Public sector management reforms in Africa: Analysis of anticorruption strategies in Kenya
1 Table of Contents Chapter 1: Introduction 9 i. The problem statement 12 ii. Rationale for the study 12 iii. Theoretical perspective 14 iv. Methodology 15 v. Data Sought 16 vi. Method of gathering data 18 Chapter 2: Understanding Corruption: Governance and Development Discourse; 22 Strategic context: 22 Definitions of corruption 26 (i) Public office-legalist definition of corruption: 27 (ii) Moralist definition of corruption: 27 (iii) Functionalist definition of corruption: 29 (iv) Public interest definition of corruption: 31 (v) Market-centered definition of corruption: 32 Types of corruption: 34 ii. Embezzlement: 35 iii. Fraud: 36 Governance and Anticorruption 41 Governance Debate: 43 Governance/Anticorruption Discourse in International Development 47 Anticorruption as International Development Agenda 49 Concluding Remarks: 56 Chapter 3: Causes and Effects of Corruption: Review of Literature 57 Strategic Context: 57 Causes of Corruption 57 i. Political Science (New Institutionalism) Explanation 57 ii. The Principal-Agency Perspective: 58 Principal-Agency Perspective and Corruption in Kenya: 61 iii. Political Systems as the Cause of Corruption 63 iv. Political Science (Rational Choice) on Causes of Corruption 65 v. Political Science (Historical Institutionalism) on Causes of Corruption 66 vi. Political Science (Neo-Patrimonial) on Causes of Corruption 66 vii. Political Science (Clientelist Networks) Causes of Corruption 71 viii. Anthropology Explanation: Causes of Corruption 74 ix. The New Institutional Economics on Causes of Corruption 81 Political Science Influence on World Bank Anticorruption Strategies 84 Effects of Corruption: A Review of Literature 84 i. Impact of Corruption on Governance 88 ii. Impact of Corruption on Politics 91 iii. Impact of Corruption on Political Rights and Participation 93 iv. Impact of Corruption on Public Sector Regulations 94 v. Corruption and Public Sector Administrative Performance 95 vi. Corruption and the Rate of Investment 96
2 vii. Corruption and Economic Growth 100 viii. Corruption and Public Sector Expenditures 101 ix. Corruption and the Ability of Open Economy Management 102 x. Corruption and Public Sector Institutions 104 xi. Corruption and Delivery of Public Services 106 xii. Corruption and Public Perceptions 108 xiii. Corruption and Inequality 108 ix) Corruption and State Capture 109 Conclusion 112 Chapter 4: Approaches to Anticorruption; 113 Strategic Context 113 World Bank Anticorruption Model: 115 Implementation of World Bank Anticorruption Model 120 i. Civil Service Reforms: 120 ii. Public Financial Management Reform 122 iii. Judicial Reform and Anticorruption Legislation/Agencies 123 iv. Economic Liberalization Policies 128 v. Political Accountability and Competitive politics, 129 vi. Civil Society Participation and Private Sector 130 vii. Political Leadership and Anticorruption 132 World Bank Anticorruption Assistance to Developing Countries 135 Concluding Remarks 137 Chapter 5: Kenya - The Nature and Sources of Corruption 138 Strategic Context 138 Nature of Corruption 138 i. Petty Corruption 140 ii. Grand Corruption 141 iii. Looting 143 iv. Weakness in Accountability 144 Institutional Sources of Corruption 145 i) Weak judiciary and law enforcement institutions 145 ii) Weak Monitoring, Transparency, and Accountability in Public Sector 148 iii) Legislative Institutions Ineffectiveness 149 iv) Weak Public Financial Management System 151 Social-Political Sources of Corruption 151 i) Tribal-based Elite Domination State 151 ii) Tribe, and Nepotism 153 Political Economy Sources of Corruption 156 i) Contestation over Political Space 156 ii) Lack of Access to Public Sector Information 157 iii) Poor Working Conditions in the Public Sector 158 iv) Fraud in Public Sector Procurement 158 v) Unwillingness to rise to challenges of nation-building 160 Concluding Remarks 161 Chapter 6: Governance and Anticorruption: Kenya and Botswana Compared 162 Strategic Context 162
3 Kenya and Botswana in perspective: 168 Botswana's Governance Experience: 170 Administration of President Seretse Khama (1960 - 1980) 172 Administration of President Ketumile Masire (1980-1998) 174 Administration of President Mogae Administration (1997 - onwards) 174 Botswana Anticorruption Experience 174 Kenya's Governance experience: 177 Administration of President Jomo Kenyatta (1963 - 1978) 180 Kenya: post-colonial nation building contract 182 Administration of President Daniel Arap Moi (1978 - 2002) 184 Administration of President Mwai Kibaki (2002 - onwards) 187 Lessons from Botswana and Kenya's governance experience: 191 i. Inclusive nation-building 191 ii. Building Comprehensive Anticorruption Strategy 193 iii. Tackling Tribe and Corruption 194 v. Governance Matters 196 Kenya: Role of Foreign Aid Actors in Poor Governance 198 Concluding remarks 200 Chapter 7. Kenya: Cases and costs of corruption 202 Strategic Context 202 Corruption in the Judicial Sector 202 i. The Case of a Senior Magistrate in the Nairobi Law Courts in 2002: 204 Corruption in the Public Sector 208 i. Export Compensation and Goldenberg Scandal 209 ii. White Elephant Corruption 224 Kenya: Costs of corruption 226 i) Corruption and relations with foreign aid donors 226 ii) Corruption and investment uncertainty 228 iii) Corruption on quality of public services 230 iv) Corruption and Economic Growth 231 v) Corruption and Insecurity 234 vi) Corruption and rising poverty 236 vii) Corruption and state capture 237 viii) Corruption and public sector expenditure wastage 239 ix) Corruption and public sector procurement fraud 242 x) Corruption and uncollected revenues 245 Concluding Remarks 247 Chapter 8. Kenya: Anticorruption activities by the government 248 Foreign donors' assistance for anticorruption work 249 Implementation of Kenya Government's Anticorruption Strategy 252 i) Phase One: Populism strategy 252 ii) Phase Two: The dream team strategy: 253 iii) Phase Three: Structural reforms strategy: 253 iv) Phase Four: Procurement and anticorruption authority strategy: 257 i) The Kenya anticorruption authority, - Part I 258 ii) The Kenya anticorruption authority, - Part II 261
4 iii) The Collapse of Kenya's anticorruption authority 261 v) Phase Five: Anticorruption legislation strategy; 263 A Critique of Kenya Anticorruption legislation 267 Chapter 9. Kenya: Anticorruption strategy by civil society and media 274 Strategic Context 274 Kenya: Civil society and anticorruption work 278 Kenya media and anticorruption work 282 Kenya: Civil Society, Media, and Challenges of Anticorruption Work 285 Kenya: Civil society anticorruption work and the challenge of Tribalism 287 Kenya: Strengths of civil society in Anticorruption 288 Kenya: Weaknesses of Civil society in anticorruption work 288 Concluding Remarks 291 Chapter 10: Kenya - Relevance and Efficacy of anticorruption strategies 293 Strategic Context: 293 Kenya: Relevance of anticorruption strategies 295 Kenya: Efficacy of anticorruption strategies 298 Kenya: Factors militating against anticorruption work 301 Monopoly, discretion, and weak accountability: 302 Tribal divisions and anticorruption strategies 304 Mapping nature and sources of corruption 306 Critique of World Bank's Anticorruption interventions in Kenya 307 Concluding Remarks 310 Chapter 11: Contributions to anticorruption practice and recommendations 312 Strategic Context 312 Study's contributions 313 Recommendations for future anticorruption work in Kenya 317 (i) Build professionalism in public sector 317 (ii) Conditionality of foreign aid assistance 318 (iii) Focus on comprehensive reform program 319 (iv) Provide incentives for societal coalition building 320 (v) Provide access to public information 321 (vi) Critical role of parliament 322 Chapter 12: Conclusion 324 Bibliography 339 Biographical data 390
5 Tables Table 1: World Bank and OECD Anticorruption Models 116 Table 2: Level of Corruption by Countries 117 Table 3: World Bank Governance Lending 135 Table 4: World Bank Anticorruption Lending 136 Table 5: FDI/GDP versus Cost of Corruption in Kenya, 1991-1996 227 Table 6: World Business Environment Survey - Quality of Public Services 231 Table 7: World Bank - Factors Affecting Corruption in Kenya 250
Abbreviations 6 ADB AIDS BDP BIP BNF BPI CGD CJSF CLARION CORI COTU CPI CSOs DCEC DFID DOD EASF EU FDI FMN FPPS G8 GDP GEMA GM GOK HIV IEA IFIs African Development Bank Acquired Immune Deficiency Syndrome Botswana Democratic Party Botswana Independence Party Botswana National Front Bribe Payers Index Centre for Governance and Development Christians for A Just Society Centre for Law and Research International Corruption Online Research and Information System Central Organization of Trades Unions Corruption Perception Index Civil Society Organizations Directorate on Corruption and Economic Crime Department of Foreign and International Development Department of Defense Enhanced Structural Adjustment Facility European Union Foreign Direct Investment Futa Magendo Action Network Family Planning Private Sector Group of Eight Gross Domestic Product Gikuyu-Embu-Meru Association Greenbelt Movement Government of Kenya Human Immunodeficiency Virus Institute of Economic Affairs International Financial Institutions
IMF KACA KANU KARA KBC KFA KG KHRA KSH MCA MI MRK NARC NCCK NEPAD NGO NIE NMG OECD OED OP PAC PAC PRSP SAP SODNET TI UB UK UNAC UNAIDS International Monetary Fund Kenya Anticorruption Authority Kenya African National Union Kenya Alliance of Residents Associations Kenya Broadcasting Corporation Kenya Farmers Association Kilograms Kenya Human Rights Association Kenya Shillings Millennium Challenge Account Mazingira Institute Moral Rearmament Kenya National Rainbow Coalition National Council of Churches of Kenya New Economic Partnership for Africa Development Non-Governmental Organizations New Institutional Economics Nation Media Group Organization for Economic Cooperation and Development Operations Evaluation Department Operation Firimbi Public Accounts Committee of Parliament Puppets Against Corruption Poverty Reduction Strategy Paper Structural Adjustment Programs Social Development Network Transparency International University of Basel United Kingdom United Nations Convention Against Corruption United Nations AIDS Program
8 UNDP United Nations Developemtn Progrm UNO United Nations Organization US AID U.S Agency for International Development WB World Bank
9 "Corruption will now cease to be a way of life in Kenya and I call upon all those members of my government and public officers accustomed to corrupt practice to know and clearly understand that there will be no sacred cows under my government," (Mwai Kibaki, 30/12/2002, president of Kenya).1 Chapter 1: Introduction Kenya was the first country to sign and ratify the United Nations Convention Against Corruption (UNCAC) on 9 December 2003.2 The UNO Convention provides for mutual legal assistance and judicial cooperation between state parties, for the purpose of detecting, searching for, and returning the proceeds of corruption. Seven days later, the "Daily Nation" of 16' December 2003 reported that Mr. Kiraitu Murungi, Kenya's Minister for Justice and Constitutional Affairs had visited London, England. The purpose of his visit related to over one billion United States Dollars, which had been traced to bank accounts in Europe by private investigators, Kroll and Associates,3 at the request of the government of Kenya. The money was said to relate to the Goldenberg scandal, which is also the subject of this project. On the 27 December 2002, Kenyan voters rejected Kenya African National Union (KANU), and its legacy of corruption, and gave the National Rainbow Coalition (NARC) a plurality mandate in excess of 70 percent,4 ushering in a new government with Mr. Mwai Kibaki at its helm. In his inauguration speech (quoted above) on December 30, 2002, President Kibaki stated that: "Corruption will now cease to be a way of life in Kenya and I call upon all those members of my government and public officers accustomed to corrupt practice to know and clearly understand that there will be no sacred cows under my government." 1 See: Inauguration speech, http://www.statehousekenya.go.ke/speeches/kibaki/2002301201.htm (online, accessed July 25,2003). 2 See: http://www.unodc.org/unodc/en/crime_convention_corruption.html (online, accessed July 25, 2003). 3 See http://www.krollworldwide.com/. Kroll is an independent risk consulting company. 4 See: http://www.transparency.org/cgi-bin/dcn-read.pl?citJD=95219 (online, accessed 30-12-03).
10 It was apparent that the fight against corruption in Kenya had gained a new lease of life. The NARC government was elected primarily because the public believed its pledge to fight corruption and to promote economic recovery. How did all this happen? Starting in 1994, anticorruption has become an issue of major economic, political, and development significance in Kenya. During that period, corruption scandals became a prominent source of media interest, and newspapers frequently published stories about illicit behavior of politicians and public officials. The media (particularly the Nation and Standard newspapers) also linked corruption to several negative outcomes, including the general low levels of accountability in the public sector, tribalism, patronage and nepotism, high poverty levels, poor public services, low investment rates, elite capture of state, and deplorable conditions of infrastructure. In addition, between 1990 and 2002, Transparency International annually released their global corruption perception indexes in which Kenya ranked as one of the top five most corrupt countries in the world. The media reports and corruption rankings, resulted in Kenya being labeled by Kenyans themselves as the "Nchiya kitu kidogo ".5 Reacting to the aforementioned perceptions, in 1994, the government acknowledged that corruption had become a governance challenge for Kenya. For instance, in 1996, the government set up an anticorruption police squad to prevent corruption. In addition, in its "Poverty Reduction Papers 1999/2000", p. 15-20, the government stated that corruption was hampering its development plans. Likewise, in its development plan, "Agenda to Restore Economic Growth 2000", p. 1-45, the government posited that corruption had resulted in the poor management of public resources, the weakened rule of law, and uneven implementation 5 Swahili word meaning "country of something small" - referring to the fact that paying small bribes was both expected and accepted as the only way to get services or any encounter with the public office in the country.
11 of policy reforms. The "Agenda to Restore Economic Growth 2000," p. 10, for instance, suggested that corruption had been inimical to good governance reforms, investments, equity, sustained economic growth, and poverty reduction. Starting in 1994, the enormity of corruption in Kenya strained relations with international financial institutions (IFIs) and created an atmosphere where foreign donors no longer had faith in the government's policy pronouncements. In 1994 through 1998, the International Monetary Fund (IMF), World Bank (WB), European Union (EU), Organization for Economic Cooperation and Development (OECD), and several multilateral and bilateral donors suspended most of their operational and development loans to Kenya due to widespread corruption and poor governance. Soon after, the World Bank and IMF made the decision where future development loans to Kenya were to be granted, subject to several anticorruption reform conditions: a) the government was required to liberalize its economy, b) it was required to implement anticorruption reforms including strengthening the judiciary and accountability mechanisms in the civil service, c) it was required to involve civil society and parliament in anticorruption activities, (d) it was required to introduce anticorruption legislation including establishing an autonomous anticorruption authority. It was assumed that the reforms in the above areas would reduce corruption through increased competition and a reduction in discretionary control over economic resources and political power. As part of a determined response to the donors and public outcry, the administration of previous President, Daniel Moi (1978-2002), from 1994 through 2002 began to implement the reform conditions under the umbrella of the World Bank public sector management strategy aimed at preventing and combating corruption.
12 i. The problem statement The purpose of this study is to discuss the steps undertaken (or proposed) by the government of Kenya and civil society groups to curb corruption in Kenya from 1994 to 2002. The study intends to examine the extent to which these steps have been, or were likely to achieve their goals given the high level of corruption, patronage, tribalism, and the dysfunctional governance contour in Kenya. The study aims to fulfill three objectives: (a) Describe and advance theoretical contributions in corruption literature. Thus, the study is organized as a literature review, extracting and evaluating the core elements of corruption research in the public sector, public administration, and political science; (b) Discuss the impact of the aforementioned literature on public sector management reforms against corruption. This is a question of reviewing the state of theoretical knowledge in anticorruption research and examining its application in the form of anticorruption strategies in Kenya; (c) Discuss the relevance, efficacy, and effectiveness of the aforementioned strategies. This is a question of the study's contribution to anticorruption practice, and recommendations for future anticorruption work in Kenya and other countries. ii. Rationale for the study In recent years, the principal-agency perspective (institutionalism) in political science and public sector management literature has influenced the design of anticorruption strategies by international financial institutions (e.g., OECD, etc.) and bilateral foreign aid donors in developed and developing nations. The work of Rose- Ackerman (1978,1994,1997), Robert Klitgaard (1988,1989,1991a, 1999b, 2000), Daniel Kaufmann and others (1999,2000,2001, 2002), Michael Johnston (1998,1999,2000,2001,2002), and Sahr Kpundeh (1998, 2000, 2001,2002), which draws heavily on agency institutional assumptions, has influenced
13 implementation of the World Bank's public sector reforms to reduce corruption in the developing world. While there is now plenty of literature on anticorruption, the scope for fresh research in the field has not been exhausted. A notable feature of much of the recent research is the tendency to focus on multi-country econometric surveys, and economic models in the absence of much analytical political economy-informed research that focuses on a single country. Using a methodology that combines desk review, documents analyses, observations, interviews, and situational knowledge of the country, this project contribution is in advancing a theoretical discussion that calls for steps to be taken in identifying and tackling the underlying causes of the dysfunctional governance contour, as part of an all-encompassing anticorruption strategy. It is argued in this study that corruption is just the tip of the iceberg of a political problem. Thus, primarily anticorruption work requires a political solution. Therefore, this study's discussion proposes reform policies that includes incorporating anticorruption within a framework of political solutions. This includes building or strengthening governance (transparency and accountability) mechanisms and a coalition building process which incorporates diverse elements of society and that presents a potential political dimension for effective anticorruption strategies. The findings of this dissertation contribute to the following: 1) general public sector governance literature on the significance of a country's governance contour in understanding corruption and thereby designing successful anticorruption strategies; 2) general political- economy literature on the governance contour in Kenya; 3) political science literature by (a) filling an important gap in anticorruption literature by offering an explanation that describes the problem of corruption in Kenya on the basis of an examination of processes of
14 governance in a post-colonial setting, and (b) how power relations are constructed in ways that allow corruption to emerge. In this regard, the study is less concerned with the question of testing econometric hypotheses or surveys - the conventional paradigm for political science analysis of anticorruption reforms - than with the question of how changes in governance create vulnerabilities for corruption and strategies for countering such vulnerabilities. iii. Theoretical perspective Whilst this study draws on the principal-agency perspective that is derived from the political science (institutionalism) and public sector management literature, it does not totally use it to build a theory. The perspective is used more for explanation. The perspective links corruption to institutional attributes by examining how institutions shape political and policy choices, and influence outcomes. The work of Rose-Susan Ackerman (1978,1994, 1997), Robert Klitgaard (1988,1989, 1991a, 1999b, 2000), Daniel Kaufmann (1998,1999,2000, 2001, 2002), Michael Johnston (1999, 2000, 2001, 2002) and Sahr Kpundeh (1998, 2000, 2001, 2002) draw heavily on this perspective and is especially pertinent to this study. Using this perspective, for instance, these authors stress three dimensions of institutional structure that they consider most critical in creating opportunities for corruption: the monopoly power of political officials, the degree of discretion that officials are permitted to exercise, and the degree to which there are systems of accountability and transparency in an institution. According to the agency perspective, it takes three parties for a transaction to qualify as corruption: a principal (P), an agent (A), and a client (C). Illicit activities will be greater when agents have monopoly power over clients, agents enjoy discretion, and accountability is poor. The case of Kenya fits into the framework: corruption has grown
15 largely to the degree in which the politicians, tribal elites, and power wielders (principals), public officials (agents), and some citizens (clients) have managed to capture and control the political instruments of mobilization and power. In the meantime, the principals and agents continue to resist or delay anticorruption reforms that might curb their power of control. Thus in public pronouncements, especially to aid donors, the principals and agents go along with anticorruption reform initiatives, while at the same time continue to engage in and abate corruption. This study has not attempted to test the agency perspective, but rather, used the principal-agency relationship in advancing explanations and theoretical assumptions regarding the causes and nature of corruption, costs of corruption, cases of corruption, and efficacy and relevance of anticorruption strategies in Kenya. In addition, I have used the perspective to describe corruption as being an integral part of Kenya's political economy that cannot be analyzed separately from the larger legal, governance, and socio-cultural challenges to which corruption owes its existence. This is a theoretical attempt to look beyond the formal structures of the central state to the informal networks of tribe, patronage, and state capture that determine how political power and corruption are actually wielded, iv. Methodology This study has used a "loosely" constructed qualitative methodology that combines historical narratives, political science theoretical assumptions, and a descriptive analysis that resulted in a detailed analysis. The study mostly relied on secondary data from literature, comparative and situational knowledge of the country, desk review, brief observations, and open-ended interviews. In addition, the study also relied on practitioner expertise and theoretical knowledge advanced by the international development organizations in this area.
16 For whilst it is essential to understand the complexity of the corruption within Kenya from an academic and theoretical background, there also exists a growing body of literature and practitioner experience that has the potential to assist in identifying the likely causes and effects of corruption and best practices in tackling it. v. Data Sought In presenting a narrative that captures the inherent linkage between corruption, institutional attributes, social factors (e.g., tribe), and politics in Kenya, I sought secondary data in five areas: a) Information about where corruption is occurring: The information included review of three surveys conducted by Transparency International Kenya Chapter in 1999,2000, and 2002. The surveys identified public and private sector activities, institutions, government agencies and relationships, such as public sector employment and the making of government contracts for goods and services; b) Information about what types of corruption are occurring: In addition to assessing the overall types of corruption prevalent in the country, information that is more detailed was also sought on the types of corruption that tend to occur in government agencies, relationships, or processes. The study, for instance, found that bribery (petty corruption) is a major problem in government contracting, while public service appointments are affected more by nepotism or ethnic connections; c) Information about costs and effects of corruption: According to Daniel Kaufmann (1999) and Robert Klitgaard (1988), understanding the relative effect of corruption is critical to assessing the efficacy of anticorruption strategies in place. Information sought included
17 the direct economic costs, plus an assessment of indirect and intangible consequences, such as relations with aid donors; d) Factors contributing to or associated with corruption: Since there is seldom a single, source of corruption, the information sought identified a number of contributing factors. These included factors such as politics, tribe, and poverty; the low social and economic status of public officials, which makes them more susceptible to bribery; the presence of specific corrupting influences, such as organized cartels and ethnic groups; structural factors, such as overly discretionary powers; and the lack of monitoring, accountability, and an overall poor governance structure. Information sought in this area was critical to understanding the nature of corruption itself and assessing the relevance of anticorruption strategies; e) The subjective perception of corruption by those involved or affected by it: According to Michael Johnston (2000) and Peter Riley (1998), assessment of anticorruption strategies should include a subjective assessment of how those involved perceive or understand what is occurring. This information was sought because the reactions of people to anticorruption efforts will be governed by their own perceptions. The following areas were looked at: • The impressions of those involved ("victims", "offenders", and others) about the types of corruption occurring; • The impression of citizens about corruption and its impact; • The impression of citizens about anticorruption strategies, and whether corruption is in breach of those strategies.
18 vi. Method of gathering data Corruption is by its nature, a covert activity. This makes accurate information hard to obtain and gives many involved a motive for distorting or falsifying any information they may provide. To obtain an accurate assessment, it was therefore necessary for me to obtain information from as many sources as possible and ensure diversity in the sources. This enabled me to identify biases due to falsification or other problems. The major techniques and sources used for gathering information included: i. Desk Review The first stage was to gather as much data as possible from pre-existing sources - mostly qualitative data which included the following; previous assessments by international financial institutions, interest groups, and the auditor-general of the government of Kenya. Additional sources included published policy documents and reports by the government of Kenya: parliamentary reports, parliament committees, records of legislation, government circulars, newsletters, workshop reports, anticorruption literature, and relevant policy reports (evaluations, program information, project assessments, and country reports from donor governments, United Nations Development Program, the World Bank, and IMF on Kenya). The desk review of government anticorruption polices was undertaken to furnish the study with models for anticorruption research and policy measures applied by the Kenya government to controlling corruption. Because the implementation of anticorruption strategy is at an early stage, the desk review only assessed its implementation and relevance. The documents assessed showed how the government intended to or applied requisite anticorruption policies.
19 ii. Open-Ended Interviews The second part was brief participatory interviews with stakeholders (former public officials and civil society activists), aimed at gathering information from the citizenry on the government's commitment to anticorruption reforms. The interviews produced detailed information, including views on corruption, precipitating causes of corruption, and ideas on how the government is tackling it. Ten informal face-to-face interviews were conducted with a wide range of informants including three members of civil society, three former senior Kenya government policymakers, two journalists, and two university students (see table). The individuals interviewed were conversant with issues of anticorruption research and initiatives undertaken to control corruption in Kenya. Observations I used the field observation as a follow-up to gather detailed information on types of corruption (bribery) that were occurring in government offices during the interactions between the civil servants and the public. Field observation, however, was limited to a three- month period, largely due to its expensive nature. In addition, the illicit nature of corruption and the involvement of politicians, public officials, police, and organized crime in corruption made field observations, extensive interviews, and/or surveys impossible in Kenya. In this situation, I relied on comparative knowledge of the country, brief informal observations, personal rapport, and brief interviews outside the country. Theoretical Assessment of Public Sector Institutions and Relationships This theoretical assessment involved consideration of their capacity to fight corruption. I also used it to analyze the nature and type of corruption within some institutions
20 (e.g., the judiciary). The assessment included desk review and advancing theoretical claims regarding government agencies and institutions as well as civil society, parliamentarians, and political parties, v. Assessment of Media: Media: Despite operating in a censored information environment under the administration of Presidents Jomo Kenyatta (1963 - 1978) and Daniel Arap Moi (1978 - 2002), Kenya's two leading newspapers (Daily Nation, East African Standard) have been able to expose corruption throughout the country. Through various informal contacts, these newspaper archives provided the study with literature and specific information on corruption. I decided to use these media because they are the most widely accessed in the country, and especially since it is very difficult to find representative "official" statistics and reports on corruption in Kenya, vi. Assessment of Governance Information: The following qualitative databases provided ready assessment documents on Kenya's leadership and institutional quality, decline in the public sector quality, and rise in corruption and breakdown in law and order. The databases I used included - Corruption Online Research and Information System (CORIS), which provided country information, and anticorruption research documents from policymakers, academia, and civil society in Africa. The World Bank Institute Governance Database (1999-2002) provided extensive anticorruption research literature. Other databases included the Transparency International Perceptions of Corruption indexes (1997-2002).6 6 The "Corruption Perception Index" (CPI) is the most comprehensive quantitative indicator of cross-country corruption available, where each single country is recognizable. The CPI assesses the degree to which public officials and politicians are believed to accept bribes, take illicit payment in public procurement, embezzle (continued)