evidence on tax administration reforms in FBR
Pakistan with particular focus on if and how effective accountability is being utilized by FBR to
control opportunities for corruption by reducing discretionary powers of tax officials.
As per Federal Board of Revenue Act, 2007, the powers and functions of the board includes:
236
“The Board may, where appropriate, issue statutory rules and orders (SROs), orders,
circulars and instructions for the enforcement of any of the provisions of fiscal law and
the provision of this Act” (FBR Act, 2007, p. 7).
As per the act, the board is empowered to exempt any products from tax by issuing a SRO and
thus superseding the approval of the parliament. The SROs are issued at the discretion of FBR to
benefit the interests of few, to make changes to the tax structures which were approved by the
parliament, thus overriding acts and finance bills. In practice, the board is severely criticized for
using the discretionary powers of issuing SROs immensely for granting favors of tax exemptions
to different powerful lobbies probably to either win friends or for rent-seeking (The Economist,
2012).
In 2008, Pakistan went for another $11.3 billion IMF program. One of the major conditions of this
IMF program included a tax reform especially targeted towards elimination of SRO powers of the
board. Pakistan vowed to remove board’s discretionary powers to issue SROs. This program
collapsed in 2010 with Pakistan failing to keep up with its promise and eliminating use of SROs.
This program failed due to strong resistance from FBR administration refusing to let go of rent-
seeking opportunities. The major failure was attributed to resistance by those in powers in
Pakistan, corrupt politicians and tax administrators alike, unwilling to abandon opportunities for
corruption. According to finance minister of Pakistan between four year period of 2008-2012 the
total amount of tax exemptions granted to some sectors via SROs exceeded the total amount of
funds borrowed from IMF during the same period (Kiani, 2013a). In other words, if these sectors
were not exempted from taxes via issuance of SROs, their contribution to Pakistan economy, in
the form of duties and taxes paid, would have been adequate to escape the expensive loan from
IMF. Ironically, one of the conditions of this IMF loan included removal of board’s discretionary
powers to issue SROs, which remained undone.

237
This issuance of SROs causes the national kitty almost Rs.180 billion every year in terms of
customs duties. The imports under the SROs blanket are equal to about 45% of the total imports
of the country. In 2011-12, the imports under SROs totaled to Rs. 1.8 trillion out of Rs. 4 trillion
total imports (Akhter, 2013). These figures were given by a special study conducted by customs
in FBR. It highlighted preferentialism in the issuance of SROs by the board, beneficial for only
certain individuals and sectors including protected managers in the industry and public sector,
and political decision makers. The SRO culture has not only lead to exemptions which deprives
the government of billion worth of duties, but also abolishes the level playing field for all sectors,
especially hitting the small businesses and exporters with no exemptions, thus creating
distortions in the economy. The use of discretionary powers for issuing SROs not only causes
loss of revenue, but also adds to the cost of doing business, promoting corruption by encouraging
ineffective businesses to develop on rent-seeking, pushing out efficient, professional businesses,
and costly but poor quality products for customers (Kiani, 2013a). Despite this, the board is
content with retaining powers to grant exemptions, beneficial to only selective individuals and
sectors of the economy.
The situation got so worse that the Public Accounts Committee (PAC) not only took strong notice
of issuance of SROs, but also summoned top officials of the MoF, FBR, and the Planning
Commission to explain why parliament was being sidestepped in this manner. The chairman PAC
while referring to issuance of almost 4500 SROs by the FBR for either increasing or decreasing
tax tariffs termed it as an encroachment on parliament’s authority by FBR towards levying or
increasing/decreasing any taxes. While highlighting the entitlement of parliament towards
imposing taxes, he noted:

238
“The FBR is doing the job of Parliament through issuance of the SROs and so far 4,500
SROs were issued, which is not fair. It seems the country’s taxation system is running on
the SROs” (Yasin, 2012, p. 1).
But what did the interview respondents think about the steps taken by FBR to reduce
discretionary powers of tax officials? When asked if FBR failed to reduce discretionary powers of
tax officials due to ineffective accountability due to interference form MoF? Most of the interview
respondents agreed. Tax Official 13 elaborated:
‘See what is happening with SROs, everybody seems to criticize it, but it’s still
happening. Obviously these exemptions are not fair. People are right to think that the
importers simply collude with the higher ups in FBR and ministry to get SROs issued in
their favors. With one SRO, they save billions worth of duty. How can it be fair for other
businesses?’.
Tax official 14 added:
‘It’s just about simple math. If you import without any concession or exemption, then you
pay so much in duties and taxes. If you pay them for the issuance of SRO in your favor
to get your specific industry exempted from duty, you will import without any duties and
taxes. If you are a businessperson and willing to collude, where would you go? Obviously
where there is more profit!’
He added:
‘I know many businessmen who are very honest otherwise. They are religious, pray five
times every day, give charity, help needy etc. but they also pay for the issuance of
SROs. That’s just the norm, that’s how business is done. If they won’t do it, they won’t be
able to compete against the ones who already have exemptions. The government has
not created a level playing field for them. They have just created it themselves’.
Tax Official 12 added on the same line:
‘For the business community paying bribes for SROs etc. is no longer shameful. It’s just
how business is done here. They know officers at the top want their share in the profits.
Otherwise why would the board still keep the power of issuing SROs to itself, despite so
many hue and cries by everyone including the donors’.
In the similar line, Businessman 4 noted:

239
‘The issuance of these SROs is making our economy hemorrhage. This is very bad
business by people of FBR but as long as I have been in business, it’s happening. I don’t
think so anybody is even seriously trying to stop it’.
Businessman 2 opined:
‘Issuance of SROs to favor certain businesses happens at the very high level and takes
a lot of money. Only very successful businessmen are able to get SROs in their favor
from FBR. Small traders like me can do nothing but to curse and keep on trying to
making ends meet. They don’t even pay all of their taxes because they have friends at
FBR’.
These views about failure to reduce discretionary powers of tax officials in FBR towards
negatively affecting control of opportunities for corruption were consistent with most interview
respondents. The main problem seems to be that due to ineffective accountability of FBR due to
interference form MoF, tax officials seem to abuse their discretionary powers to use issuance of
SROs as a rent-seeking tool. This was happening in collusion with MoF and in the absence of
any sufficient checks by the oversight bodies on the discretionary powers of tax officials in FBR,
not helping with curbing opportunities for corruption.
The above micro level of analysis was aimed at analyzing research hypothesis 2(c) for FBR. This
analysis illustrated that FBR made less than optimal use of effective accountability to reduce
discretionary powers of tax officials, due to interference from MoF, and was less effective in
controlling opportunities for corruption. Hence, FBR lend support to the basic argument behind
research hypothesis 2(c) as per micro level of analysis. In FBR, effective accountability proved
ineffective in controlling opportunities for corruption by failing to reduce discretionary powers of
tax officials. Therefore, research hypothesis 2(c) was supported by FBR.

240
6.3.4 Integration of Sub-Analyses for Research Hypothesis 2
As highlighted at the start of this section, the micro level of analysis was aimed to analyse
research hypothesis 2. This research hypothesis comprised of three further sub-hypotheses (2a,
2b, and 2c), which have already been examined in sections above. As it was proposed at the
start of section 6.3, analysis of these three sub-hypotheses will be undertaken with an aim to test
main research hypothesis 2, and any answers or findings as per analysis of these sub-
hypotheses will mean to respond to main research hypothesis 2. Consequently this section aims
to integrate all three sub-analyses discussed in sections above for the cumulative analysis of
research hypothesis 2. The research hypothesis was adapted for FBR and proposed that FBR
has been ineffective in controlling opportunities for corruption due to ineffective accountability by
MoF. Table 6.5 illustrates the accumulation of all three sub-analyses discussed in sections 6.3.1,
6.3.2, and 6.3.3 of the micro level of analysis with respect to the link between effective
accountability and all three preventive strategies to control opportunities for corruption, in addition
to providing latest CPI rating for Pakistan.
Table 6.5: Overview of FBR concerning Effective Accountability and all three preventive
strategies to control Opportunities for Corruption
Effective Accountability
Corruption
Country Reduction of Perception Index
Introduction of
De-politicization Discretionary (CPI) 2012*
Oversight
of Tax Officials Powers of Tax
Mechanisms
Officials
Pakistan - - - 27
+ means effective accountability resulted in increase in the specific preventive strategy to control opportunities for
corruption, with no interference from MoF, as per micro level of analysis in the sub-hypotheses analyses.
- means effective accountability did not resulted in increase in the specific preventive strategy to control
opportunities for corruption, due to interference from MoF, as per micro level of analysis in the sub-hypotheses
analyses.
to control opportunities for corruption by ensuring effective accountability. As depicted by the -

241
symbols for all three preventive strategies in table 6.5, FBR provided considerable support for
research hypothesis 2 by showing that ineffective accountability (in terms of all three preventive
strategies) resulted in FBR being ineffective in controlling opportunities for corruption. In other
words, FBR did not made an optimal use of effective accountability in terms of all three
preventive strategies, due to interference from MoF, and was not able to effectively control
opportunities for corruption. Table 6.5 also illustrates the findings in terms of individual sub-
hypotheses. It highlights that in terms of all sub-hypotheses (2a, 2b, and 2c) the micro level of
analysis provided considerable support for all three preventive strategies (signified by - symbols).
The discussions in the micro level of analyses in the previous sections showed how FBR did not
made good progress in terms of ensuring effective accountability due to interference from MoFs
undermining introduction of oversight mechanisms, de-politicization of tax officials, and reduction
of discretionary powers of tax officials, leading to ineffectiveness in controlling opportunities for
corruption. This contention for FBR was also supported by Pakistan’s latest CPI rating as
illustrated in tables 6.2 and 6.3 above, which is one of the lowest among its regional developing
countries and selected SARA developing countries, showing Pakistan not doing well to control
corruption over time and placing it as one of most corrupt countries among the groups.
In sum, this section was aimed at analyzing the overall proposition contending that if FBR has
ineffective accountability due to interference form MoF, then it would also be ineffective in
controlling opportunities for corruption. FBR fully supported this contention by showing how
ineffective accountability (in terms of all three preventive strategies), due to interference form
MoF, resulted in FBR being ineffective in controlling opportunities for corruption. Hence, it is
pertinent to conclude that research hypothesis 2 was supported in case of FBR.

242
As highlighted at the start of this chapter, the micro level of analysis was aimed to analyse
research hypotheses 1 and 2 for FBR. The research hypothesis 1 comprised of four further sub-
hypotheses (1a, 1b, 1c and 1d), and research hypothesis 2 comprised of three further sub-
hypothesis (2a, 2b, and 2c), which have already been examined in sections above. As it was
proposed at the start of section 6.2 and 6.3, analysis of all sub-hypotheses will be undertaken
with an aim to test main research hypotheses 1 and 2, and any answers or findings as per
analysis of these sub-hypotheses will mean to respond to research hypotheses 1 and 2.
Consequently this section aims to integrate all seven sub-analyses discussed in sections above
for the cumulative analysis of research hypotheses 1 and 2. Next, Table 6.6 illustrates the
accumulation of all sub-analyses discussed in sections 6.2.5 and 6.3.4 of the micro level of
analysis for research hypotheses 1 and 2, in addition to providing latest CPI ratings for Pakistan.
243
Table 6.6: Overview of FBR accumulating sub-analyses of the micro level of analysis for research hypotheses 1 and 2
Research Hypothesis 1 Research Hypothesis 2

Personnel Autonomy Effective Accountability
Corruption
RH 1(a) RH 1(b) RH 1(c) RH 1(d) RH 2(a) RH 2(b) RH 2(c)
Perception
Country Increases in Increases in Increases in Instilling Ethics Introduction of De- Reduction of
Index (CPI)
Wages and the Probability and Stricter in Tax Officials Oversight politicization of Discretionary
2012*
Rewards of Detection Enforcement of Mechanisms Tax Officials Powers of Tax
Penalties Officials
Pakistan - - - - - - - 27
+ means personnel autonomy resulted in increase in the specific preventive strategy, which led to
effectively controlling motivations for corruption, as per micro level of analysis in the sub-hypotheses + means effective accountability resulted in increase in the specific preventive
analyses. strategy to control opportunities for corruption, with no interference from MoF, as
- means personnel autonomy resulted in increase in the specific preventive strategy, which did not led to per micro level of analysis in the sub- hypotheses analyses.
effectively controlling motivations for corruption, as per micro level of analysis in the sub- hypotheses - means effective accountability did not resulted in increase in the specific
analyses. preventive strategy to control opportunities for corruption, due to interference from
MoF, as per micro level of analysis in the sub- hypotheses analyses.
244
to ensure personnel autonomy and effective accountability for controlling motivations and
opportunities for corruption. Table 6.6 also illustrates the findings in terms of individual sub-
hypotheses. In terms of individual sub-hypotheses, research hypothesis 1 was not provided
support by any of the individual sub-hypotheses (indicated by – symbols for 1a, 1b, 1c and 1d).
Hence, as per individual sub-hypotheses, research hypothesis 1 was not supported in case of
FBR. Conversely, in terms of individual sub-hypotheses, research hypothesis 2 was provided
considerable support by all individual sub-hypotheses (indicated by – symbols for 2a, 2b, and 2c).
Hence, as per individual sub-hypotheses, research hypothesis 2 was fully supported in case of
FBR.
By referring to the - symbols for 4/4 preventive strategies of personnel autonomy for research
hypothesis 1, and - symbols for 3/3 preventive strategies of effective accountability for research
hypothesis 2 in table 6.6, FBR provided considerable support for research hypothesis 2 but not 1.
The micro level of analysis for both research hypothesis 1 and 2 showed that FBR made less
than optimal use of personnel autonomy to control motivations for corruption by increasing wages
and rewards for tax officials, increasing the probability of detection, increasingly and strictly
enforcing penalties for corruption, and instilling ethics in tax officials (indicated by - symbols for all
four preventive strategies in table 6.6). In addition, FBR’s progress against controlling
opportunities for corruption remained poor, due to persistent ineffective accountability, due to
continued interference from MoF, undermining introduction of oversight mechanisms, de-
politicization of tax officials, and reduction of discretionary powers of tax officials (indicated by -
symbols for all three preventive strategies in table 6.6). This poor progress against corruption by
not effectively controlling either motivations or opportunities for corruption led to worsening of
FBR’s progress against corruption. This finding for FBR is also supported by Pakistan’s latest
245
CPI rating as illustrated in tables 6.2 and 6.3 above, which is one of the lowest among its regional
developing countries and selected SARA developing countries, showing Pakistan not doing well
to control corruption over time and placing it as one of most corrupt countries among the groups.
FBR showed how adoption of personnel autonomy did not resulted in tax administration reform
being effective in controlling motivations for corruption, in addition to ineffective accountability,
due to interference form MoF, resulted in FBR being ineffective in controlling opportunities for
corruption. Hence, FBR remained ineffective in controlling both motivations and opportunities for
corruption. This section highlighted that there was a positive finding for FBR for research
hypothesis 2, but a negative finding for research hypothesis 1.
In a nutshell, the specific contribution of chapter six to the overall thesis argument lies in
conceptually testing the Anti-Corruption SARA framework for FBR by analyzing individual design
components in their separate contributions towards controlling motivations and opportunities for
corruption. This was to contribute towards filling a gap in literature for research on the practical
side of designing and implementing SARA reform against controlling corruption. This chapter
tried to contribute towards filling this gap by doing a practical work of designing an Anti-
Corruption SARA framework (chapter three) and applying it to FBR in this chapter.
In particular, this case study for the micro analysis also aimed to confirm if the findings reached
for SARA countries in the macro analysis in chapter five are also applicable to FBR Pakistan. It is
concluded that although FBR, like many SARAs, preferred adoption of personnel autonomy over
ensuring effective accountability for controlling corruption. But such preference did not led to
even partial progress against controlling motivations for corruption, and FBR remained ineffective
in controlling both motivations and opportunities for corruption.

246
This chapter also examined why FBR was unable to adopt effective accountability? Similarly to
many SARAs, in case of FBR as well, it was found that presence of MoF in the accountability
arrangements rendered FBR ineffective in controlling opportunities for corruption. Continued
interference from MoF was found to undermine not only the effective oversight mechanisms of
FBR, but also lead to continuation of politicization of tax officials, and resulting in increases in
discretionary powers available to tax officials in the face of ineffective accountability mechanisms.
In sum, by conducting the micro level of analysis, progress was made towards a two part answer
to the main question of this study. To explore why tax administration reform failed to effectively
control opportunities for corruption in FBR Pakistan? It was found that FBR remained ineffective
in controlling both motivations and opportunities for corruption despite focusing more on
controlling motivations for corruption (through personnel autonomy) and lesser focus on
controlling opportunities for corruption (through effective accountability). Continued interference
from MoFs was found to undermine not only effective accountability for FBR, but also
undermining control of opportunities for corruption. Based on the findings obtained in macro
analysis in chapter five, in case of FBR it was expected that opportunities for corruption might not
be controlled effectively, but it was unexpected that motivations for corruption were not being
controlled as well. So it was surprising in case of FBR that the situation against corruption was
so bad that motivations for corruption were not being controlled despite employing personnel
autonomy to curb them.
Chapters five and six have provided the results in the form of macro and micro level of analyses
examining SARA reforms in developing countries with special reference to Pakistan. The next
chapter reviews and discusses the key findings arising out of macro and micro level of analyses
and concludes.
247
CHAPTER SEVEN
DISCUSSION AND CONCLUSION
7.0 INTRODUCTION
The aim of the thesis has been to examine why some SARAs have failed to effectively control
opportunities for corruption. To explore this question this thesis set out an Anti-Corruption SARA
framework by analyzing individual SARA design components towards controlling motivations and
opportunities for corruption in tax administrations in developing countries. This analytical
framework was developed by utilizing two SARA frameworks including SARA design components
proposed by Taliercio (2004) and control of corruption framework for tax administrations by
Martinez-Vazquez et al. (2004). This thesis tried to contribute towards filling a gap in literature for
research on the practical side of designing and implementing SARA reform against controlling
corruption, by doing a practical work of designing an Anti-Corruption SARA framework (chapter
3) and applying it to SARAs (Chapter 5) and FBR Pakistan (Chapter 6).
In order to analyze the validity of the framework, a two-staged analysis of SARAs was conducted
in chapters five and six of the thesis. In chapter five, a detailed mapping of SARA cases, labelled
as the macro level of SARA analysis, helped to conclude that SARAs preferred adoption of
personnel autonomy over ensuring effective accountability for controlling corruption. In other
words, SARAs preferred adoption of autonomy-enhancing design components over
accountability-enhancing design components for controlling corruption. This imbalance led to
some progress in controlling motivations for corruption, but failing to effectively control
opportunities for corruption. It was also found that presence of MoFs in the accountability
arrangements rendered SARAs ineffective in controlling opportunities for corruption. Continued
interference from MoFs was found to undermine not only the effective oversight mechanisms of
248
SARAs, but also lead to continuation of politicization of tax officials, and resulting in increases in
discretionary powers available to tax officials. In chapter eight, a detailed case study of tax
administration reforms in Pakistan, labelled as the micro level of analysis, was conducted to
explore why tax administration reform failed to effectively control opportunities for corruption in
FBR Pakistan? It was found that FBR remained ineffective in controlling both motivations and
opportunities for corruption despite focusing more on controlling motivations for corruption
(through personnel autonomy) and lesser focus on controlling opportunities for corruption
(through effective accountability). Continued interference from MoFs was found to undermine not
only effective accountability for FBR, but also undermining control of opportunities for corruption.
This chapter summarises and discusses the analyses presented in previous chapters of the
thesis, in addition to making concluding remarks. This chapter is further organised into five
sections. Section 7.1 presents a summary of the thesis. Section 7.2 presents recommendations
for possible reforms and is sub-divided in three sections. Section 7.3 analyzes the limitations of
the study. Section 7.4 considers the contribution of the thesis to public management literature, in
addition to presenting suggestions for further research in the thesis area. Section 7.5 presents
concluding comments. The structure of Chapter seven is illustrated in Figure 7.1 below.
249
7.0
INTRODUCTION
7.1
SUMMARY OF THE THESIS
7.2
RECOMMENDATIONS FOR
POSSIBLE REFORMS
7.2.1 7.2.2 7.2.3

Culprit Ministries of A Case for Pakistan Adoption of SARAs:
Finance? Revenue Authority? Balancing Autonomy
with Accountability
7.3
LIMITATIONS OF THE STUDY
7.4
CONTRIBUTIONS OF THE THESIS
AND AVENUES FOR FURTHER
RESEARCH
7.5
CONCLUSION
250
7.1 SUMMARY OF THE THESIS
In recent years, there has been an increasing adoption of SARA reform model in developing
countries to achieve dual objectives of improving efficiency and controlling corruption in tax
administrations (McCourt, 2002; Fjeldstad, 2003, 2005a, 2005b; Kidd & Crandall, 2006; Mann,
2004; Martinez-Vazquez et al., 2004; McCarten, 2006; Taliercio, 2004; Zuleta, 2007). Although
review of literature in chapter two succeeded in improving our understanding of the reform model
at a conceptual level, it also pointed towards a theory-practice paradox. Literature highlighted that
experience of SARAs against corruption has not always been as expected or explained by
conceptual models’ consequences. This observation pointed to a need to closely analyse the
theory behind SARA reform, and try to find out why the literature is suggesting that practice
deviated from theory. In order to achieve this objective, this thesis analysed prominent
frameworks and strategies to control corruption in tax administrations, not only to throw some
light on theories behind SARA reform idea, but to look for answers to the theory-practice paradox
(See Chapter 2). The over-arching objective of this exercise was to further discussion on how to
better control opportunities for corruption in SARAs by analyzing prominent conceptual
frameworks in this field, resulting in development of analytical framework.
The development of analytical framework resulted in formulation of two research hypotheses
(See Chapter 3). Research hypothesis 1 proposed that by preferring personnel autonomy over
effective accountability, SARAs have been effective against motivations but not opportunities for
corruption. Also, research hypothesis 2 proposed that SARAs have been ineffective in controlling
opportunities for corruption due to ineffective accountability by MoF. With an aim to add focus
and specificity to both research hypotheses, so as to make the forthcoming analysis of SARA
cases more focused and to the point, sub-hypotheses were also developed. All sub-hypotheses,
with each one referring to one specific preventive strategy to control motivations and
251
opportunities for corruption, helped to logically break down both research hypotheses. The
specific contribution of chapter three to the overall thesis argument lies in conceptually
developing an Anti-Corruption SARA framework by drawing on literature reviewed in chapter two.
To explore why some SARAs have failed to effectively control opportunities for corruption, a
detailed mapping of SARA cases, labelled as the macro level of SARA analysis, was undertaken
for conceptually testing the Anti-Corruption SARA framework for multiple SARA countries (See
Chapter 5). This was to contribute towards filling a gap in literature for studies on SARAs
differentiating between motivations and opportunities for corruption. The macro level of SARA
analysis was aimed at theorizing what is going wrong with practice by approaching old evidence
(secondary literature on SARAs) in a new way (i.e., analytical framework). The macro level of
analysis found that SARAs have made partial progress to control corruption by focusing more on
controlling motivations for corruption (through autonomy-enhancing design components) and
lesser focus on controlling opportunities for corruption (through accountability-enhancing design
components). Continued interference from MoFs was found to undermine not only effective
accountability for SARAs, but also undermining control of opportunities for corruption.
To explore why tax administration reform failed to effectively control opportunities for corruption in
FBR Pakistan, a detailed case study of tax administration reforms in Pakistan, labelled as the
micro level of analysis, was conducted for conceptually testing the Anti-Corruption SARA
framework for FBR by analyzing individual design components in their separate contributions
towards controlling motivations and opportunities for corruption (See Chapter 6). The micro level
of analysis for FBR was aimed at theorizing what is going wrong with practice by approaching
both new and old evidence (semi-structured interviews and secondary literature) in a new way
(i.e., analytical framework). It was concluded that although FBR, like many SARAs, preferred
252
adoption of personnel autonomy over ensuring effective accountability for controlling corruption.
But such preference did not led to even partial progress against controlling motivations for
corruption, and FBR remained ineffective in controlling both motivations and opportunities for
corruption. It was also questioned why FBR was unable to adopt effective accountability?
Similarly to many SARAs, in case of FBR as well, it was found that presence of MoF in the
accountability arrangements rendered FBR ineffective in controlling opportunities for corruption.
Continued interference from MoF was found to undermine not only the effective oversight
mechanisms of FBR, but also lead to continuation of politicization of tax officials, and resulting in
increases in discretionary powers available to tax officials in the face of ineffective accountability
mechanisms.
Having detailed the summary of the thesis, the next section discusses recommendations for
possible reforms emerging out of findings of macro and micro analyses.
7.2 RECOMMENDATIONS FOR POSSIBLE REFORMS
The following section elaborates three major recommendations for SARA reform.
7.2.1 Culprit Ministries of Finance?
The first recommendation is related to findings coming out of macro level of SARA analysis and
is applicable to SARA developing countries. In chapter five, the macro level of SARA analysis
helped to conclude that presence of MoFs in the accountability arrangements rendered SARAs
ineffective in controlling opportunities for corruption. Continued interference from MoFs was
found to undermine not only the effective oversight mechanisms of SARAs, but also lead to
continuation of politicization of tax officials, and resulting in increases in discretionary powers
available to tax officials. Findings suggesting how SARAs’ accountability mechanisms were
253
undermined due to interference by MoFs, leading to ineffectiveness in controlling opportunities
for corruption, pointed towards a case of not keeping SARAs directly accountable to MoFs, in the
presence of other effective oversight bodies including Revenue Boards/Superintendents and
Parliament etc. These findings pointed towards a SARA reform design loophole, arising due to
insufficient level of disaggregation from MoFs, leading to ineffective accountability mechanisms of
SARAs with oversight bodies, and hence making it ineffective against opportunities for corruption.
The proposed solution to tackle opportunities for corruption lies in reforming SARAs into an
organizational form which are far more disaggregated from the parent ministry, such that SARAs
have no accountability link with MoFs. Furthermore, the oversight of SARAs can be replaced by
other suitable oversight bodies, such as parliament select committees. To elaborate this
recommendation further, three different organizational forms are illustrated on the disaggregation
continuum in figure 7.2 below.
Figure 7.2 Nomenclature for Solving SARAs Ineffectiveness against Corruption
‘Closer to’ (political control) ‘Further out’ (from political control)
More Autonomous
Bodies
(MABs)
Ministries Agencies SARAs
State Enterprises
and other
Corporate
Commercial
Organizations
Source: (Adapted from Pollitt et al., 2004)

254
This figure illustrates that the basic difference between agencies and (further out from ministries)
MABs is the level of disaggregation granted by parent ministries to both organizational forms. As
illustrated above, agencies enjoy more disaggregation, in comparison to government
departments working under ministerial control, hence the gap between ministry and agency in
box 1. But in comparison to MABs, agencies remain under ministerial control, and hence are put
closer to ministerial control on the disaggregation continuum in box 1. In contrast, MABs enjoy
complete disaggregation from ministerial control, and have far more independence than
agencies. In terms of level of disaggregation granted, MABs are put further away from ministerial
control than agencies on the disaggregation continuum in box 2. As per discussion in paras
above, to be effective against opportunities for corruption, SARAs are recommended to be
designed not like agencies or MABs, but in the middle of these organizational forms without
accountability link to parent MoFs, with such oversight replaced by other suitable oversight
bodies. The point is SARAs should have more disaggregation from parent ministries than
available to agencies, but still they should not be free from any sort of democratic control as
MABs, hence the appropriate position of SARAs on the disaggregation continuum is
recommended to be at box 3 as illustrated in Figure 7.1.
This recommendation is aimed at improving, revising and re-forming the existing SARA reform
model, to make it more robust against corruption, by incorporating changes in existing SARA
model as identified by analyses. SARAs are recommended to be more disaggregated from
parent ministries than agencies on account of findings obtained in this thesis which highlighted
not-enough disaggregation from MoFs made SARAs ineffective against corruption. This
recommendation comes with a caveat that SARAs should not be completed disaggregated from
any governmental control such as MABs, hence making them wholly democratically
unaccountable, as it could be very dangerous for the central tax collection agency not being
255
answerable to the parliament in developing countries. Therefore, in order to improve SARAs’
effectiveness against corruption, it is recommended that influence of parent ministries such as
MoFs should be removed from accountability arrangements of SARAs, and replaced by other
suitable oversight bodies. SARAs are recommended to be conceptualized unlike both agencies
and MABs, such that they are more disaggregated from MoFs than agencies, but a little short of
complete disaggregation as in the case of MABs.
7.2.2 A Case for Pakistan Revenue Authority?
The second recommendation is related to findings coming out of micro level of analysis and is
applicable for reforming tax administration in FBR Pakistan. The micro level of analysis was
aimed to explore the reasons behind why tax administration reform failed to effectively control
opportunities for corruption in FBR Pakistan and what should be done about it. In addition, this
analysis also aimed to examine if there is potential and suitability of adopting Anti-Corruption
SARA framework for FBR to control corruption in its tax administration. It is pertinent to conclude
here that findings suggesting how FBR’s accountability mechanisms with oversight bodies were
undermined due to interference by MoF, leading to ineffectiveness in controlling opportunities for
corruption, point towards a case of not keeping FBR directly accountable to MoF, in the presence
of other effective oversight bodies such as parliament select committee in the form of Cabinet
Committee for Federal Revenue. MoF is perceived as so strong in Pakistan that one senior
official of the Planning Commission of Pakistan noted:
“The Finance Ministry, precisely the finance secretary, is running a mini-government”

(Kiani, 2013b, p. 1).
In terms of disaggregation, the status of FBR lies in between a revenue authority and a
centralized government department working under close supervision of the ministry. FBR is
neither a SARA nor a conventional government department. Officially FBR is thought of an

256
organizational arrangement, which after reform has remained under governmental control, but
with significant autonomy in terms of recruitment, compensation, and promotions etc. But as
highlighted by the micro analysis, FBR in its present form, is only autonomous on paper and in
practice remains in the clutches of the parent ministry. It is high time that the ‘confused’ form of
organizational arrangement of FBR, i.e., ‘under governmental control but autonomous’, should be
done away by granting it a more clearer status of a SARA in the form of Pakistan Revenue
Authority, such that it is free from any direct oversight by the MoF, and in contrast should be
subjected to effective oversight by other oversight bodies, such as already existing Cabinet
Committee for Federal Revenue. Moreover, the Cabinet Committee for Federal Revenue, which
has been formed for oversight of FBR, should be free from any influence/control of MoF as well.
There is no point in having a Cabinet Committee for oversight of FBR in place of MoF, if it is still
heavily influenced by MoF.
It is recommended that there is a case for making FBR more autonomous in Pakistan to reduce
political meddling in its affairs, especially by the MoF. As pointed out by the World Bank (2000)
SARAs can be employed by developing countries to react to particular political problems,
especially to protect important activities from conflicting political influences. As highlighted in the
micro analysis, FBR was not robust enough to effectively deal with political influence and
corruption in tax administration. Owing to this reason, there is an argument for considering the
case of appropriateness of SARA reform model for Pakistan in the form of creation of Pakistan
Revenue Authority to make its tax administration fair. It is recommended that there is potential for
adopting Anti-Corruption SARA framework for Pakistan to control corruption in its tax
administration. In other words, in terms of components of agencification (Talbot, 2004) it is
recommended that FBR should be entrusted with more autonomization (by law) and more
disaggregation from MoF. Therefore, it is recommended that there is a case to be considered for
257
the creation of Pakistan Revenue Authority to protect priority activity of collection of taxes from
opposing political influences.
7.2.3 Adoption of SARAs: Balancing Autonomy with Accountability
The third recommendation is related to findings coming out of macro level of analysis and is
applicable to not only existing SARAs but also developing countries contemplating SARA model
to reform tax administration. One of the findings emerging out of macro level of SARA analysis
related to balancing both autonomy and accountability-enhancing SARA design components to
effectively control both motivations and opportunities for corruption. The macro level of SARA
analysis highlighted that SARAs preferred adoption of personnel autonomy over ensuring
effective accountability for controlling corruption. In other words, SARAs preferred adoption of
autonomy-enhancing design components over accountability-enhancing design components for
controlling corruption. This imbalance led to some progress in controlling motivations for
corruption, but failing to effectively control opportunities for corruption. These findings also
highlight the issue of improper implementation of SARAs in developing country context due to
imbalance in ensuring both autonomy and accountability mechanisms. In other words, SARAs
made a half-hearted attempt by utilizing more of autonomy than accountability. As a result when
SARA reforms are criticised for increasing opportunities for corruption, then it is a case where
SARAs has been adopted with more focus on ensuring autonomy, rather also balancing it with
proportionate effective accountability.
It seems as if SARA countries have failed to fully comprehend the importance of balancing
autonomy with accountability. There seems to be lack of understanding that if strong effective
accountability mechanisms are employed then they are not supposed to lessen autonomy in any
way, as both autonomy and accountability mechanisms are supposed to be ensured through
258
separate design features, and hence are not contradictory in nature. This observation is
consistent with both Taliercio (2004) and Schick (1998). As Schick (1998) pointed out, the
underlying principle behind the creation of agencies is a combination of ‘letting managers
manage’ and ‘making managers manage’, which in essence is about granting increased
autonomy in operations in exchange for greater accountability for results. It is concluded that
SARA countries have undergone a paradox of autonomy-accountability mechanisms, where
autonomy features have been adopted more and given preference over accountability features.
As a result SARAs have made only partial progress against corruption with some progress
against controlling motivations, but not opportunities for corruption. It is recommended that SARA
countries should develop this understanding about importance of balancing both autonomy and
accountability mechanisms to be fully effective against corruption by controlling both motivations
and opportunities for corruption. Efforts should be made towards developing this understanding
by SARAs, especially by reformers/donors, of how importance it is to adopt SARA model in its
essence to be fully effective against corruption.
Having outlined some recommendations for making SARAs more effective against corruption, it
is appropriate to recognize that this research, like most researches of this nature, has some
limitations, as discussed in section below.
7.3 LIMITATIONS OF THE STUDY
The macro level of SARA analysis entailed analyzing SARA cases which have not performed well
against corruption. Although quite a few SARA cases fitted this purpose, but were geographically
dispersed all around the globe. Since the researcher lacked resources of time and money to
carry out in-depth case studies of selected SARA cases in pursuit of primary data to analyze
research question, the researcher was left with no other choice but to rely on secondary literature
259
available to conduct macro level of SARA analysis. Also this research, especially the macro level
of SARA analysis, relying heavily on use of secondary literature, is limited by the lack of publicly
available literature. Therefore, the macro level of SARA analysis can only be as good as the
quality and quantity of the secondary sources on which it has relied. As is common in such
research, individuals and corporate organisations alike seldom offer information on how to pay
bribes or engage in corruption (Johnston, 2002).
Some of the issues discussed in this thesis were based on reports of international donor
organizations readily available from the internet. Their acquisition proved quite straightforward.
But same cannot be said about obtaining internal documents and reports of FBR, such as cases
of pending inquiry on account of corruption charges against tax officials. The secrecy and lack of
access to reports and documents in FBR constrained the comprehensive micro level of analysis
examining progress of FBR towards corruption. Therefore, non-availability of such secondary
literature significantly limited the analyses of tax administration reform in FBR. Moreover, the
evidence in this thesis has unavoidably been constrained by what is available in the public
sphere. Although, a detailed analysis of criminal court proceedings might have led to provision of
more information, leading to better understanding of the topic of this thesis, but resources of time
and finances as well as secrecy in the public sectors in Pakistan acted as major constraints in
doing so.
The study conducted interviews with a number of respondents to explore why tax administration
reform failed to effectively control opportunities for corruption in FBR, and gathered the opinions
of interviewees who shared their own personal experiences and perceptions, leading to some
significant findings. In this study limited numbers of respondents were interviewed (31 semi-
structured interviews) owing to problems of access and the duration of fieldwork. Due to this
260
reason, this study does not claim to offer a comprehensive account, rather only providing some
illustrations from a limited number of cases, representing the views of some of the respondents
about the main topic of this thesis. In addition, most interviewees shared their experiences about
corruption but none of the respondents, for obvious reasons, agreed to have personally engaged
in corruption. To solicit views of actors having engaged in corruption could have been obtained
by interviewing those having received punishments for corruption in FBR Pakistan, but such
access has not been possible. Lack of access to this important fragment of participants has
therefore limited this study. Moreover, although this study interviewed 16 tax officials, but
interviews of tax officials of very high level such as Members of the Board and Chairman FBR
could not be conducted on account of access issues. Same is true of interviews of high level
officials in MoF such as Secretary Finance and Minister of Finance. Hence their views are not
represented in this study about why tax administration reform failed to effectively control
opportunities for corruption in FBR and limited the findings of the micro level of analysis.
Nonetheless, an effort was made towards tackling this limitation, by represented their views, as
available in media outlets, in the micro level of analysis by conducting newspaper analysis for an
extended period of three years (2011-2013).
It is hoped that this study examining SARAs’ progress against corruption will generate prospects
for further research into wider issues relating to corruption in tax administration not only in
Pakistan but generally in developing countries. In this setting, the next section discusses the
contributions this study has made to the literature, in addition to proposing avenues for further
research.
261
7.4 CONTRIBUTIONS OF THE THESIS AND AVENUES FOR FURTHER RESEARCH
This thesis has contributed towards filling a gap in literature for research on the practical side of
designing and implementing SARA reform against controlling corruption, by doing a practical
work of conceptualizing an Anti-Corruption SARA Framework (chapter 3) and testing it for SARA
developing countries (Chapter 5) and FBR Pakistan (Chapter 6). This analytical framework was
developed by utilizing two SARA frameworks including SARA design components proposed by
Taliercio (2004) and control of corruption framework for tax administrations by Martinez-Vazquez
et al. (2004). Another important contribution behind developing this framework lies in bringing
together ideas by Taliercio (2004) and Martinez-Vazquez et al. (2004), which have previously
remained separate. The novel contribution of this thesis lies in developing the new lens
(analytical framework) which causes us to see the topic of SARAs against corruption quite
differently (due to differentiating between motivations and opportunities for corruption).
One of the contributions of this research lies in critiquing Taliercio’s (2004) SARA framework.
This critique rested on the ground that Taliercio’s framework might be effective in achieving one
of the objectives behind SARA reform adoption of improvement in revenue performance, but
displays a weakness when applied to achieve control of corruption, by failing to convince
taxpayers about insulation of SARAs from politics, due to persistent accountability link with
politicians in MoFs. It was proposed that politicians in MoFs needed to completely disaggregate
themselves from SARAs’ operations to enable taxpayers to perceive meaningful insulation from
politics. It is stressed that the proposed solution to identified weakness in Taliercio’s framework is
conceptualized to improve SARA reform’s robustness against corruption, in addition to revenue
improvement. Also, another contribution of this research lies in not only analyzing Klitgaard’s
(1988, 1997) corruption framework from a different lens to understand how corruption might be
controlled in centralized tax administrations. But more importantly Klitgaard’s corruption

262
framework was also adapted to make it suitable to analyze how to control opportunities for
corruption in decentralized tax administrations after SARA reforms.
One of the significance of macro level of SARA analysis lies in analyzing relevant secondary
literature sources on SARAs under a new lens and for a different purpose than originally
envisaged. The contribution of the macro level of SARA analysis lies in expanding on previous
research on SARAs by expanding interpretation of previous findings on autonomy and
accountability of SARAs towards controlling motivations and opportunities for corruption in
SARAs. While an effort was made towards an optimum use of secondary literature towards
analyzing the analytical framework in different developing countries, nonetheless it would be
worthwhile to revise and test this framework by applying it in the form of detailed case study in
different SARA developing countries and collect primary data. In addition, this analytical tool can
also be employed by reformers and analysts in developing countries which are contemplating to
adopt SARA reform to improve tax administration. The application of the analytical framework
before adopting reform might highlight the need and suitability of SARA reform model to the
specific developing country context.
The analytical framework in this study concentrated on conceptualizing the effect of SARA design
components of personnel autonomy and effective accountability towards controlling motivations
and opportunities for corruption. The future research can benefit by adapting the analytical
framework for other SARA design components/ types of autonomy such as financial autonomy,
legal autonomy, structural autonomy towards analysing their effect on controlling motivations and
opportunities for corruption.
Also, another important area of future research involves specifically examining SARA success
stories, to analyse what they have done differently to succeed against corruption. Based on the
263
logic of convenience sampling, in terms of quality and quantity of secondary literature available,
this thesis analysed a group of SARA cases, most of which underperformed against corruption. It
would be worthwhile to apply the analytical framework of this thesis on SARA success stories
against corruption, to analyse if they have adopted the SARA reform model in a way experiencing
no interference from MoFs in SARAs operations, which neither undermined their autonomy nor
accountability. This proposed area of research would further enhance the validity of hypotheses
developed in the analytical framework.
7.5 CONCLUSION
The aim of the thesis has been to examine why some SARAs failed to effectively control
opportunities for corruption. To explore this question, this thesis set out an Anti-Corruption SARA
framework and analyzed the validity of the framework by conducting a two-staged analysis of
SARAs in chapters five and six of the thesis.
In chapter five, a comprehensive mapping of existing SARA cases, labelled as the macro level of
SARA analysis, helped to conclude that SARAs have made partial progress to control corruption
by focusing more on controlling motivations for corruption (through autonomy-enhancing design
components) and lesser focus on controlling opportunities for corruption (through accountability-
enhancing design components). Continued interference from MoFs was found to undermine not
only effective accountability for SARAs, but also undermining control of opportunities for
corruption. In other words, SARAs, conceptualized as islands of ‘integrity’ failed to expand in
‘archipelagos’ due to not-enough disaggregation from MoFs affecting corruption control by
undermining both autonomy and accountability. Findings suggesting how SARAs’ accountability
mechanisms were undermined due to interference by MoFs, leading to ineffectiveness in
controlling opportunities for corruption, pointed towards a case of not keeping SARAs directly
264
accountable to MoFs, in the presence of other effective oversight bodies including Revenue
Boards/Superintendents and Parliament etc. The proposed solution to tackle opportunities for
corruption lies in reforming SARAs into an organizational form which are far more disaggregated
from the parent ministry, such that SARAs have no accountability link with MoFs. Furthermore,
the oversight of SARAs by MoFs can be replaced by other suitable oversight bodies, such as
parliament select committees. These findings also highlight the issue of improper implementation
of SARAs in developing country context due to imbalance in ensuring both autonomy and
accountability mechanisms. In other words, SARAs made a half-hearted attempt by utilizing more
of autonomy than accountability. It is concluded that SARA countries have undergone a paradox
of autonomy-accountability mechanisms, where autonomy features have been adopted more and
given preference over accountability features. As a result SARAs have made only partial
progress against corruption with some progress against controlling motivations, but not
opportunities for corruption. It is recommended that SARA countries should develop this
understanding about importance of balancing both autonomy and accountability mechanisms to
be fully effective against corruption by controlling both motivations and opportunities for
corruption.
In chapter six, a detailed case study of tax administration reforms in Pakistan, labelled as the
micro level of analysis, was conducted to explore why tax administration reform failed to
effectively control opportunities for corruption in FBR Pakistan? It was found that although FBR,
like many SARAs, preferred adoption of personnel autonomy over ensuring effective
accountability for controlling corruption. But such preference did not led to even partial progress
against controlling motivations for corruption, and FBR remained ineffective in controlling both
motivations and opportunities for corruption. It was also questioned why FBR was unable to
adopt effective accountability? Similarly to many SARAs, in case of FBR as well, it was found that
265
presence of MoF in the accountability arrangements rendered FBR ineffective in controlling
opportunities for corruption. It is pertinent to conclude here that findings suggesting how FBR’s
accountability mechanisms were undermined due to interference by MoF, leading to
ineffectiveness in controlling opportunities for corruption, point towards a case of not keeping
FBR directly accountable to MoF, in the presence of other effective oversight bodies such as
parliament select committee in the form of Cabinet Committee for Federal Revenue. The
proposed solution lies in converting FBR into a SARA in the form of PRA or Pakistan Revenue
Authority, such that it is free from any direct oversight by the MoF, and in contrast should be
subjected to effective oversight by other oversight bodies, such as already existing Cabinet
Committee for Federal Revenue. Moreover, the Cabinet Committee for Federal Revenue, which
has been formed for oversight of FBR, should be free from any influence/control of MoF as well.
Therefore, it is recommended that there is a case to be considered for the creation of Pakistan
Revenue Authority to protect priority activity of collection of taxes from opposing political
influences.
266
Appendix 1
Semi-Structured Interview Schedule
Research Proposition 2(a) Wages and Rewards

Indicators: Increases in wages for tax officials and utilization of performance-linked
bonuses
1. Has there been any increase in wages and rewards for tax officials in FBR?
2. If yes, do you think increase in wages and rewards have been sufficient enough/effective to
deter tax officials from committing corruption in FBR?
3. Is the wages and reward structure at FBR enough to retain employees with the right skills?
Research Proposition 2(b) Increases in the Probability of Detection

Indicators: Increases in the quality and frequency of internal audits and probes by
establishing internal audit and anti-corruption divisions;
Increases in the number of supervisory personnel assigned to internal audit and anti-
corruption divisions;
Statutory condition of declaration of assets by all employees
4. Has there been any increase in the quality and frequency of internal audits and probes for tax
officials by establishing Directorate General Intelligence & Investigation along with allied regional
Directorates at FBR?
5. If yes, do you think the increase in the quality and frequency of internal audits and probes by
establishing Directorate General Intelligence & Investigation along with allied regional
Directorates at FBR has deterred tax official from committing corruption?
6. Has there been any increase in the number of supervisory personnel assigned to Directorate
General Intelligence & Investigation along with allied regional Directorates at FBR?
7. If yes, do you think the increase in the number of supervisory personnel assigned to
Directorate General Intelligence & Investigation along with allied regional Directorates at FBR has
deterred tax official from committing corruption?
8. Are you aware about reformed system of statutory condition of declaration of assets by tax
officials in FBR?
9. If yes, have you noticed any effectiveness of the reformed system of statutory condition of
declaration of assets by FBR towards deterring tax officials from committing corruption?
Research Proposition 2(c) Increases in and Stricter Enforcement of Penalties for

Corruption
267
Indicators: Increases in imposition of high levels of penalties including monetary

sanctions, job dismissals and prison sentences.
10. Has there been any increase in imposition of high levels of penalties including monetary
sanctions, job dismissals and prison sentences for tax officials in FBR?
11. If yes, do you think increase in imposition of high levels of penalties including monetary
sanctions, job dismissals and prison sentences in FBR has deterred tax official from committing
corruption?
Research Proposition 2(d) Instilling Ethics in Tax Officials

Indicators: Increases in ethics training and adoption of code of conduct.
12. Are you aware about adoption of reformed code of conduct by tax officials in FBR?
13. If yes, do you think adoption of reformed code of conduct by FBR has deterred tax officials
from committing corruption?
14. Has there been any increase in ethics training for tax officials in FBR?
15. If yes, do you think tax officials are less likely/less motivated to commit corruption in FBR as a
result of increased ethics training?
Research Proposition 4(a) Introduction of Oversight Mechanisms

Indicators: Effective accountability relationships with all supervising bodies including
Ministry of Finance, Revenue Boards/Superintendents (whichever applicable), and
Parliament etc.
16. Has FBR taken any steps to introduce effective oversight mechanisms for tax officials?
17. If No, has FBR failed to introduce effective oversight mechanisms for tax officials due to
ineffective ‘accountability relationships with oversight bodies’ due to interference form MoF?
Research Proposition 4(b) De-politicization of Tax Officials

Indicators: Effective accountability relationships with all supervising bodies especially
Revenue Boards/Superintendents (whichever applicable), and Parliament etc., in addition
to Ministry of Finance.
18. Has FBR taken any steps to de-politicize tax officials?
19. If No, has FBR failed to de-politicize tax officials due to ineffective ‘accountability relationships
with oversight bodies’ due to interference form MoF?
Research Proposition 4(c) Reduction of Discretionary Powers

Indicators: Balancing effective accountability relationships between all supervising bodies
especially between Ministry of Finance and other supervising bodies including Revenue
Boards/Superintendents (whichever applicable) and Parliament etc.
268
20. Has FBR taken any steps to reduce discretionary powers of tax officials?
21. If No, has FBR failed to reduce discretionary powers of tax officials due to ineffective
‘accountability relationships with oversight bodies’ due to interference form MoF?
269
Appendix 2
270
271
272
273
274
Appendix 3
GOVERNMENT OF PAKISTAN
CABINET SECRETARIAT
ESTABLISHMENT DIVISION
***
DECLARATION OF INCOME AND ASSETS

Financial Year Ending on 30th June__________
1. Name
NIC No.
N.T.N
o
2. Basic Position Held
Occupational
Group/Service/Dept.
Present Position
Held
3. Present Residential Address
Phone(R) Mobile
Salary Rs. Rental Income Rs. Agri. Income Rs.

4. Income (During the year)
Other sources (dividend, profit, prize money, gift, loan etc.) Total
Rs.
Rs.
5. Expenses Utilities (Electricity, Total House Hold

(Approx) Gas, Telephone etc.) Expenses
Rs. Rs.
6. Private Foreign Traveling Country/Countries Period of stay Approx. expense

(Self, Spouse & Children) visited from ___to___ Rs.
During F.Y.___________ …………………. ………………. ………………….
…………………. ………………. ………………….

275
7. Children’s Education Name(s) of children Educational Institutions

(inland & abroad) attended during F.Y._________
……………………….. ……………………… ………………………….

……………………….. ……………………… ……………………….....
………………………… ……………………… ……………………….....
8. Club Membership Name of Club(s) Membership No.

_______________ ______________ ______________
ASSETS & LIABILITIES

9. Immoveable Assets (Agri. & Non-Agri Lands, House properties, Commercial & Industrial
properties, Open plots of all types)
Identification & nature of Asset(s) Mode of acquisition/year Cost of acquisition

a)
b)
c)
d)
e)
f)
g)
h)
i)
j)
k)
l)
m)
10. Moveable Assets (Cash in hand, Motor vehicles, Jewelry, Household items, Equipment, Business
capital etc.)
Identification & nature of Asset(s) Mode of acquisition/year Cost of acquisition
a)
b)
c)
d)
e)
f)
g)
h)
i)
j)
k)
l)
m)
276
Identification & nature of Asset(s) Date of disposal Amount received as

sale proceed (Rs.)
a)
b)
c)
d)
11. Assts held as Attorney

Identification & nature of Asset(s) Nature of Power of Name & Address of the
Attorney (Recoverable/ Legal Owner.
Irrecoverable)
a)
b)
c)
12. Assets disposed off during year
13. Investments (Bonds, Shares, Certificates, deposits/Advances, Loans granted etc.)
Details of Bonds held Investments

Bond No(s) Denomination Rs. Rs.
a)
b)
c)
d)
14. Bank Accounts (Current, Saving, Deposit A/C & F.C. A/C)
A/c No. & Bank Branch Year of opening Main Source of Balance on 30.6
deposit (Rs)
a)
b)
c)
d)
15. Total Assets (9-14)
16. Liabilities (Departmental/Bank loans, Over drafts, Mortgages secured, private loans etc.)
Outstanding Liabilities (A) Liabilities paid off during the year (B)
a) Rs.
b) Rs.
c) Rs.
d) Rs.
[15-16 (A)] Net worth Rs.___________
As on 30.6___________
Net worth declared previously Rs.___________
277
Signature____________________
Name _____________________
Designation_______________
Name of the Organization/Dept.
________________________
Place____________________
Date_____________________
INSTRUCTIONS
1. If the space provided in the form is found inadequate or some explanation is

required, a separate page may be attached / annexed.
2. All assets should be valued at cost and in the cases of assets acquired through
gift name, address of the donor and donees relationship with him is to be
declared.
3. Income declared at Serial 4 must include income earned by the spouse &
children as well.
4. Information requested must be completed. No column should be left blank.

Columns which are not applicable should be crossed.
5. All assets owned by the officer & his family members (family as defined in
Rule-3 (1) (c ) of conduct Rules, 1964) should be declared. Assets acquired by
major children dependents & others where funds have been provided by the
officer are also to be declared.
6. Assets owned partly or acquired on “Hire purchase Agreement” or installments

should also be declared.
7. If any exact figure cannot be inserted an estimated/approx figure may be given.
8. Sale proceeds of assets disposed off during the relevant Financial Year must be
declared under the head “other sources” (Serial 4).
9. If there is no change in assets over the previous year (for which the declaration
had been filed) relevant column (serial 9, 10 & 11) may be marked “As
Before”.
10. At serial 11 assets held by others as attorney on behalf of declarant, his spouse
or dependent children are also to be declared.
11. Expenses against utilities (Serial 5) should include bills paid against all meters
(Gas & Electricity installed on the residence) and Telephone collections
(including Mobile) in use of the officer, spouse & dependent children.
12. Notwithstanding the applicability of any other law for the time being in – force,
this declaration is being filed under Conduct Rules, 1964 and any breach thereof
278
(including concealment of assets or giving wrong information) is punishable

under RSO, 2000.
…………………………. .
279
Appendix 4
280
281
Appendix 5
The names of Secretaries/Ex-officio Chairmen, full time Chairmen Secretary, Revenue

Division/Chairmen and Vice Chairman, who headed the FBR/ Revenue Division from August 14, 1947
are given below:
Chairmen, FBR
1) Mr. Tariq Bajwa 02.07.2013 Till Date
2) Mr. Ansar Javed 10.04.2013 30.06.2013
3) Mr. Ali Arshad Hakeem 11.07.2012 09.04.2013
4) Mr. Mumtaz Haider Rizvi 13.02.2012 11.07.2012
5) Mr. Mehmood Alam (Member SP&S) 25.01.2012 13.02.2012
Secretary Revenue Division/ Chairman, FBR

1) Mr. Salman Siddique 24.12.2010 21.01.2012
Chairman, FBR
1) Mr. Sohail Ahmad 18.05.2009 24.12.2010
Secretary Revenue Division/ Chairman, FBR

1) Mr. Ahmad Waqar 23.07.2008 18.05.2009
Secretary General Revenue Division/ Chairman, FBR

1) Mr. M. Abdullah Yusuf 14.06.2006 23.07.2008
Secretary Revenue Division/ Chairmen, CBR

1) Mr. M. Abdullah Yusuf 12.03.2004 14.06.2006
2) Mr. Riaz Ahmad Malik 03.07.2001 11.03.2004
3) Mian Iqbal Farid 07.11.1998 06.11.1999
4) Mr. Riaz Hussain Naqvi 08.11.1999 02.07.2001
Vice Chairmen, Chairman CBR

1) Mr. Moinuddin Khan 02.01.1998 06.11.1998
2) Mr. HafeezullahIshaq 11.11.1996 02.01.1998
3) Mr. Shamim Ahmed 28.08.1996 11.11.1996
4) Alvi Abdul Rahim (Remained vice 13.07.1995 28.08.1996
Chairman of CBR before becoming
Chairman CBR)
Secretary Revenue Division/ Chairmen, CBR

1) Mr. A.R. Siddiqi 11.07.1994 11.01.1995
2) Mr. Javed Talat 26.07.1993 01.07.1994
3) Qazi M. Alimullah 03.05.1993 17.07.1993
4) Mr. M. Mubeen Ahsan 03.11.1992 03.05.1993
282
5) Mr. Sajjad Hassan 03.10.1991 03.11.1992
Chairmen, CBR
1) Mr. Sajjad Hassan 24.07.1991 03.10.1991
2) Mr. Ahadullah Akmal 16.08.1990 24.07.1991
3) Mr. Ghulam Yazdani Khan 22.01.1989 11.08.1990
4) Syed Aitezazuddin Ahmed 20.08.1988 02.01.1989
5) Mr. I. A. Imtiazi 11.08.1985 20.08.1988
6) Mr. Fazlur Rahman Khan 14.12.1980 11.08.1985
7) Mr. N. M. Qureshi 12.11.1975 14.12.1980
8) Mr. M. Zulfiqar 01.10.1974 12.11.1975
9) Mr. Riaz Ahmad 17.11.1973 30.09.1974
10) Mr. M. Zulfiqar 11.10.1971 17.11.1973
Finance Secretaries/ Ex-Officio Chairmen, CBR

1) Mr. A.G.N. Kazi 08.09.1970 10.10.1971
2) Mr. Ghulam Ishaq Khan 31.05.1966 08.09.1970
3) Mr. M. M. Ahmed 06.03.1963 30.05.1966
4) Mr. Mumtaz Mirza 19.06.1961 06.03.1963
5) Mr. M. Ayub 29.07.1960 19.06.1961
6) Mr. H. A. Majid 01.11.1958 29.07.1960
7) Mr. Mumtaz Hassan 25.02.1952 01.11.1958
8) Mr. Abdul Qadir 01.02.1950 25.02.1952
9) Sir Victor Turner 14.08.1947 01.02.1950
283
Bibliography
Abed, G. T. And Davoodi, H. (2002) Corruption, Structural Reforms, and Economic Performance
in the Transition Economies, in Abed George T. and Gupta Sanjeev (eds) Governance,
Corruption and Economic Performance, Washington, DC: International Monetary Fund, 487-537.
ADB (2006) Anti-Corruption Policies in Asia and the Pacific. Progress in Legal and Institutional
Reform in 25 Countries. ADB/OECD. Manila : ADB.
ADB (2008) Combating Corruption in Asia-Pacific: Pakistan’s Measures to Implement the Anti-
Corruption Action Plan for Asia-Pacific. Manila: ADB.
ADB (2013) Key Indicators for Asia and the Pacific 2013. Manila: ADB.
Akhter, S. (2013) Rs180 billion in Duties Lost Annually due to SROs Culture. The News, 4 May.
[Accessed 4 May, 2013].
Akindele, S. T. (2005) A Critical Analysis of Corruption and its Problems in Nigeria',

Anthropologist, 7 (1): 7-18.
Ali, K. (2012) Ministers, Lawmakers Evading Tax: Study. DAWN, 13 December. [Accessed 20
December, 2012].
Alm, J. (1998) Tax Compliance and Administration. In Handbook on Taxation, W. B. Hildreth & J.
A. Richardson (eds.). 741-768. New York: Marcel Dekker, Inc.
Amundsen, I. (1999) Political Corruption: An Introduction to the Issues, Chr. Michels Institute,
Development Studies and Human Rights, WP 7, 1- 2.
Andreoni, J., Erard, B., & Feinstein, J. (1998) Tax Compliance. Journal of Economic Literature 36
(2), 818-60.
Archer, M. (1995) Realist Social Theory: The Morphogenetic Approach, Cambridge: Cambridge
University Press.
Archer, M. (1996) Social Integration and System Integration: Developing the Distinction
,SocioIogy, 30 (4): 679- 700.
Arnold, G., Llewellyn, J. and Mao, Q. (1998) International Trends in Public Administration–Notes.
Canberra Bulletin of Public Administration. 88 (May).
Asad, M. (2013) IHC Removes FBR Chairman. DAWN, 6 June. [Accessed 6 June, 2013].
Baker, C. R. and Battner, M. S. (1997) Interpretive and Critical Research in Accounting: A

Commmentary on its Absence from Mainstream Accounting Research, Critical Perspectives on
Accounting, 8: 293-310.
284
Bakre, O. M. (2004) Accounting and the Problematique of Imperialism: Alternative

Methodological Approaches to Empirical Research in Accounting in Developing Countries,
Advances in Public Interest Accounting, Re-Inventing Realities, 10: 1-30.
Bakre, O. M. (2007) The Unethical Practices of Accountants and Auditors and the Compromising
Stance of Professional Bodies in the Corporate World Evidence from Corporate Nigeria,
Accounting Forum, 31(3): 277-303.
Bale, M. and Dale, T. (1998) Public Sector Reform in New Zealand and its Relevance to
Developing Countries, The World Bank Research Observer, 13, 1: 103-121.
Bardhan, P. (1997) Corruption and Development: A review of Issues. Journal of Economic

Literature, 35 (3), 1320-1346.
Batley, R. (1999a) The New Public Management in Developing countries. Journal of International
Development. 11(5), pp. 755-760.
Batley, R. (1999b) The New Public Management in Developing Countries: Implications for Policy
and Organizational Reform. Journal of International Development. 11(5), pp. 761-765.
Bekke, Hans A. G. et al. (1996) Introduction: Conceptualizing Civil Service Systems. In: Bekke
et.al, eds. (1996) Civil Service Systems in Comparative Perspective. Indianapolis: Indiana
University Press.
Bennett, A. and Mills, S. (1998) Government capacity to contract: health sector experience and
lessons, Public Administration and Development, 18: 307–336.
Berger, P. L. and Luckmann, P. (1966) The Social Construction of Reality: A Treatise in the
Sociology of Knowledge, Britian: Penguin Books Ltd.
Bersselaar, D.V. and Decker, S. (2011) “No Longer at Ease”: Corruption as an Institution in West
Africa. International Journal of Public Administration, 34: 741-752.
Bhaskar, R. (1978) A Realist Theory of Science, Sussex: The Harvester Press, London.
Bhaskar, R. (1979) The Possibility of Naturalism: A Philosophical Critique of the Contemporary

Human Science, Sussex: The Harvester Press, London.
Borrins, S. and E. Warrington (1996) The new public administration: global challenges, local
solutions: a report on the second biennial conference of CAPAM, London: Commonwealth
Secretariat.
Brown, L. (1999) Institutional Structures and Capacity-Building for Policy Formulation and
Implementation. Paper presented at a seminar on Restructuring the State: Managing Tensions
between Economic Reform and Social Equity, Barbados, 7–8 October.
Bukhari, H. and Haq, I. (2013) Tax Delinquents and Elections. The News, 31 March. [Accessed 1
April, 2013].
285
Burrell, G. and Morgan, G. (1979) Sociological Paradigms and Organisational Analysis: Elements
of the Corporate Life, London: Ashgate Publishing Ltd.
Burrell, G. and Morgan, G. (2005) Two Dimension, Four Paradigms, in C. C. Lundberg and C. A.
Young (eds) Foundation of Inquiry: Choices and Trade-offs in the Organisational Science,
Standford University Press: 49-53.
Carney, T. (1972) Content Analysis: A Technique for Systematic Inference from
Communications. Winnipeg: University of Manitoba Press.
Chand, S. and Moene, K. (1997) Controlling Fiscal Corruption. WP/97/100. Washington D.C.:
International Monetary Fund.
Chand, S. and Moene, K. (1999) Controlling Fiscal Corruption. World Development. 27(7), pp.
1129-1140.
Chand, S. K. and Moene, K. O. (2002) Controlling Fiscal Corruption, in Abed George T. and
Gupta Sanjeev (eds) Governance, Corruption and Economic Performance, Washington, D C:
International Monetary Fund, 89-110.
Cheema, U. (2012a) Big Taxpayer Aitzaz Raps Lawyers, MPs for not Paying their Taxes. The
News, 18 December. [Accessed 20 December, 2012].
Cheema, U. (2012b) Long Arm of Law finally catches up with a Rogue Officer. The News, 26
December. [Accessed 26 January, 2012].
Chimphamba, Blessings F.S. (1999) Public Health Reform in Malawi: Should Hospitals Have
Autonomy?. MA (Econ) Dissertation, University of Manchester.
Christensen, T. and Laegreid, P. (2001) New Public Management: Undermining Political Control?
In: Christensen, T. and Laegreid, P., eds., (2001) New Public Management: The Transformation
of Ideas and Practice. Ashgate: Aldershot, pp. 93- 119.
Chua, W. F. (1986a) 'Radical developments in Accounting Thought', Accounting Review, LXI (4):
601-632.
Chua, W. F. (1986b) 'Theoretical Construction of and by the Real' A ti Organization and Society,
11(6) : 583-598.
Clark, J. and Wood, D. (2001) New Public Management and Development: The Case of Public
Service Reform in Tanzania and Uganda. In: McCourt, W. and Minogue, M., eds., (2001) The
Internationalization of Public Management: Reinventing the Third World State. Cheltenham and
Northampton: Edward Elgar, pp. 70-89.
Cohen,I. J. (1998) Anthony Giddens, in R. Stones (ed) Key Sociological Thinkers, London:
Macmillan, 279-290.
Common, R. (1999). Accounting for Administrative Change in Three Asia– Pacific States: The
Utility of Policy Transfer Analysis. Public Management 1(3): 429–38.
286
Commonwealth Secretariat (1996) Current Good Practices and New Developments in Public
Service Management. London: Commonwealth Secretariat.
Conrad, L. (2005) A Structuration Analysis of Accounting Systems and Systems of Accountability

in the Privatised Gas Industry, Critical Perspectives on Accounting, 16 (1): 1-26.
Da Silva, M. (1999) TQM in the Brazilian public sector. In Ho, S. (ed.) TQM and Innovation:
Proceedings of the Fourth International Conference. School of Business: Hong Kong Baptist
University.
David A. Nadler and Michael Tushman (1989) Organization, Congruence and Effectiveness: A
Model for Diagnosing Organizational Behavior: Applying a Congruence Perspective. The
Management of Organizations: Strategies, Tactics and Analyses. New York: Harper & Row
Publishers.
DAWN (2010) Two Customs Officials Convicted in Graft Cases. 27 February. [Accessed 2
March, 2010].
DAWN (2011a) Tax Fraud. 7 February. [Accessed 11 February, 2011].
DAWN (2011b) Police, Customs Officials Convicted in Graft Case. 24 February. [Accessed 24
February, 2011].
DAWN (2013) SC Suspends Tax Official’s Sentence. 2 February. [Accessed 4 February, 2013].
De Merode, L. and Charles, S. (1994) Implementing Civil Service Pay and Employment Reform
in Africa: The Experiences of Ghana, the Gambia, and Guinea. In: Lindauer, D. and B. Nunberg,
B., eds., (1994) Rehabilitating Government: Pay and Employment Reform in Africa. Washington,
D.C.: World Bank, pp. 160–94.
de Soto, H. (2000) The Mystery of Capital: Why Capitalism Triumphs in the West and Fails
Everywhere Else. New York: Basic Books.
Delay, S., Devas, N., and Hubbard, M. (1998) The Reform of Revenue Administration. DFID and
Development Administration Group: University of Birmingham.
DeLeon, L. (1997) Administrative Reform and Democratic Accountability. In: Kickert, W.J.M., ed.,
(1997) Public Management and Administrative Reform in Western Europe. Cheltenham and
Northampton: Edward Elgar, pp. 237–252.
Devas, N., Delay, S., and Hubbard, M. (2001) Revenue Authorities: Are they the Right Vehicle for
Improved Tax Administration? Public Administration and Development. 21(3), pp. 211-222.
Di Tella, R., and Schargrodsky, E. (2003) The Role of Wages and Auditing during a Crackdown
on Corruption in the City of Buenos Aires. Journal of Law and Economics. 46 (1), pp. 269-292.
Dodoo, R. (1997) Performance Standards and Measuring Performance in Ghana. Public

Administration and Development 17(1): 115–21.
287
Doig, A. (1997) The Privatisation of the Property Services Agency: Risk and Vulnerability in
Contract-Related Fraud and Corruption. Public Policy and Administration. 12(3), pp. 6–27.
Doig, A., & Theobald, R. (2000) Corruption and Democratization. London: Frank Cass.
Dunleavy, P. and Hood, C. (1994) From Old Public Administration to New Public Management.
Public Money & Management. 14(3), pp. 9-16.
Dunleavy, P. and O'leary, B. (1987) Theory of the State, the Politics of Liberal Democracy,
London: the Macmillan Press Ltd.
Durand, F. and Thorp, R. (1998) Reforming the State: A Study of the Peruvian Tax Reform.
Oxford Development Studies. 26(2), pp.133-151.
Easterby-Smith, M., Thorpe, R. and Jackson, P. R. (2008) Management Research, London: Sage
Publications.
Easterby-Smith, M., Thorpe, R. and Lowe, A. (2003) Management Research: An Introduction,

London: Sage Publications.
Eldridge, D. and McCourt, W. (1998) Human Resource Management and Development. London:
School of Oriental and African Studies.
FBR (2008) A Review of Resource Mobilization Efforts of Federal Board of Revenue. FBR
Quarterly Review. 7(4), pp. 30-40.
FBR (2013) About Directorate General, Intelligence & Investigation-FBR. Available online at:
http://www.cbr.gov.pk/ShowArticle.aspx?actionID=572.
Ferlie, E. and Fitzgerald, L. (2000) The Sustainability of the New Public Management in the UK:
An Institutionalist Perspective. Symposium Paper for the Annual Conference of the American
Academy of Management. Toronto: AoM.
Ferlie, E., Ashburner, L., Fitzgerald, L., and Pettigrew, A. (1996) The New Public Management in
Action. Oxford: Oxford University Press.
Financial Times (2011) Pakistan ends IMF Loan Deal amid Dispute. 31 October. [Accessed 26
January, 2013].
Fjeldstad, O. (2002) Fighting Fiscal Corruption: The Case of the Tanzania Revenue Authority.
CMI Working Paper 2002. Bergen: Chr. Michelsen Institute.
Fjeldstad, O. (2003) Fighting Fiscal Corruption: Lessons from the Tanzania Revenue Authority.
Public Administration and Development. 23(2), pp.165–175.
Fjeldstad, O. (2005a) Corruption in Tax administration: Lessons from Institutional Reforms in

Uganda. CMI Working Paper 2005:10. Bergen: Chr. Michelsen Institute.
Fjeldstad, O. (2005b) Revenue Administration and Corruption. U4 Issue 2:2005. Bergen: Chr.
Michelsen Institute.
288
Fjeldstad, O. and Moore, M. (2009) Revenue Authorities and Public Authority in Sub-Saharan
Africa. Journal of Modern African Studies. 47 (1), pp. 1-18.
Fjeldstad, O. and Ranker, L. (2003) Taxation and Tax Reforms in Developing Countries:
Illustrations from Sub-Saharan Africa. Bergen: Chr. Michelsen Institute.
Fjeldstad, O., and Tungodden, B. (2003) Fiscal Corruption: A Vice or a Virtue? World
Development. 31 (8), pp. 1459-1467.
Fjeldstad, O., Kolstad, I., and Lange, S. (2003) Autonomy, Incentives and Patronage: A Study of
Corruption in the Tanzania and Uganda Revenue Authorities. Bergen: Chr. Michelsen Institute.
Fleischman, R. K., Kalbers, L. P. and Parker, L. D. (1996) Expanding the Dialogue: Industrial
Revolution Costing Historiography, Critical Perspectives on Accounting, 7 (3): 315-337.
Friedman, E., Johnson, S., Kaufmann, D., & Zoido-Lobaton, P. (2000). Dodging in the Grabbing
Hand: The Determinants of Unofficial Activity in 69 Countries. Journal of Public Economics, 77
(3), 459-493.
Ghumman, K. (2010) FBR lets its Official off in Rs140m Tax Refund Case. DAWN, 13 February.
[Accessed 15 February, 2010].
Ghura, D. (2002) Tax Revenue in Sub-Saharan Africa: Effects of Economic Policies and
Corruption, in Abed George T. and Gupta Sanjeev (eds. Governance, Corruption and Economic
Performance, Washington, DC: International Monetary Fund, 369-395.
Ghura, D., and Katz, M. (1998) Tax Revenue in Sub-Saharan Africa: Effects of Economics
Policies and Corruption. IMF Working Paper WP/98/135. Washington D.C.: International
Monetary Fund.
Giddens, A. (1977) Studies in Social and Political Theory, London: Hutchinson.
Giddens, A. (1984) The Constitution of Society: Outline of the Theory of Structuration,

Cambridge: Polity Press.
Giles, D. E., and Caragata, P. J. (1999). The Learning Path of the Hidden Economy: The Tax
Burden and Tax Evasion in New Zealand. Reference Number EC/96461.
Gill, Jit B.S. (2000) A Diagnostic Framework for Revenue Administration. World Bank Technical
Paper No. 472. World Babk: Washington, D.C.
Greenaway, J. (1995) Having the Bun and the Halfpenny: Can Old Public Service Ethics Survive
in the New Whitehall? Public Administration. 73(3), pp. 357–374.
Grindle, Merilee S. (1997). Divergent Cultures? When Public Organizations Perform Well in
Developing Countries. World Development 25(4): 481–95.
289
Guba, E. G. and Lincoln, Y. S. (1994) 'Competing Paradigms in Qualitative Research', in N. K.

Denzi and Y. S. Lincoln (eds) Handbook ofQualitative Research, Sage, Thousand Oaks, CA:
105-117.
Guess, G. (1989) The Relevance of Comparative International Administration to Management

and Policy Issues. International Journal of Public Administration. 12(3)
Gupta S., de Mello, L., & Sharan, R. (2001). Corruption and military spending, European Journal
of Political Economy 17 (4), 749-777.
Gupta, S., Davoodi, Rand Tiongson, E. R. (2002) Corruption and the Provision of Health Care
and Education Services, in Abed George T. and Gupta S. (eds) Governance, Corruption and
Economic Performance, Washington, D. C.: International Monetary Fund, 245-279.
Gupta, S., de Mello, L. and Sharan, R. (2000) Corruption and Military Spending, IMF Working
Paper 00/23, Washington, DC: International Monetary Fund.
Hakim, C. (1993) Research Analysis of Administrative Records. In: Hammersley, M., ed., (1993)
Social Research: Philosophy, Politics and Practice. London: Sage, pp. 131-45.
Haldenwang, C. (2010) Taxation, Fiscal Decentralisation and Legitimacy: The Role of Semi-
Autonomous Tax Agencies in Peru. Development Policy Review. 28 (6), pp. 643-667.
Hall, J. M. and Jenkins, G. P. (1995) The Transition to a Revenue Authority. International Tax
Program: Harvard Institute for International Development.
Hall, P. A. and Soskice, D. (2001) An Introduction to Varieties of Capitalism', in P. A. Hall and D.

Soskice (eds) Varieties of Capitalism: The Institutional foundations of Comparative Advantage,
New York: Oxford University Press, 1-68.
Halligan, J., ed., (2003) Civil Service Systems in Anglo-American Countries. Cheltenham and
Massachusetts: Edward Elgar.
Haltiwanger, J. and Singh, M. (1999) Cross-Country Evidence on Public Sector Retrenchment.

The World Bank Economic Review. 13 (1), pp. 23–66.
Haque, M. S. (2000) Significance of Accountability under the New Approach to Public

Governance. International Review of Administrative Sciences. 66(4), pp. 599–617.
Hlophe, D. and Friedman, S. (2002) “’…And Their Hearts and Minds Will Follow…’ Tax
Collection, Authority and Legitimacy in Democratic South Africa.” IDS Bulletin. 33(3), pp. 67-76.
Hood, C (1991) A Public Management for All Seasons? Public Administration. 69 (1), pp. 3-19.
Hood, C. (1989) Public Administration and Public Policy: Intellectual Challenges for the 1990s.
Australian Journal of Public Administration. 48 (4)
Hood, C., Scott, C., James, O., Jones, G. and Travers, T. (1999) Regulation inside Government:
Waste-Watchers, Quality Police, and Sleazebusters. Oxford: Oxford University Press.
290
Hoogenboom, A. (1982) Outlawing the Spoils: A History of the Civil Service Reform Movement,
1865-1883. Westport: Greenwood Press.
Hopper, T. M. and Powell, A. (1985) Making Sense of Research into the Organisation and Social
Aspects of Management Accounting: A Review of Its Underlying Assumptions, Journal of
Management Studies, 22(5): 429-465.
Hoque, Z. (2006) Methodological Issues in Accounting Research: Theories, Method and Issues,
Spiramus.
Hors, I. (2001) Fighting Corruption in Customs Administration: What can We Learn from Recent
Experiences? OECD Development Centre Working Papers No. 175. OECD.
House of Commons (2013) Pakistan: Tenth Report of Session 2012–13. International

Development Committee. London: House of Commons.
Huang, H. Shan-Jin, W. (2003). Monetary policies for Developing Countries: The Role of
Corruption. NBER Working Papers, WP#10093. Cambridge, MA: NBER.
Humphrey, S., Charles, C., Economou, D. & Drury, H., (2015) Peeling the onion – A textual
model of critical analysis, Journal of English for Academic Purposes, 17: 37–50.
Huntington S. P.(1968). Political order in changing societies. New Haven: Yale U. Press.
Huther, J., & Shah, A. (2000). Anti-corruption Policies and Programs: A Framework for
Evaluation. Policy Review Working Paper 2501, World bank, Washington, DC.
IMF (2013) Pakistan Article IV Consultation and Request for an Extended Arrangement under the
Extended Fund Facility. IMF Country Report No. 13/287. Washington D.C.: IMF.
Iqbal, A. (2010) Hillary urges Rich Pakistanis to Pay more Tax for Self-Reliance. DAWN, 26
February. [Accessed 26 February, 2010].
Islam, N. (1993) Public Enterprise Reform: Managerial Autonomy, Accountability and

Performance Contracts. Public Administration and Development. 13(2), pp. 129-152.
Jackson, M C. (2000) Systems Approaches to Management. New York KIuwer Academic/Plenum

Publishers.
Jamal, N. (2011) The Yawning Tax Gap. DAWN, 14 February. [Accessed 14 February, 2011].
Jang (2011) Commissioner Income Tax Sargodha Imprisoned for One Year with Fine of Rs. 10
million. 22 February. [Accessed 22 February, 2011].
Jenkins, G. (1995) An Autonomous Revenue Authority for South Africa. International Tax
Program: Harvard University.
Jenkins, G. P. (1994) Modernization of Tax Administrations: Revenue Boards and Privatization

as Instrument for Change. International Bureau of Fiscal Documentation. 48 (2), pp. 75- 81.
291
Jenkins, G.P. and Khadka, R. (2000) Modernization of Tax Administration in Low-Income

Countries: The Case of Nepal. CAER II Discussion Paper 68. Massachusetts: USAID.
Johnson, S., Kaufmann, D., McMillan, D., & Woodruff, C. (1999). Why Do Firms Hide? Bribes
and unofficial Activity after Communism. (Preliminary Version).
Johnson, Simon, Daniel Kaufmann, and Pablo Zoido-Lobaton (1998) Regulatory Discretion and
the Unofficial Economy. American Economic review 88 (2): 387- 92.
Johnston, M. (2002) Measuring the New Corruption Ranking: Implications for Analysis and
Reform. In: Heidenheimei, A. and Johnston, M. eds., (2002) Political Corruption: Concepts and
Contexts. New Jersey: Transaction Publishers, pp. 865-884.
Johnston, M. (2005) Civil Society and Corruption Mobilising for Reforms University Press of
America.
Jupp, V. and Norris, C. (1993) Traditions in Documentary Analysis. In: Hammersley, M., ed.,
(1993) Social Research: Philosophy, Politics and Practice. London: Sage, pp. 37-51.
Kaufmann, (1997). Corruption: The Facts. Foreign Policy, Summer (1997).
Kaufmann, D., Kraay, A., & Mastruzzi, M. (2003). Governance Matters III. World Bank Policy
Research Working Paper 3106.
Kemal, A. (2013) Why Pay Taxes. The News, 1 April. [Accessed 1 April, 2013].
Kernaghan, K. (1992) Empowerment and Public Administration: Revolutionary Advance or

Passing Fancy? Canadian Public Administration. 35(2), pp. 194–214.
Khan, M. (2011) Officials’ Transfer to Hurt FBR Performance. DAWN, 29 January. [Accessed 10
February, 2011]
Khan, M. (2012a) NADRA’s Disclosure. Over 2m Rich Pay No Taxes. DAWN, 3 October.
[Accessed 3 October, 2012].
Khan, M. (2012b) Many Leading Cricketers don’t File Tax Returns. DAWN, 8 October. [Accessed
8 October, 2012].
Khan, M. (2013) Another Attempt to Reform Tax Administration. DAWN, 3 April. [Accessed 3
April, 2013].
Kiani, K. (2010) World Bank, Govt in $300m Dispute over PRSC Programme. DAWN, 20
Kiani, K. (2013a) Damaging Discretionary Powers. DAWN, 18 February. [Accessed 22 February,

2013].
Kiani, K. (2013b) Finance Ministry on a ‘Solo Flight’. DAWN, 6 May. [Accessed 6 May, 2013].
Kidd, M. and Crandall, W. (2006) Revenue Authorities: Issues and Problems in Evaluating their
Success. IMF Working Paper WP/06/240. Washington D.C.: IMF.
292
Klingner, D. (2000) South of the border: Progress and problems in implementing New Public
Management reforms in Mexico today, paper presented at symposium on New Public
Management in Mexico, http://www.fiu.edu.
Klitgaard, R. (1988) Controlling Corruption. Berkeley and Los Angeles: University of California
Press.
Klitgaard, R. (1997) Cleaning Up and Invigorating the Civil Service. Public Administration and
Development. 17(5), pp. 487–509.
Klitgaard, Robert (1997a) Unanticipated Consequences in Anti-Poverty Programs. World

Development 25(12): 1963–72.
Kpundeh, S. (1997) Combating Corruption: An Assessment with International Cases. Washington

D.C.: Management Systems International.
Langseth, P. and J. Pope (1999) Building integrity to fight corruption: learning by doing,
Washington, DC: World Bank, EDI Working Paper.
Langseth, P., Stapenhurst, R., & Pope, J. (1997). The Role of a National Integrity System in
Fighting Corruption. EDI Working Papers. The World Bank, Washington D.C.
Langseth, Petter (1995). Civil Service Reform in Uganda: Lessons Learned. Public Administration
and Development 15: 365–90.
Larbi, G. (1999) Management Decentralization in Practice: A Comparison of Public Health and

Water Services in Ghana. In: Minogue, M., Hulme, D. and Polidano, C., eds., (1999) Beyond the
New Public Management: Changing Ideas and Practices in Governance. Cheltenham and
Laughlin, R. (1995) Empirical Research in Accounting: Alternative Approaches and a Case for
'Middle-Range Thinking', Accounting, Auditing and Accountability Journal, 8 (1): 63-87.
Layder, D. (1987) Key Issues in Structuration Theory: Some Critical Remarks, Current
Perspectives in Social Theory, 8: 2-46.
Lee, A. S. (1991) Integrating Positivist and Interpretive Approaches to Organisational Research,

Organisation Science, 2 (4): 342-365.
Leff, N. H. (1964) Economic Development through Bureaucratic Corruption. The American

Behavioral Scientist, 8,(2), 8-14.
Lindauer, D. and B. Nunberg (eds.) (1994) Rehabilitating Government: Pay and Employment
Reforms in Africa, Washington DC: World Bank.
Livingstone, Ian, and Roger Charlton (1998). Raising Local Authority District Revenues Through
Direct Taxation in a Low-Income Developing Country: Evaluating Uganda’s GPT. Public
Administration and Development 18(5): 499– 517.
293
Macintosh, N. B. and Scapens, R. W. (1990) Structuration Theory in Management Accounting,

Accounting, Organisations and Society, 15(5): 455-477.
Macintosh, N. B. and Scapens, R. W. (1996) Structures and Agency in Management Accounting:

A Response to Boland's Interpretive Accounting, Accounting, Organisations and Society, 21(7/8):
675-690.
Magee, B. (1985) Popper, 3rd edn., London: Fontana.
Makumbe, J.M. (1997) The Zimbabwe Civil Service Reform Programme: A Critical Perspective,
Role of Government in Adjusting Economies Paper No. 16. Birmingham: Development
Administration Group.
Mallon, R. (1994) State-owned enterprise reform through performance contracts: the Bolivian
experience, World Development, 22: 925–34.
Manasan, R.G. (2003) Tax Administration Reform: Semi-Autonomous Revenue Authority,

Anyone? Philippine Journal of Development. 56(2).
Mann, A. J. (2004) Are Semi-Autonomous Revenue Authorities the Answer to Tax Administration
Problems in Developing Countries? – A Practical Guide. Washington, D.C.: USAID.
Manzetti, L., & Drake, C. (1906). Market Reforms and Corruption in Latin America: New Means
for Old Ways. Review of International Political Economy. 3 (4), 662-697.
Martinez-Vazquez, J., Arze, F. J. and Boex, J. (2004) Corruption, Fiscal Policy, and Fiscal
Management. Washington, D.C.: USAID.
Marx, K. (1926) The Eighteenth Brumaire of Louis Banaparte by Karl Marx Translated from
German by Eden and Ceder Paul, London: G. Allen and Uniwin Ltd.
Matheson, A. (2000) New Public Management: How well does it travel?’, Rethinking Approaches
to Government Reforms. Seminar Series, No.1. Oxford: Oxford Policy Institute.
Mauro, P. (1995) Corruption and Growth', Quarterly Journal of Economics, 110 (3): 681-712.
Mauro, P. (1995). Corruption and Growth. The Quarterly Journal of Economics, 10 (3), 681-712.
Mauro, P. (1998) Corruption and Composition of Government Expenditure, Journal of Public

Economics, 69: 263-279.
Mauro, P. (1998). Corruption and the Composition of Government Expenditure. Journal of Public
Economics, 69, 263-279.
Mauro, P. (2004) The Persistence of Corruption and Slow Economic Growth, IMF Staff Paper, 51
(1): 1-18.
Mayville, W. (2005) Codes of Ethics and Conduct in Revenue Administrations: What does
International Practice Tell Us. World Bank Tax Policy and Tax Administration Thematic Group:
World Bank.
294
McCarten, W. (2006) Focusing on the Few: The Role of Large Taxpayer Units in the Revenue
Strategies of Developing Countries. In: Alm, J., Martinez-Vazquez, J. and Rider, M., eds., (2006)
The Challenges of Tax Reform in a Global Economy. New York: Springer.
McCourt, W. (1998) The New Public Selection? Competing Approaches to the Development of
the Public Service Commission of Nepal. Public Policy and Management Working Paper No. 8.
Manchester: University of Manchester.
McCourt, W. (2001a) The NPM Agenda for Service Delivery: A Suitable Model for Developing
Countries? In: McCourt, W. and Minogue, M., eds., (2001) The Internationalization of Public
Management: Reinventing the Third World State. Cheltenham and Northampton: Edward Elgar,
pp. 107-128.
McCourt, W. (2001b) The New Public Selection? Anti-corruption, Psychometric Selection and the
New Public Management in Nepal. Public Management Review. 3(3), pp. 325–343.
McCourt, W. (2001c) Moving the Public management Debate Forward: A Contingency Approach.
In: McCourt, W. and Minogue, M., eds., (2001) The Internationalization of Public Management:
Reinventing the Third World State. Cheltenham and Northampton: Edward Elgar, pp. 220-253.
McCourt, W. (2002) New Public Management in Developing Countries. In: McLaughlin, K.,
Osborne, S.P., and Ferlie, E., eds., (2002) New Public Management: Current Trends and Future
Prospects. London: Routledge, pp. 227-242.
McCourt, W. (2006) Sri Lanka: Political Patronage. In: McCourt, W., ed., (2006) The Human
Factor in Governance: Managing Public Employees in Africa and Asia. Hampshire and New York:
Palgrave Macmillan, pp. 135-157.
McCourt, W. (2008) Public Management in Developing Countries', Public Management

Review,10:4,467-479.
McCourt, W. and Minogue, M., eds., (2001) The Internationalization of Public Management:
Reinventing the Third World State. Cheltenham and Northampton: Edward Elgar.
McCourt, W. and Ramgutty-Wong, A. (2000) Strategic Human Resource Management in the

Mauritian civil service, Manchester: Institute for Development Policy and Management,
unpublished manuscript.
McCourt, W. and Sola, N. (1999) Using training to promote civil service reform: a Tanzanian local
government case study’, Public Administration and Development, 19: 63–75.
McEvoy, P. and Richards, D. (2006) A Critical Realist Rationale for Using a Combination of
Quantitative and Qualitative Methods. Journal of Research in Nursing. 11(1), pp. 66-78.
McLaughlin, K., Osborne, S.P. and Ferlie, E, (2002) New Public Management: Current Trends
and Future Prospects. London: Routledge.
Mello, L and Barenstein, M. (2002)'Fiscal Decentralisation and Governance: A Cross Country

Analysis, in Abed George T. and Gupta Sanjeev (eds) Governance, Corruption and Economic
Performance, Washington, DC: International Monetary Fund, 333-367.
295
Meyers, R. and Lacey, R. (1996) Customer satisfaction, performance and accountability in the
public sector, International Review of Administrative Sciences, 62: 331–350.
Miles, M. B. and Huberman, A. M. (1994) An Expanded Sourcebook: Qualitative Data Analysis.

London: Sage Publications.
Mills, Anne (1997) Improving the Efficiency of Public Sector Health Services in Developing
Countries: Bureaucratic versus Market Approaches, pp. 245–74 in Christopher Colclough, ed.,
Marketizing Education and Health in Developing Countries: Miracle or Mirage? Oxford:
Clarendon.
Minh Le, T. (2007) Combating corruption in revenue administration: An overview, pp. 335-338 in
J.E. Campos and S. Pradhan (eds.) The many faces of corruption: Tracking vulnerabilities at the
sector level. Washington, D.C: World Bank.
Minogue, M. (1999) Changing the State: Concepts and Practice in the Reform of the Public
Sector. In: Minogue, M., Hulme, D. and Polidano, C., eds., (1999) Beyond the New Public
Management: Changing Ideas and Practices in Governance. Cheltenham and Northampton:
Edward Elgar, pp. 17-37.
Minogue, M. (2001) The Internationalization of New Public Management. In: McCourt, W. and
Minogue, M., eds., (2001) The Internationalization of Public Management: Reinventing the Third
World State. Cheltenham and Northampton: Edward Elgar, pp. 1-19.
Mitchell, A., Sikka, P. and Willmott, H. (1998) Sweeping It Under the Carpet: The Role of
Accountancy Firms In Money Laundering, Accounting, Organisation and Society, 23 (5/6): 589-
607.
Mitchell, A.and Sikka, P.(1993) Accounting for Change: The Institutions of Accounting, Critical
Perspectives on Accounting, 4 (1): 29-52.
Mkandawire, T. and Soludo, C. C. (1999) Our Continent, Our Future: African Perspectives on
Structural Adjustment, Africa World Press, Online. Available at:
http://www.idrc.ca/openebooks/855-4/#page 18 (accessed 13 June 2010).
Mollel, R.H. (1998) Constraints and Challenges in Establishing Executive Agencies in Tanzania:
The Lessons of Experience. MSc dissertation, University of Manchester.
Mookherjee, D. (1997) Incentive Reforms in Developing Country Bureaucracies: Lessons from

Tax Administration. Paper prepared for the Annual Bank Conference on Developing Economics.
World Bank: Washington D.C.
Morgan, G. (1983) Research as Engagement: A personal View, in G. Morgan (ed.) Beyond

Method: Strategies for Social Research, London: Sage: Publications.
Morgan, G. and Smircich, L. (2005) The Range of Core Assumptions, in C. C. Laundberg and C.
A. Young (eds.) Foundation of Inquiry: Choices and Trade-offs in Organisational Science,
Stanford University Press: 99-103.
296
Morris, T. and Wood, S. (1991) Testing the Survey Method: Continuity and Change in British
Industrial Relations. Work Employment and Society. 5 (2), pp. 259-282.
Mouzelis, N. (2000) The Subjective-Objective Divide: Against Transcendence, Sociology, 34 (4):

741-762.
Mustafa, Z. (2013) Tax Reform, not Aid Please. DAWN, 10 April. [Accessed 10 April, 2013].
Njoku, O. N. (2004) Era of Misguided Good Feeling: Yakubu Gowon and the Birth of New
Nigeria, 1970-1975, in L. A. Nwachuku and G. N. Uzoigwu (eds) Troubled Journey Nigeria Since
the Civil War, University Press of America Inc., 51-77.
Norton, et al., (2004) Casuistry and Social Category Bias. Journal of Personality and Social
Psychology. 87, pp. 817-831.
Nunberg, B. (1995) Managing the Civil Service: Reform Lessons from Advanced Industrialized
Countries. World Bank Discussion Paper 204. Washington, D.C.: World Bank.
O’Donnell, G. (1999) Horizontal Accountability in New Democracies. In: Schedler, A., Diamond,
L. and Plattner, M.F., eds., (1999) The Self-Restraining State: Power and Accountability in New
Democracies. Colorado and London: Lynne Rienner, pp. 29- 52.
Odekunle, F. (ed.) (1986) Nigeria: Corruption in Development, Ibadan: University Press, Nigeria.
OECD (1995) Governance in Transition: Public Management Reforms in OECD countries, Paris:
OECD.
Ofosu-Amaah, W., Soopramanien, R., and Uprety, K. (1999) A Comparative Review of Selected
Legal Aspects of State Practice and Major International Initiatives. Washington, D.C.: World
Bank.
Osoba, S. O. (1996) Corruption in Nigeria: Historical Perspectives, Review of African Political

Economy, 69: 371-386.
Osoro, N., Mpabngo, P., and Mwinyimvua, H. (1999) Enhancing Transparency in tax
Administration in Tanzania. University of Dar-es-Salaam: Department of Economics.
Parry, Taryn Rounds (1997) Achieving Balance in Decentralization: A Case Study of Education
Decentralization in Chile. World Development 25(2): 211–25.
Patton, M. Q. (1990) Qualitative Evaluation and Research Methods. Newbury Park: Sage
Publications.
Paul, S. and Sekhar, S. (1997) A report card on public services: a comparative analysis of five
cities in India, Regional Development Dialogue, 18, 2: 119–35.
Persson, T. and Tabellini, G., eds., (1994) Monetary and Fiscal Policy: Credible Commitment
(Volume 1). Massachusetts: The MIT Press.
Peters, B. G. (1996) The Future of Governing. Kansas: University Press of Kansas.

297
Pettigrew, A. M., Ferlie, E. and McKee, L. (1992) Shaping Strategic Change - Making Change in
Large Organizations: The Case of the National Health Service. London: Sage.
Pettigrew, A.M. (1990) Longitudinal Field Research on Change: Theory and Practice.
Organization Science. 1(3), pp. 267-92.
Polidano, C. (1999b). The New Public Management in Developing Countries. IDPM Public Policy
and Management Working Paper No. 13: University of Manchester.
Polidano, C. (2001) Administrative Reform in Core Civil Services: Application and Applicability of
the New Public Management. In: McCourt, W. and Minogue, M., eds., (2001) The
Internationalization of Public Management: Reinventing the Third World State. Cheltenham and
Polidano, C., Hulme, D. and Minogue, M. (1999a) Conclusions: Looking Beyond the New Public
Management. In: Minogue, M., Hulme, D. and Polidano, C., eds., (1999) Beyond the New Public
Management: Changing Ideas and Practices in Governance. Cheltenham and Northampton:
Edward Elgar, pp. 278-294.
Pollitt, C. (1995) Justification by Works or by Faith?: Evaluating the New Public Management.
Evaluation. 1(2), pp. 133-154.
Pollitt, C. and Talbot, C., eds., (2004) Unbundled Government: A Critical Analysis of the Global
Trend to Agencies, Quangos and Contractualisation. London and New York: Routledge.
Pollitt, C., Talbot, C., Caulfield, J., Smullen, A. (2004) Agencies: How Governments Do Things
Through Semi-Autonomous Organizations. New York: Palgrave Macmillan.
Pope, J. (1995) Ethics, Transparency and Accountability: Putting Theory into Practice. In:
Langseth, P., Nogxina, S., Prinsloo, D., and Sullivan, R., eds., (1995) Civil Service Reform in
Anglophone Africa. Pretoria: Economic Development Institute, Overseas Development
Administration, and Government of South Africa.
Purohit, M.C. (2007) Corruption in Tax Administration. In: Shah, A., ed., (2007) Performance
Accountability and Combating Corruption. Washington, D.C.: World Bank, pp. 285- 302.
Qazi, R.S. (2008). Tax Administration Reforms Project. Available at

http://www.fbr.gov.pk/reforms/ Artical.asp
Rana, P. (2013a) Worn-out Approach to Tax Collection. DAWN, 1 April. [Accessed 1 April, 2013].
Rana, P. (2013b) FBR Promoting Institutionalised Corruption. DAWN, 16 January. [Accessed 28

January, 2013].
Raza, I. (2012) NAB Chief Raises Graft Estimate to Rs13bn. DAWN, 13 December. [Accessed
13 December, 2012].
Riege, A. M. (2003) Validity and Reliability Test in Case Study: A Literature Review with ‘Hands-
on’ Application for each research phase, Qualitative Market Research' A International Journal,
6(2): 75-86.
298
Robert, J. and Scapens, .R. W. (1985) Accounting Systems and Systems of Accuntability
Understanding Accounting Practices in their Organisational Contexts. Accounting, Organisations
and Society, 10(4): 443-456.
Robson, C. (2002) Real World Research. Oxford: Blackwell.
Rose-Ackerman, S. (1999) Corruption and Government: Causes, Consequences, and Reform.

Cambridge: Cambridge University Press.
Russell, Steven, Sara Bennett and Anne Mills (1999). Reforming the Health Sector: Towards a
Healthy New Public Management. Journal of International Development 11(5): 767–75.
Samaratunge, R., Alam, Q. and Teicher, J. (2008) Public Sector Reforms and Accountability: The
Case of South and Southeast Asia. Public Management Review. 10(1), pp. 101- 126.
Sarji, A. (1995) The Civil Service of Malaysia: Towards Efficiency and Effectiveness. Kuala
Lumpur: Government of Malaysia.
Saunders, M., Lewis, P. and Thornhill, A. (2003) Research Method for Business Students.
Prentice Hall.
Schick, A. (1998) Why Most Developing Countries should not Try New Zealand’s Reforms. World
Bank Research Observer. 13(1), pp. 123-131.
Schneider, F. (2003). The Size and Development of the Shadow Economy Around the World and
the Relaxation to the Hard Tax.
Schneider, Friedrich and Dominik Enste (2000) Shadow Economies around the World: Size,
Causes, and Consequences. IMF Working Paper WP/00/26, International Monetary Fund,
Washington, DC.
Scott, G. (2001) Public Management in New Zealand: Lessons and Challenges. Wellington: Astra
print.
Shah, A. (2007) Tailoring the Fight against Corruption to Country Circumstances. In: Shah, A.,
ed., (2007) Performance Accountability and Combating Corruption. Washington, D.C.: World
Bank, pp. 233- 254.
Shah, H. (2011) Cosmetic Façade. DAWN, 31 January. [Accessed 11 February, 2011].
Sikka, P. (2008) Enterprise Culture and Accountancy Firms: The New Master of Universe,
Accounting, Auditing and Accountability Journal, 21(2): 268-295.
Silvani, C. and Baer, K. (1997) Designing a Tax Administration Reform Strategy: Experiences
and Guidelines. IMF Working Paper 97/30. Washington D.C.: IMF.
Skocpol, T. (1984) Vision and Method in Historical Sociology, Cambridge University Press, UK.
299
Sobh, R. and Perry, C. (2006) Research Design and Data Analysis in Realism Research.
European Journal of Marketing, 40 (11/12): 1194-1209.
Talbot, C. (1994) Reinventing Public Management: A Survey of Public Sector Managers’

Reactions to Change. Corby: Institute of Management.
Talbot, C. (2004) Executive Agencies: Have they Improved Management in Government? Public
Money & Management. 24(2), pp. 104-112.
Taliercio, R. (2004) Designing Performance: The Semi-Autonomous Revenue Authority Model in

Africa and Latin America. World Bank Policy Research Working Paper 3423. Washington D.C.:
World Bank. Available from: http://elibrary.worldbank.org/doi/book/10.1596/1813-9450-3423
Taliercio, R. (2004a) Administrative Reform as Credible Commitment: The Impact of Autonomy

on Revenue Authority Performance in Latin America. World Development. 32(2), pp. 213-232.
Tanzi, V. (1998) Corruption Around the World: Causes, Consequences, Scope and Cures, IMF
StaffPapers, 54 (4): 559-594.
Tanzi, V. (1998). Corruption and the Budget: Problems and Solutions. In Economics of
Corruption, Ed. Arvind, K. J. Massachusetts, Kluwer Academic.
Tanzi, V. and Davoodi, H. R. (2002) Corruption, Growth and Public Investment, in Abed George
T. and Gupta Sanjeev (eds) Governance, Corruption and Economic Performance, Washington,
DC: International Monetary Fund, 197-222.
Tanzi, V., and Davoodi, H. (2000) Corruption, Growth, and Public Finances. IMF Working Paper
WP/00/182, International Monetary Fund, Washington, DC.
Taylor, N. (2001) Understanding Taxpayers Attitudes through Understanding Taxpayers'

Identities, Centre for Tax System Integrity, Working Paper No. 14, Canberra; AustralIan National
University.
Terkper, S. (1994) International: Improving the Accountancy Context of Tax Reform in

Developing Countries. Bulletin for International Fiscal Documentation. 48 (1), pp. 21-32.
Terkper, S. (1999) Revenue Authorities- A Comparison of the Experience of Ghana and East
African Countries. International Bureau of Fiscal Documentation. 53 (4), pp.171-179.
The Economist (2012) Pakistan’s Economy. Plugging Leaks, Poking Holes, Who will Pay for
Pakistan’s State? [Accessed 11 December, 2012].
The News (2012a) Pakistan among Top Seven Most Corrupt, Most Insecure Nations. 29
November. [Accessed 3 December, 2012].
The News (2012b) Patwari made Osama Pay Rs. 50,000 Bribe. 26 December. [Accessed 26
January, 2012].
The News (2012c) Daily Corruption Hovers around Rs.12 bn: NAB Chief. 14 December.
[Accessed 20 December, 2012].
300
The News (2012d) FBR Chief Plans to Get Evaders to Pay Up.17 December. [Accessed 20
December, 2012].
The News (2012e) Traders Call for Making FBR Independent Body to Broaden Tax Net. 15
December. [Accessed 20 December, 2012].
The News (2013a) FTO Terms Less Number of Returns Filing Shameful for Tax Machinery. 22
March. [Accessed 22 March, 2013].
The News (2013b) Tax Authorities Preparing Proposals to Qualify for World Bank Loan. 24
The News (2013c) UK Committee Confronts Pak Politicians on Tax Issues. 5 February.
[Accessed 5 February, 2013].
The News (2013d) Pak Politicians Taking our Taxes, not paying their Own: UK Minister. 30
January. [Accessed 30 January, 2013].
The News (2013e) Nawaz Picks Dream Team, all from Brother’s Closet. 24 May. [Accessed 24
May, 2013].
Theobald, R. (1997) Enhancing Public Service Ethics: More Culture, Less Bureaucracy?
Administration & Society. 29(4), pp. 490–504.
Therkildsen, O. (2004) Autonomous Tax Administration in Sub-Saharan Africa: The Case of the
Uganda Revenue Authority. Forum for Development Studies. 31(1), pp. 59-88.
TRA (1998) Performance indicators quoted are from Tanzania Revenue Authority Corporate
Plan, November, Unpublished.
nd
Transparency International (1997). TI Source Book, 2 Edition.
Transparency International. (2003) Tax Evasion and Corruption: Is there a Trade-off?-Moldova.

Lilia Carasciuc: Author.
Turner, Mark and David Hulme (1997) Governance, Administration and Development: Making the
State Work, Basingstoke: Macmillan.
Venard, B. and Hanafi, M. (2008) Organisational Isomophism and Corruption in Financial

Institutions: Empirical Research in Emerging Countries, Journal of Business Ethics, 81: 481-498.
Verhoest K, Peters GB, Bouckaert G, Verschuere B. (2004) The study of organisational

autonomy: a conceptual review. Public Administration and Development 24(2): 101–118.
Wade, R. (1982) The System of Administrative and Political Corruption: Canal Irrigation in South
India. Journal of Development Studies 18: 287–327.
World Bank (1999) An Anticorruption Strategy for Revenue Administration. PREM Notes Number
33. Washington D.C.: World Bank.
301
World Bank (2000) Difficulties with Autonomous Agencies. Washington D.C.: World Bank.
Available from: http://go.worldbank.org/JDHO04ZK90 [Accessed 20 June 2009].
World Bank (2001) Strengthening Peru’s Tax Agency. PREM Notes Number 60. Washington
D.C.: World Bank.
World Bank (2002) Organizational Design Profiles of Semi-Autonomous Revenue Authorities in

Developing Countries. Washington D.C.: World Bank. Available from: www1.worldbank.org./
public sector/ tax /autonomy.html [Accessed 20 June 2009].
World Bank (2004) Mainstreaming Anticorruption Activities in World Bank Assistance: A Review
of Progress since 1997.Washington, DC: World Bank.
World Bank (2004a) Salary Bonuses in Revenue Departments: Do They Work? PREM Notes
Number 83. Washington D.C.: World Bank.
World Bank (2004b) Project Appraisal Document for the Tax administration Reform Project in
Pakistan. Report No. 30374 – pk. Washington D.C.: World Bank.
World Bank (2004b) The Costs of Corruption. Web Site: News/Feature Stories (April 8, 2004).
World Bank (2009) Pakistan Tax Policy Report. Tapping Tax Bases for Development. Report No.
50078-PK. Washington D.C.: World Bank.
World Bank (2013) Data on Pakistan. Available online at

http://data.worldbank.org/country/pakistan.
World Bank. (1997) Helping Countries Combat Corruption. The Role of the World Bank. PREM
Notes Public Sector,4. Washington D.C.: The World Bank.
Yasin, A. (2012) Why Taxation System is Being Run on SROs, asks PAC. The News, 12
December. [Accessed 20 December, 2012].
Yin, R. K. (1981) The Case Study Crisis: Some Answers. Administrative Science Quarterly. 26
(1), pp. 58-65.
Yin, R. K. (1984) Case Study Research: Design and Methods. Beverly Hills CA: Sage.
Yin, R. K. (1994) Case Study Research: Design and Methods. Thousand Oaks: Sage Publishing.
Yin, R. K. (2003) Case Study Research: Design and Methods. Thousand Oaks: Sage
Publications.
Yitzhaki, S. (1974) Income Tax Evasion: A Theoretical Analysis, Journal of Public Economics,3
(2): 201-202.
Zakiuddin, A. and Haque, W. (2002) Corruption in Bangladesh: An Analytical and Sociological

Study, Online. Available at:
http://unpan1.un.org/intradoclgroups/publicldocumentslapcity/unpan004881 (accessed 28
October 2010).
302
Zuleta, J. C., Leyton, A. and Ivanovic, E.F. (2007) Combating Corruption in Revenue
Administration: The Case of VAT refunds in Bolivia. In: Campos, J.E. and Pradhan, S., eds.,
(2007) The Many Faces of Corruption: Tracking Vulnerabilities at the Sector Level. Washington,
D.C: World Bank, pp. 339-366.

Final Thesis Sobia Khurram

Uploaded by

Copyright:

Available Formats

Final Thesis Sobia Khurram

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Final Thesis Sobia Khurram

Uploaded by

Copyright:

Available Formats

CORRUPTION IN THE DEVELOPING WORLD: THE CASE OF SEMI-

AUTONOMOUS REVENUE AUTHORITIES WITH SPECIAL

A Thesis Submitted in Partial Fulfilment of the Requirements for the Degree

Stirling Management School

late supervisor Professor. Rob Ball

2.4 THEORETICAL UNDERPINNINGS OF SEMI-AUTONOMOUS REVENUE AUTHORITY MODEL 32

4.3.2 Document Analysis 118

in Reduction of Discretionary Powers of Tax Officials

Figure 1.1: Structure of the Thesis 6

Appendix 1 Semi-Structured Interview Schedule…………………………………………………………... 266

CBR Central Board of Revenue

DFID Department for International Development

DMG District Management Group

FBR Federal Board of Revenue

FPSC Federal Public Service Commission

FTO Federal Tax Ombudsman

GoP Government of Pakistan

HRM Human Resource Management

IIMU Internal Investigation and Monitoring Unit

IMF International Monetary Fund

IRS Inland Revenue Service

KRA Kenyan Revenue Authority

MABs More Autonomous Bodies

MoFs Ministries of Finance

NAB National Accountability Bureau

NADRA National Database and Registration Authority

NPM New Public Management

NRS National Revenue Secretariat Ghana

NTN National Tax Numbers

PAC Public Accounts Committee

QUANGOS Quasi-Autonomous Non-Governmental Organizations

Rs. Rupees (Pakistan Currency).

SARAs Semi-Autonomous Revenue Authorities

SARS South African Revenue Service

SAT Mexican Tax Administration Service

SAT Superintendencia de Administración Tributaria Guatemala

SENIAT National Integrated Tax Administration Service

SIN National Intelligence Service Peru

SRO Statutory Rules and Orders

SUNAT Superintendencia Nacional de Administración Tributaria Peru

TARP Tax Administration Reform Project

TRA Tanzania Revenue Authority

URA Uganda Revenue Authority

WJP World Justice Project

INTRODUCTION TO THE THESIS

against successful implementation of SARA reform. Several other researches highlighted

consideration of SARA reform as a universal remedy for reforming tax administrations in

2006; Mann, 2004; McCarten, 2006).

explained by conceptual models’ consequences. This observation pointed to a need to closely

countries by analyzing prominent conceptual frameworks in the field.

reform efforts to enhance tax collection.

1.1 AIMS AND OBJECTIVES

 Why some SARAs failed to effectively control opportunities for corruption?

This study is aimed at analysing corruption in tax administrations in developing countries

Martinez-Vazquez et al. (2004). The development of this framework is aimed to contribute

corruption in tax administrations in developing countries, but also to improve revenue

performance as a consequence. In order to analyze the validity of the framework through

analytical framework resulted in two research hypotheses:

Research hypothesis 1: By preferring personnel autonomy over effective accountability, SARAs