Sub-National Budget Transparency:

Transcription

1 Sub-National Budget Transparency: Case of Ten Counties in Kenya 1

2 Sub - National Budget Transparency: Case of Ten Counties in Kenya

3 Published by With Funding From All rights reserved 5th Floor ACK Garden House 1st Ngong Avenue P.O. Box Nairobi, Kenya Tel: , Fax: Cell: , Website: Design & Layout Oscar Ochieng 2

4 Foreword The Institute of Economic Affairs (IEA-Kenya) considers the introduction of devolution as the most decisive and far-reaching instance of public sector reforms ever undertaken in Kenya. Devolution is significant because it is the result of constitutional reforms without precedent in many nations. Understandably, such reforms have, not only teething problems, but also conceptual factors that require public education and broad policy discourse. Being that Kenya has had an enduring problem with accountability and transparent utilization of public funds, it was predictable that the counties would exhibit the same difficulties. This project was based on the need to examine the practice of public finance management in selected counties, and identify any difficulties that are in contravention of the established norms for transparency and the prescriptions contained within Kenya s Public Finance Management Act of We opted to utilize a derivative of the Open Budget Survey (OBS) to assess the adherence to these principles by ten selected counties in Kenya. The results are presented in the form of a Sub-National Open Budget Index that ranks the ten counties, thereby giving an idea of the extent to which these counties provide unimpeded access to budget documents related to the county. We are pleased to present the results in this publication, and note that there are lessons on the record keeping and a difference in the effort made by different counties in granting reasonable access to Kenyan citizens. The results also demonstrate that, while counties have had less than two years to consolidate their administrative systems, it is prudent for them to consider transparency in financial management throughout the budget process. By considering transparency in all management activities, they would be respecting the constitutional requirement to include citizens in the management of public affairs. We are immensely grateful for the support of the World Bank Office in Kenya that has enabled us to conduct the study and publish the results of this project. This publication is intended as an input into public education, inform policy reform, and to demonstrate to county governments that transparency in public affairs and financial management has value. We are confident that it is the first in future collaborations to contribute towards consolidating the constitutional promise of better delivery of public services in Kenya through devolution. 3

5 Acknowledgements The Institute of Economic Affairs (IEA-Kenya) would like to express its gratitude to all those who played a role in ensuring the completion of this report. In particular, we sincerely thank the Public Finance Management Programme team, Mr. John Mutua, Mr. Raphael Muya and Mr. Duncan Otieno for the successful implementation of this project. Further appreciation goes to Mr. Kwame Owino, the Chief Executive Officer, IEA-Kenya, for his leadership and professional guidance during the entire life of this project. We appreciate the role of the entire staff at IEA-Kenya who committed their experience, time and effort during the implementation of the project. In addition, we would also like to thank all our field enumerators in the ten counties who committed not only their time, but also their experience in the project. We wish to extend our sincere appreciation to Dr. Jacob Chege (Kenya Institute of Public Policy Research and Analysis), Ms. Eunice Fedha (Council of Governors), Dr. Jason Lakin (International Budget Partnership, Kenya), Ms. Annette Omollo (World Bank, Kenya Office), Ms. Rosemary Irungu (The Institute for Social Accountability), Ms. Lonah Losem (County Assemblies Forum), and Mr. Duncan Gituanja (Chairman, Nyeri County Assemblies Budget and Appropriation Committee) for their invaluably constructive criticism, comments and suggestions. We also acknowledge and appreciate all the county government officials and members of the civil society organizations from all the ten counties who provided us with data and helpful feedback that helped us in coming up with the report. Last but not least, IEA-Kenya thankfully acknowledges the World Bank, Kenya Office for providing us with the financial support and opportunity to successfully undertake this research project. Special thanks go to Christopher Finch and Annette Omollo, World Bank, Kenya Office, for their outstanding role in managing and coordinating the project and Lucy Musira for administrative support provided. In addition, we also grateful for Sanjay Agarwal s and Harika Masud s the invaluable input and comments both from the World Bank Head office. Finally, we do recognize that it is impossible to mention all who contributed to the completion of this project. We, however, wish to thank all our partners, various individuals and agencies for their continued support. Your support is highly appreciated. 4

6 Table of Contents Abbreviations and Acronyms Context Purpose of SNOBS Objectives of SNOBS Methodology Assessment Parameters Ranking Counties using County Open Budget Index (COBI) The Second Part of Findings - Comprehensiveness of Budget Documents Counties Covered Findings from the Survey State of Budget Transparency Availability of Budget Documents and Mode of Distribution Form and structure of public consultations in the budget process is wanting Information on service offered by counties not available at a single source Announcement of procurement tenders made public but limited information on award of tenders and on procurement complaints decisions Assessment of Comprehensiveness of County Fiscal Strategy Paper and County Executive Budget Estimates Dissemination of preliminary results Conclusions and Recommendations...34 References...36 Annexe...37 Annex 1: Summary of Questionnaire and Calculation of COBI...37 Annex 2: Synthesis of Field Reports...40 Annex 3: Synthesis of Dissemination Report...42 List of Tables and Figures List of Tables Table 1: Summary of classification of SNOBS questions by budget documents...16 Table 2: Counties selected for pilot survey...19 Table 3: Summary results from the ten counties on the availability of eight key budget documents...23 List of Figures Figure 1: Key budget documents in the budget cycle...15 Figure 2: County Open Budget Index...20 Figure 3: Average score on comprehensiveness of CFSP by counties...29 Figure 4: Score on comprehensiveness of County Budget Estimates

7 List of Abbreviations and Acronyms BAC Budget and Appropriation Committee BPS Budget Policy Statement CB Citizens Budget CBEFs County Budget and Economic Forums CBEs County Budget Estimates C-BROP County Budget Review and Outlook Paper CEDGG Centre for Enhancing Democracy & Good Governance CFSP County Fiscal Strategy Paper CIDP County Integrated Development Plan COBI County Open Budget Index CSO Civil Society Organization EB Enacted Budget EBP Executive Budget Proposal ECDE Early Childhood Development Education IBP International Budget Partnership IEA Institute of Economic Affairs IYRs In-Year Reports MCA Member of County Assembly MTEF Medium Term Expenditure Framework MYRs Mid-Year Reports OBS Open Budget Survey OCoB Office of the Controller of Budget PFMA Public Finance Management Act 2012 PBB Program Based Budgeting SNOBS Sub-National Open Budget Survey YERs Year End Reports 6

8 Executive Summary Context County governments are now slightly over two years old since coming to effect in March The Constitution of Kenya 2010, that heralded onset of devolution and creation of a two-tier system of government, accords county government substantive, administrative, fiscal, and political responsibilities. As a result, this has heightened expectations that bringing the government closer to the people will address existing resource allocation and service delivery challenges. In the same breath, there are fears that devolution may actually exacerbate inequalities and corruption. On this basis, the IEA, during June-August 2014, implemented a Sub-National Open Budget Survey (SNOBS). Since this survey allows for cross-county comparison, it serves as a useful tool to assess progress of counties in terms of budget transparency, and will in turn be useful in informing policy reform processes. Methodology The Sub-National Open Budget Survey measures the extent to which county governments provide timely, useful and accessible budget information throughout the budget process as provided for in the law. In addition, this survey also assesses the degree of overall public engagement in budgetary processes at the sub-national level, and also whether information on public procurement and on the services offered by counties is made public. This survey, while appreciating that county governments are still very young two years in the transition period goes not only beyond assessing whether counties meets minimum legal requirements, but also benchmarks their performance on international budgeting practices. The SNOBS is based on 70 multiple-choice questions adopted from the International Budget Partnership (IBP). Majority of these questions are based on whether four key budget documents are produced and available to the public. These documents are: County Fiscal Strategy Paper and County Budget Estimates, produced at the formulation stage; County Appropriation Act and County Finance Act, produced at the enactment stage; and Annual Financial Statement, at the execution stage. The County Audit Report 2013/14 is not included in the survey because, legally, it is supposed to be released in early 2015, several months after the research period of June-August Six researchers, trained by the IEA, sought the four budget documents and conducted interviews to complete the questions between June and August Of the six, four researchers covered two counties each, and the other two researchers covered a county each. These questions were later peer-reviewed by the IEA for final answers/responses to each of the 70 questions. The ten counties selected were covered in the survey using simple criterion of urban, peri-urban, and rural differential, as well as geographical proximity. 7

9 The scores used to rank counties are based on the County Open Budget Index (COBI) that assigns each county a score from 0 to 100 based on the simple average of the numerical value of each of the responses to the total 70 questions in the questionnaire. Findings The average score for COBI of 16.2 out of 100 implies that counties provide scant budget information to the public. As shown in the map of Kenya (next page), Nakuru, with a score of 30.5 and far ahead of the other nine counties, provides minimal budget information. Unlike the other counties, the County Fiscal Strategy Paper (CFSP) 2014/15 and the County Budget Estimates 2014/15 for Nakuru were accessible online. CFSP 2014/15 was available to the public in seven out of the ten counties with Mombasa, Garissa and Kajiado missing out and hence their low respective COBI scores. In these seven counties, CFSP 2014/15 was availed online. With the exception of Nyeri, this document was released by the other counties four months prior the start of the financial year as provided for in the law. County Appropriation Act and County Finance Act 2013/14 (Enacted Budgets), as well as the Annual Financial Statement 2013/14 (Year End Report) were not available in any of the ten counties, although it is produced in all the counties for internal use. Taita Taveta, Machakos, and Kajiado have established County Budget and Economic Forums (CBEFs) as a formalized platform for public consultations on matters of county planning and budget. However, these CBEFs are not functional and majority of civil society felt the process of constituting them was not transparent and integral. Granted that capital expenditure is a driver to development, the scope of public discussion and consultation is expected to cover both recurrent and development expenditure as well as revenue side of the budget, which was not the case for majority of the counties. Bungoma and Kisumu provide information on the services they offer at a single source, through service charters; but in the other counties, this information is available from each responsible department. All the counties surveyed make information on the launch announcement of procurement tenders available to the public through various sources, but information on the award of public tenders and on procurement complaints decision is limited. County Budget Estimates in majority of counties lack important details in terms of presentation of expenditure information by economic classification (only Nakuru, Taita Taveta and Nyeri present expenditure by economic classification), prior financial year expenditure and revenue information, and generally very non-financial information is presented. 8

10 Recommendations County governments should publish and publicize CBEs, County Appropriation Act and County Finance Act and YERs already produced for internal use at no extra cost. County governments should develop, publish and publicize Citizens Budget, non-technical and simple version of the main county budget estimates. Counties have an opportunity to partner with CSOs (for example, Nakuru) or they can take the initiative. Besides, they should consult with the public on the content. Improve comprehensiveness of CBEs by focusing on providing information on revenue, expenditure and other fiscal information including disaggregated information, multi-year projections, and so on. County government should adopt PBB because this budgeting approach enhances improved information on budget narrative, non-financial performance targeting, etc., to allow for effective budget analysis. Budgets are poor on outputs and outcomes; need to provide link between counties policies, spending, and macroeconomic policy. County governments should establish CBEFs in a transparent and integral manner as a way of formalizing public consultations and inclusivity. 9

11 How Much Budget Information is Available to the Public: Case of 10 Counties in Kenya 10

12 1.0 Context During June to August 2014, the Institute of Economic Affairs (IAE-Kenya) implemented a Sub-National Open Budget Survey (SNOBS) to assess or measure the extent to which county governments 2 provide timely, useful and accessible budget information throughout the budget process. In addition, this survey also assesses the degree of overall public engagement in budgetary processes at the sub-national level. This project is part of the Institute s work to promote budget transparency and accountability in the country. The onset of devolution, following the promulgation of Kenya s Constitution in 2010, heralded the creation of a two-tier system of government, which is comprised of the national government and 47 county governments. This implies a substantive increase in the political, administrative, and fiscal responsibilities for counties and heightened expectations that bringing the government closer to the people will help to mitigate existing resource allocation and service delivery challenges. Indeed, public reports and media articles show that majority of Kenyans feel that devolution holds the key to dealing with past challenges of centralized governance, including: inequitable distribution of resources; uneven development; and poor service delivery. At the same time, based on decentralization literature 3 and experiences from other parts of the world, there is recognition that devolution is not a panacea for resolving service delivery challenges, and that it may in fact entail risks such as transfer of corruption, weak devolved institutions, low capacity for public participation and exacerbation of existing inequalities. The Constitution has established the right for the public to access all government information, and infused the importance of public participation into all aspects of public administration. For example, Article 201 (a) requires that there be openness and accountability, including public participation in financial matters. Article 196 (2) of the Constitution requires that the county assembly facilitate public participation and involvement in the legislative and other business of the assembly and its committees. The Public Finance Management Act (PFMA), 2012 and other county-related laws include new provisions in relation to PFM in the newly created 47 county governments, as well as relations between national and county governments 4. Further, the national level Open Budget Survey (OBS) results for 2010 and 2012 confirm that, the national government only provides some fiscal and budget information to the public. A budget transparency assessment at the county level would contribute to a more holistic picture regarding public disclosure of budget information in Kenya. In addition, since such endeavours have the potential to expose leakages, enhance tracking and monitoring of budget and increase accountability to the public (Kroth, 2012), it is expected that the findings of this survey exercise will help to identify and implement further reforms in this area. 2 County government used interchangeably with the word county. 3 Fjeldstad (2003) shares similar concerns by noting that devolution can indeed import poor governance to the sub-national level. 4 See PFMA Part IV and Part V. 11

13 Therefore, the increased responsibility of county governments for resource allocation and service provision, combined with constitutional provisions for transparency and public participation in the budget process, have lent to the imperative of piloting a SNOBS in Kenya. 1.1 Purpose of SNOBS County governments, by 27th March 2015, marked their second year since coming to effect. This is obviously telling that county governments are still finding their feet in terms of, for example, setting up their Public Finance Management (PFM) systems, and establishing platforms such as County Budget and Economic Forums (CBEFs) as a vehicle for public consultation in public finances, recruitment of personnel, and adoption of devolved functions, and so on. SNOBS would form a useful assessment tool on the progress of the counties in a number of areas with regard to budget transparency. Besides, SNOBS provides an opportunity for cross-county comparison despite some variations in county fiscal capacity, and in other areas. Budget transparency is defined as the full disclosure of all relevant fiscal information in a timely and systematic manner (Centre on Budget and Policy Priorities, 2006). Fiscal information is concerned with taxes, government spending and public debt operations in provision of public services. Why budgets, you may ask. Budgets are essential for ensuring the delivery of critical services such as basic education, maternal health, water, security and roads, and it is important that we ask if counties are investing our taxes efficiently in these services. Therefore, if budgets are not sufficiently open, citizens will not have access to information therein on how counties propose to raise funds and how much is being spent in a given financial year. Moreover, inadequate budget information implies that, citizens cannot meaningfully participate in budgetary policies and decisions and, therefore, cannot hold the government to account on its expenditure. To this end, this survey, in its first section, assesses whether eight key county budget documents, as provided for in PFMA, are published at every point of the budget process, and more importantly, if they are availed to the public. The second section of this survey assesses the level of detail or comprehensiveness of these documents and availability to the public as per legal requirements and international good budgeting practices. The PFMA provides some legal requirements of the content that should be reflected in some of these budget documents. For example, Section 130, not only states that County Executive Committee Member for Finance should submit to the County Assembly a Budget Summary and County Budget Estimates, but also mentions the content to be expected from each of these two documents. 12

14 Public participation throughout the budget process Budget transparency is not enough to instil accountability. It needs to be accompanied by meaningful public participation in budget decision making and budget monitoring. This is the reason why the SNOBS include a few questions on public consultations in budget formulation and in other stages of budget process to assess opportunities for public participation. The Constitution variously provides legal basis for stakeholder public access to information and participation in planning and budgeting processes through Article 35 of the Bill of Rights; Article 118(1) a-b under the Legislature; Article 196 (1) a-b under the Devolved Government Chapter, and Article 201 (a) under the Principles of Public Finance. In addition, Article 225 (5) explicitly stipulates that, in discussion and reviewing of the budget, a Committee of the National Assembly shall seek representatives from the public and their recommendations shall be taken into account in the final resolution by the National Assembly. In a follow up to Article 201 of the Constitution, the revised Public Finance Management Act, 2013 makes references to a number of clauses that speak to the need for public participation to be incorporated as part and parcel of planning and budgetary processes at the two levels of government. This law, in Section 36 (5) and 125 (2), specifically instructs the Cabinet Secretary for Finance and their counterpart at the county level, the County Executive Committee Member for Finance to ensure the public is engaged throughout the budget process. However, this law does not define the how, when and where of public participation. This is defined in the Regulations of the PFMA. Section 137 of the same law requires each county to establish a platform, CBEF, whose purpose is to formalize public consultation on matters related to county economics, planning, and budgeting. CBEF s composition is made of the Governor as the chair, other Members of County Executive Committee and an equal number of nominated representatives from professionals, businesses, labour groups, faith based groups, women groups, the elderly, and person with disabilities. Recently, and as was captured in the media, the Commission on Revenue Allocation came up with new rules and guidelines that seek greater public input in county budgets, in an attempt to fill the void on the details for how CBEF will function. One of the outstanding rules is that, budget documents must be released two weeks prior to any formal discussion, and that notice for convening meetings must take a similar time. These rules are also intended to deter county meetings from elite capture and enhance wide scope in consultations. Assessment beyond meeting legal requirement Whereas most county administration may feel that the assessment of budget transparency should be restricted to meeting legal requirements, this SNOBS goes beyond this. While appreciating that county governments are still relatively young, the assessment goes, not only beyond examining whether counties have met legal requirements, but also to benchmark counties to international best budgeting practices. After all, some counties are already beyond just meeting legal requirements. For example, PFMA is not explicit on presentation of forward years (multi-year projections) of expenditure by county government ministries/department, but a number of counties are already doing this. 13

15 In sum, SNOBS will compare and ranks counties on their level of budget transparency, and use the findings to inform policy reform processes with regard to budget transparency and accountability at the county level in future 1.2 Objective of SNOBS It is expected that this survey and its findings will be instrumental in improving access to budget information and overall budget transparency at the county level, not only in line with the local legal PFM framework, but also in line with international benchmarks, and as such, contribute to strengthening county PFM systems. In addition, the survey is aimed at empowering citizens to be more involved throughout the budget process at the county government level, right from influencing budget priorities and needs, to monitoring the budget and evaluating public service delivery. 14

16 2.0 Methodology 2.1 Assessment Parameters The questionnaire used for the County Open Budget Survey was adapted from the IBP s Open Budget Survey (OBS) which collects information on budget transparency, oversight and participation at the national level on a biennial basis. The OBS assesses whether eight key budget documents, as illustrated in Figure 1 below, are published and availed to the public. The figure shows the international standard terms used for the documents at every point of budget cycle and the equivalent name of the same documents that should be produced at the county government level. Figure 1: Key budget documents in the budget cycle Source: Adopted from IBP. 15

17 Borrowing from the OBS, it is noteworthy that, the SNOBS questionnaire that we adopted and customized has 75 multiple-choice questions and is based on five out of the eight key county budget documents, as shown in Table 1 below. The table shows the five documents and matching the fiscal year to be assessed on the SNOBS. The first section of the survey documents public availability of five out of the eight key budget documents that each county is supposed to produce as per the Constitution of Kenya 2010 and the Public Finance Management Act, As listed in Table 1, we will assess whether each of the ten counties produced and published the five documents, and for the cited fiscal years. Table 1: Summary of classification of SNOBS questions by budget documents Stage of County Budget Process Question Numbers Budget Documents Considered for the Survey Fiscal Year Used Budget Formulation Questions County Fiscal Strategy Paper 2014/15 Section Two - Questions 6-27 Budget Execution Section Three - Questions Budget Reporting and Oversight Section Four - Questions 61-63; Section four - Questions 68; County Budget Estimates 2014/15 Supporting Documents such as Budget Summary County Appropriations Act and County Finance Act 2014/ /14 Annual Financial Statement 2013/14 County Audit Report 2013/14 According to the IBP, and for the purpose of this exercise, budget documents are considered or described to be publicly available if they meet the following criteria: If any member of the public can access these documents online. If any member of the public, upon making a formal direct request from the county offices, receive the document(s) within ten working days. If the documents are available at a minimal cost to any person who want to use them. If the documents are available in public institutions such as public libraries/resource centres, public universities, at ministries, and so on. What follows after establishing whether these documents are published and availed to the public is to assess the comprehensiveness (level of content) of each respective document. The response to this is based on the answers to the questions aligned to each document as shown in second column of Table 1. This questionnaire does not assess availability of county mid-year report, county in-year report, and citizen budget of the County Budget Estimates, and hence there are no questions based on these three 16

18 documents. These three documents are deliberately omitted from the survey but should be considered in future scale up of this exercise. The second parameter, crucial to this survey, is on the timeliness of release of these budget documents to the public or to the county assembly, depending on which document we are referring to. Further, and indeed specific to CBE, there are two questions on whether it is submitted to the county assembly in advance to enable parliamentarians ample time to scrutinize, debate, and approve it before the beginning of a new financial year. Question on this parameters are numbers 1, 29, 30, 60, 69 and 73. The extent to which these documents reach the public is dependent on the range of dissemination mechanisms used. Questions 28, 48 and 67 test this on three documents, namely, the CBE, County Appropriation and Finance Act, and on the Annual Financial Statement. Given the importance of public input into these documents, or the processes culminating into publication of these documents, there are a few questions, 2-4, that focus on whether there exists a formal mechanism for the public to influence budget priorities, its scope of discussion and whether forums are representative. Related to this are questions 5, 64 and 72 that assess whether the public has an opportunity to input or engage with the other budget documents published at the enactment, implementation, and audit stage of the budget process. Public procurement comprises a substantial part of public spending at both the national and county level of government, and as such, issues of transparency and competitiveness of the entire process are critical. Kenya has a procurement law and the Constitution also speaks on issues of procurement. As such, questions seek answers on how transparent and fair this process is. The final parameter, service delivery oversight, assesses the extent to which citizens are able to access necessary information on the kind of services offered at the county. In addition, this parameter examines whether information on who is charged with providing these services is available and also whether there is a platform to channel complains or lobby for better or more appropriate services (questions 53-59). 2.2 Ranking Counties using County Open Budget Index (COBI) In order to rank and compare the ten counties surveyed on their level of budget transparency, the number of questions used in the County Open Budget Survey are whittled down to 70 out the total set of 75 questions. There are five questions based on the audit phase of the budget cycle and largely on the county audit reports. The research period for this survey was June to August The 2013/14 county audit reports are legally supposed to be released around December 2014 or January 2015, several months post the research period. As such, this document was not available and hence the reason for excluding it from the survey including the five questions aligned to it. As such, it would be unfair to penalize counties for a document that was not yet released during the survey/research period because it was legally supposed to be released much later. 17

19 The scores used to rank counties are based on the County Open Budget Index (COBI) 5 that assigns each county a score from 0 to 100 based on the simple average of the numerical value of each of the responses to the total 70 questions in the questionnaire. The COBI reflects the timeliness and comprehensiveness of publicly available budget information in the four key budget documents as shown in Table 1 above (excluding the audit report). The other aspects of COBI, as discussed above, include: assessment of public consultation avenues and engagement or scrutiny of budget documents, information related to service delivery at the county level, and the county procurement system. 2.3 The Second Part of Findings - Comprehensiveness of Budget Documents This report includes a section that provides findings on comprehensiveness of budget documents. From the preceding section, CFSP was publicly available from seven of the ten counties surveyed; while it is only Nakuru that publicly availed its CBEs. According to SNOBS, assessment of the level of comprehensiveness would only be focused on these aforementioned documents. In other words, this survey would only assess CFSP in the seven counties for comprehensiveness. In so far as the CBE is concerned, it is only the one from Nakuru that would be assessed. However, to make this project worthwhile, and based on lessons learnt from three countries Brazil, Indonesia and Tanzania that conducted a similar SNOBS project in 2013, we asked enumerators to use secondary or third party means to obtain the CBE from the other nine counties. The objective of this section is not to rank the counties per se but to assume that if CBEs were publicly available, how counties would score on the level of comprehensiveness so as to inform future scale up of SNOBS. We managed to obtain copies of CBEs from nine counties. For CFSP, our efforts to obtain copies from three counties were futile. We were, however, not able to obtain the full set of Enacted Budgets and Annual Financial Statements from any of the counties. Therefore, the section on comprehensiveness of budget documents focus will be limited to both CFSP and CBEs. 2.4 Counties Covered The survey was piloted in ten counties that were selected based on selection criteria including urbanrural differentials, fiscal capacity, population, and geographic proximity, and so on (see Table 2 below). The majority of selected counties also overlap with areas where the World Bank could potentially conduct performance monitoring to complement the SNOBS exercise. 5 Details of how the COBI is computed are discussed in the Annex. 18

20 Table 2: Counties Selected for Pilot Survey Regions Rural Urban Peri-Urban Rift valley Nandi and Kajiado Nakuru Central Machakos Nyeri Western Bungoma Kisumu Northern Garissa Coast Taita Taveta Mombasa 19

21 3.0 Findings from the Survey 3.1 State of Budget Transparency Majority of Counties Provide Scant Budget Information The average score for COBI is 16.2 out of 100. This indicates that, nine out of the ten counties provide scant budget information to the public. Nakuru, with a score of 30.5, leads the pack by a considerable margin and is the exception as it provides minimal budget information. Three counties, Kisumu, Taita Taveta, and of course Nakuru, have scores higher than the average, as shown in Figure 2 below. The Constitution, the PFM Act 2012, and the County Government Act all emphasize public participation in public finance management processes, including planning and budgeting processes at the national and county government levels. To this end, public disclosure of budget information is an important pre-requisite for the public to provide input for the budget and engage effectively in the entire budget process. Given the above scenario in the ten counties where scant budget information is provided, this is likely to impede debates and discussions on the budget, as well as public understanding of budget policies, county revenue and spending plans. In fact, the status quo may run contrary to counties current efforts towards establishing citizen participation mechanisms, as is legally required of them. Without improvement in increased public disclosure of budget information, the likely outcome is less transparent budgets that are not responsive budgets to citizens voices. Figure 2: County Open Budget Index 20

22 Cluster of COBI Scores Based on Degree of Budget Information No. of Counties Extensive Information (COBI Scores of ) Nil None Significant Information (COBI Scores of 61-80) Nil None Some Information (COBI Scores of 41-60) Nil None COBI Minimal Information (COBI Scores of 21-40) 1 Nakuru Name of Counties Scant Information (COBI Scores of 0-20) 9 Taita Taveta, Kisumu, Bungoma, Nyeri, Machakos, Nandi, Kajiado, Mombasa, and Garissa Given that COBI scores are based on public availability of four budget documents, Nakuru comes out tops because it publishes both CFSP and CBEs online. The three lowest ranked counties Garissa, Mombasa and Kajiado, do not avail any of these two budget documents to the public, whereas the other seven posted their CFSP on their respective websites. None of the ten counties publishes the other two remaining budget documents assessed in the survey; that is, the County Appropriations Act and County Finance Act and the Quarterly Financial Statement, and hence reason for the low average COBI score. 3.2 Availability of Budget Documents and Mode of Distribution The following section specifically discusses availability of each budget document, the mode of distribution and findings on whether they were released in a timely fashion. County Fiscal Strategy Paper: The County Fiscal Strategy as the pre-budget statement is made available to the public by seven of the ten counties (Kisumu, Nyeri, Nandi, Bungoma, Nakuru, Taita Taveta and Machakos) via their own websites as earlier mentioned. This document is not publicly available in Kajiado, Mombasa and Garissa. According to a Mombasa County officer, this document was still at the precincts of county assembly at the time we were conducting this survey. In their own understanding, the officer mentioned that this document could only be shared to the public upon its approval by the county assembly, contrary to the provisions of the PFMA, Insofar as the other two counties are concerned, despite making numerous attempts to obtain this document, we did not get any response or explanation worth noting as to why this document is not accessible. In order to ensure that the county assembly has ample time to review, discuss and provide their inputs for the CFSP, the PFMA, 2012 stipulates that it be approved by the County Executive Committee and submitted to the county assembly by 28th February. Consequently, this document must be made available to the public within seven days after being tabled in the county assembly. The score on timely release of this document to the public, at least four months prior to the beginning of the financial year, shows that all the counties, except Nyeri, adhered to this legal requirement via their websites. Nyeri managed to do so at least two months prior to the start of the financial year. 21

23 County Budget Estimates: Are considered as the Executive Budget Proposal at the county government level. Despite the importance of this document, Nakuru is the only county that made this document available online via its website. In Nyeri, Bungoma, Garissa, Nandi, Kajiado and Machakos counties, this document was only available in response to formal requests for access, but not within the confines of the definition of publicly available, as in most of this it took more than ten days. Secondly, the introduction letter used by our enumerators was a major influence to obtaining this document. The public has no access to this document in Mombasa, Taita Taveta, and Kisumu counties as it is produced for internal use. The case for Kisumu is, however, rather interesting as the availability of this document is limited to those who attended a public hearings in May where the document was disseminated to participants. It is worth noting that in Nyeri and Kisumu, this document is accompanied by supporting documents as required, as per Section 130 (a) of the PFMA. In the case of Nyeri, this document is titled Nyeri County Highlights of the County Budget 2014/15 and Medium Term Expenditure Framework (MTEF): April 2014 and Nyeri County Budget Summary for the Fiscal Year 2014/15. On the other hand, Kisumu, their budget estimates were accompanied by budget notes as supporting documents as prepared by individual ministries/department. The date for the release of County Budget Estimates to the county assembly is set in the PFMA, 2012, that is, not later than 30th April each year. Our assessment of the ten counties shows that they all adhered to this legal provision. Box 1: Case study of how county governments can partner with civil society to produce citizen budget Nakuru County Government and a CSO from Nakuru by the name Centre for Enhancing Democracy & Good Governance (CEDGG) produced a Citizen Budget 2014, but after the Nakuru County Budget Estimates 2014/15 had already been approved, in July. We sincerely appreciate your interest in our work towards promoting budget transparency in the counties. This work is anchored in the Governance and Devolution Programme whose objective is to promote participatory issue-based decisionmaking processes in the devolved system of government. 1. The need for a Citizen Guide for the budgets was occasioned by our experience in trying to enable residents to monitor public projects and expenditure. It is clear that, in order for citizens to conduct budget monitoring, they need to understand the budget as well as other planning documents, including the CIDP, Annual Plans, County Fiscal Strategy Paper, County Budget Review and Outlook Paper (C-BROP), etc. However, this is often marred by the fact that most of these documents are voluminous and in most cases technical, making it difficult for majority of citizens to internalize them. The preparation of the guides, therefore, was meant to simplify the budgets for citizens to enable them monitor their implementation and execute their oversight role more effectively. We were also able to prepare popular versions of the CIDP for Nakuru and Baringo Counties. 2. CEDGG has adopted a collaborative approach in her work. Through engagements with budget and planning units of the two counties, the need to create awareness on the CIDP and budget was agreed upon between us and the units. We offered to facilitate the formulation of the guides. 3. The guides have been disseminated to various stakeholders, but mostly residents associations, community oversight committees, and civil society forums in the counties. 22

24 Table 3: Summary results from the ten counties on the availability of eight key budget documents Which of the eight key budget documents does the public have access to? Publication status by County Key Budget Document Kenyan equivalent document Description of documents Kisumu Bungoma Nyeri Garissa Mombasa Taita Nandi Hills Machakos Nakuru Kajiado Taveta Pre-Budget Statement County Fiscal Strategy Paper Provides information that links government policies and budgets and typically sets forth the broad parameters that will define the budget proposal that is presented to the legislature Published Published Published Published Published Published Published Published Published Produced for internal use Executive Budget Proposal County Budget Estimates Presents the government plans to raise revenues through taxes and other sources and spend these monies to support its priorities, thus transforming policy goals into action Produced for internal use - only disseminated during public hearings Published Published Published Produced for internal use Produced for internal use Published Published Published Published Citizen Budget Citizen Budget A non-technical presentation to enable broad public understanding of a government s plans for raising revenues and spending public funds in order to achieve policy goals Not produced Not produced Not produced Not produced Not produced Not produced Not produced Not produced Not produced Not produced Enacted Budget County Appropriation Act and County Finance Act The legal instruments authorizing the executive to raise revenues, make expenditures, and incur debt Published only Appropriation Act 2013 Published only Appropriation Act 2013 Published only Appropriation Act 2013 Published only Appropriation Act 2013 Produced for internal use Produced for internal use Produced for internal use Not published Published only Appropriation Act 2013 Not published 23

25 Key Budget Document Kenyan equivalent document In-Year Report County Quarterly Financial Statementr Mid-Year Reportl County Mid-Year Report Year-End Report County Annual Financial Statement Audit Report County Audit Report Description of documents Periodic (monthly or quarterly) measures of the trends in actual revenues, expenditures, and debt, which allow for comparisons with the budget figures and adjustments Summarizes the actual budget data for the first six months of the year (revenues, expenditures, and debt), reassesses the economic assumptions upon which the budget was initially drafted, and adjusts the budget figures for the remaining six months accordingly Information comparing the actual budget execution relative to the Enacted Budget Independent evaluation of the government s accounts by the country s supreme audit institution. It typically assesses whether the executive has raised revenues and spent monies in line with the authorized budget, and whether the government s accounts of its revenues and expenses are accurate and provide a reliable picture of the fiscal situation Kisumu Bungoma Nyeri Garissa Mombasa Taita Not used in the survey Not used in the survey Not produced Not produced Not produced Not applicable during this survey period Not produced Not produced Taveta Not produced Nandi Hills Machakos Nakuru Kajiado Not produced Not produced Not produced Not produced 24

26 Enacted Budget (EB): The Enacted Budget is comprised of both County Appropriation Act and County Finance Act. Five counties, namely: Kisumu, Bungoma, Nyeri, Garissa, and Nakuru have only one set of the EB, i.e., the Appropriations Act which is published in an online portal for legislations called Kenya Law Reports. Apparently, the second set, Finance Act, was not available to the public at all. In the remaining counties the EB was produced for internal use and was therefore not publicly available. Annual Financial Statement: PFMA provides that no later than three months after the end of the financial year that county government shall submit the annual financial statement to the Office of the Auditor General as the Year-End Report. The survey shows that this document is not published for public use by any of the ten counties. As such, they all score zero on questions County Audit Report: Since county governments came to effect in March 2013, evaluation of the Audit Report was not relevant since first audit report 2013/14 will only be released in early Other Key Budget Documents Not Assessed: The survey did not have any questions assigned to assess public availability and comprehensiveness of three important budget documents, namely: Citizen Budget, Mid-Year Report and Quarterly Financial Statement (In-Year Report). Although it is not a legal requirement, but good budgeting practices require government to produce a simplified and nontechnical version of the County Budget Estimates to enhance uptake and overall engagement by the public. None of the ten counties produced and published a Citizens Budget. In this area, counties can learn from the national government that produced and published its first Citizen Budget of the National EBP in 2011/12. Equally, the law is silent on requirement for Mid-Year Report. As for the Quarterly Financial Statement, although county governments are legally required to produce, this survey did not include any questions on the same. We anticipate that subsequent scale up of SNOBS will include questions to assess these three documents. Although some county governments are still setting up new structures or upgrading, majority have websites which are working at varying levels; posting a document online is by far the easiest and most cost-effective way to begin to make it publicly available to the members of the public. After all, internet penetration is slowly but surely increasing in the urban centres and even in the rural communities in Kenya. It worth noting that, of all the published documents by the ten counties surveyed, CFSP and the CBE were only available in English as the official language, and that none of the counties went the extra mile to publish them in the national language of Kiswahili for wider outreach. Views from a Nakuru County Respondent Accessibility of budget documents should be centralized in one place. Customer care and complaint service should be harmonized and made user-friendly. The county websites should be furnished with budget information and documents. Customer care contacts should also be made functional for improved access to information. 25

27 3.3 Form and Structure of Public Consultations in the Budget Process is Wanting. PFMA, 2012 provides for county governments to establish County Budget and Economic Forum (CBEF), a formalized platform for public consultations on all matters of county planning and budgeting. This survey shows that three counties, Taita Taveta, Machakos and Kajiado have all established CBEF for public consultation and perspective on budget priorities. For example, in Taita Taveta, the CBEF was established on 15th March, For this and the other remaining counties that have not established CBEF, they all understand that public participation is a constitutional requirement. The challenge again was that we did not manage to confirm on the regularity of consultations. Legally, consultations are expected in their design to be as representative as much as possible but in reality this is not always the case. For example, the law provides that the CBEFs include diverse public representation from various groups, including professionals, business, labour issues, women, persons with disabilities, the elderly and faith based groups, etc. However, for some counties, the process of constituting CBEFs by the County Governor is not always transparent. Counties that have already established CBEF have a head start in ensuring that the consultation process is representative. On another point, the consultation processes in some counties is seen as a mere public relations exercise; for example, in Kisumu, in 2014/15 some of the budget formulation forums were held in few pre-selected areas within the sub-county. The scope of consultation and discussions is primarily focused on development expenditure within the discretionary budget in virtually all the counties, with the exception of Machakos where the discussion involved all elements of expenditure and revenue. Granted that majority of the public are concerned more about developmental issues, however, budget consultation fora should allow for discussion on recurrent expenditure issues such as wages and salaries and the revenue side of the budget and the linkage thereof. Upon submission of the County Budget Estimates to the County Assembly, the Budget and Appropriation Committees organize public hearings regarding the draft budget before it is approved. All the ten counties held pre-budget hearings but testimony from the public was limited. As noted in some counties such as Kisumu, sometimes the reason for limited public testimony is inadequate capacity to discuss budget estimates. See the chart on the next page. 26

28 3.4 Information on Services Offered by Counties Not Available at a Single Source For the citizens to exercise oversight on service delivery, they need information on the range of services offered at the county as well as information on the basis on which services are provided and whether available to all for free or to a segment of the population or at a fee. Bungoma and Kisumu are the only counties that provide information on the range of services offered consolidated at a single source through services charters. For the other counties, this information is available at each responsible department or agency. Equally, it is these two counties that provide full information on conditions for entitlements at a single source. On whether these services are responsive to citizens need, all counties have indeed established mechanism for registering complaints or comments available at varying sources 27

29 3.5 Announcement of Procurement Tenders Made Public but Limited Information on Award of Tenders and on Procurement Complaints Decisions In Kenya, public procurement consumes over 65% of the national budget, including local government procurement. It consumes 45% of the national government budget, excluding local government procurement 6. The close relationship between procurement and development demonstrates that there is need for transparency and accountability in the manner in which procurement is conducted 7. Secondly, as documented, procurement matters because public resources are often amenable to poor management. Besides, Article 227 of the Constitution and Public Procurement and Disposal Act, 2005 require public procurement to be fair, equitable, transparent, competitive, and cost effective. Good budgeting practices require that sub-national levels of government prepare annual procurement plan, a document that shows all the purchases of goods and the contracting for services for the forthcoming year. This document although initially produced for internal use, citizens should have access to it, consequently to monitor transparency and fairness of the public procurement procedures. 6 Centre for Governance and Development and National Tax Payers Association. Citizens Guide to Public Procurement: Public Procurement Procedures for Constituency Development Funds 7 Ibid. 28

30 On a positive note, though, all the counties surveyed made public information on the launch announcement of procurement tenders through a set of different sources, including online such as county websites, newspapers, and on radio. In contrast, information from the ten counties on the award of public tenders from a regular known location or a set of sources was limited. In two counties, Taita Taveta and Garissa, some department and agencies publish this information. Further, as a way of ensuring transparency and fairness, procurement law provides for dispute resolution mechanism that can be used by the public to register complaints or, if necessary, appeal against the award of a tender. In majority of the counties, the independent procurement complaints body does make public decision on any registered complains. Nonetheless, these decisions can be accessed by the public from procurement entities within 30 days in Bungoma, Kisumu, and Taita Taveta. 3.6 Assessment of Comprehensiveness of County Fiscal Strategy Paper and County Executive Budget Estimates In addition to assessment of public availability of budget documents, comprehensiveness and relevance of the budget information is equally important. This section, therefore, assesses the comprehensiveness of the County Fiscal Strategy Paper and County Executive Budget Estimates produced and published by the ten county governments. On average, CFSP provides more information relative to CBE; but overall level of detail in county budget information is below average. The average sub-score for CFSP is 48.5 out of 100, and that for CBE is 37.9 out of 100, albeit the latter being associated with more questions, twelve, versus two for CFSP. Figure 3 below captures the average sub-scores on the level of detail for the published CFSP by counties based on the two questions specific to comprehensiveness of CFSP. Seven counties scored above the average sub-score of 48.5 out of 100; and three counties, Kajiado, Garissa and Mombasa scored zero because CFSP was not accessible at all. Figure 3: Average score on comprehensiveness of CFSP by counties Source: Author s compilation. 29

31 For the seven counties, sub-scores for the CFSP varied considerably, with Nyeri leading the pack with a score of 100 as it is the only county that presents extensive presentation of key fiscal and macroeconomic policy information as well as information on expenditure policy priorities to guide the development of detailed budget estimates. Overall, these seven counties provided more comprehensive information on expenditure policy priorities than fiscal and macroeconomic policy information. It seems that this score is perhaps a reflection of the impressive national score on the Budget Policy Statement (79 in the last Open Budget Survey 2012) which is the equivalent of the CFSP at the county government level. We noted that all the seven counties borrowed heavily from the BPS in terms of structure and expected content but also due to the fact that, logically and legally, CFSP are supposed to be aligned to the BPS. It is important to mention however that, according to the PFM Act, although the CFSP is supposed to be submitted to the county assembly by 28th February for debate and approval, there may be challenges in the extent to which it will be aligned with the BPS. This is because the BPS is supposed to be submitted to parliament on 15th of the same month, and any delay in its approval may delay the release of the CFSP or impact its thorough preparation and comprehensiveness. After completion of the actual survey, we noted that we had omitted the issue of information on expenditure ceilings in CFSP and hence the high scores on the comprehensiveness of this document. We therefore note that, future SNOBS should include a question to test or assess whether CFSP provide this information that is useful in guiding preparation of county budgets, and the boundaries they should observe. County Budget Estimates lack important details Figure 4 below shows that, for five out of ten counties, sub-scores for County Budget Estimates were above the average sub-score of 38. Only Taita Taveta and Nakuru counties, with similar scores of 54.5, provide slightly more than half of the desired level of information contained in the County Budget Estimates. In contrast, low CBE sub-scores for Kajiado and Mombasa indicate that there is scope for both counties to improve comprehensiveness of this document. To get a clearer picture of the level of detail in the CBEs, it is useful to look at the results of the set of 12 questions. 30

32 Figure 4: Score on comprehensiveness of County Budget Estimates Source: Author s compilation. Mixed results on classification of county budget information All the ten counties surveyed, with the exception of Mombasa, present expenditure information for budget year 2014/15 classified by administrative unit. Information on expenditure classified by administrative unit is important because it indicates which county government ministry/department/directorate is responsible for spending the funds assigned to it, and which should therefore be held accountable for their use. However, the extent to which counties present information on expenditure by economic classification is much more limited. That is, by the nature of expenditure such as funds to be used to pay salaries and wages, acquisition of goods and services, transfers, capital projects and so on. Nakuru, Taita Taveta, and Nyeri are the only counties that had all its expenditure by economic classification attributed to their adoption of Programme Based Budgeting. Seven counties presented further disaggregation of expenditure by programme level. For example, the ministries under the health ministry/department showed budget for hospitals, dispensaries, and so on. Related to this is presentation of expenditure for the year preceding the budget year, in this case 2013/14 (prior budget year), which is demanded as good budgeting practice. Besides, this makes comparisons easy. Results show that majority of counties do not fare well in this area; only Bungoma, Nakuru, and Nandi present expenditure by administrative and economic classification for prior budget year. Further, Kisumu, and to some extent Taita Taveta and Mombasa, featured some information on value of transfers to sub-county levels such as constituencies and wards. On the revenue side, all counties provided information on all the different sources for the budget year, including their own sources such as market fees, land rates, parking fees, and transfers from the national 31

33 government. Indeed, a majority of these counties presented information on the value of transfers from the national government in a disaggregated manner, including equitable share; and for some counties like Kisumu, presented information on conditional grants for level 5 hospital and donor grants. Three counties, Mombasa, Garissa, and Nyeri presented only major sources of revenue. Budgets for six counties contain only a single year of revenue estimates contrary to PFM Act 2012 requirement that this information aggregate revenue for at least two years following the budget year (multi-year perspective). Counties show a good attempt at comparing revenue estimates for 2014/15 with estimates for prior year 2013/14 in the same format as the budget year. The culprits in this case are Nandi, Mombasa, Garissa, and Machakos, which do not provide any estimates of revenue for prior budget year. As far as donor support is concerned, whether grants channelled through county budget system of project grants earmarked for a specific project, not much information is provided. This was not an easy question to assess as getting information on whether the counties surveyed will benefit from donor support was not forthcoming. Budget information on critical issues such as fiscal deficit/surplus, financing sources and stock of debt, if any, is widely unavailable in virtually all the county budget estimates of the surveyed ten counties. The counties were encouraged by the Office of the Controller of Budget (OCoB) to avoid budget deficits during the three-year transition period from 2013, but many counties have deficits. Sub-score results show that, only four counties present explicit information on planned deficit and the rest present implicit information, excluding Machakos that that does not show any information. Apparently, none of the ten counties showed information on how they planned to finance any deficits or information on the use of finances from surplus. As far as data on stock of debt is concerned, it is only Kisumu that captured some information, but again this issue is not clear. The other remaining counties may not have any stock of debt in the first place, but then again it is not clear whether counties inherited debt owed by the former local authorities or whether this is supposed to be written-off by the national government. Missing information on complementary and non-financial information Just like at the national government, there is little or no information at all on extra budgetary funds and expenditure arrears. The same applies to missing information on contingent liabilities, if any, that is for things like loans guaranteed by counties. Further, financial assets such as bank deposits or shareholding owned county government information is hard to come by; but the picture is marginally better in terms of availability of information on non-financial assets such as buildings and other physical assets, as captured by Taita Taveta and Nakuru in their budget estimates. This kind of information is critical in understanding the financial position of the county and how this will affect revenue and spending plans. It is also important to mention that, there is confusion as the process of transfer of assets and liabilities under the Transition Authority is ongoing. Budget narrative information highlighting the link between individual major policy goals is evident in Bungoma, Nandi and Machakos County Budget Estimates, but this information lacks some details. Without this kind of information, whoever is engaging in county budget analysis and advocacy work or reviewing county budget would be at a loss in understanding the basis of expenditure priorities and choices. Related to this is information on non-financial performance targeting, such as the number of targeted beneficiaries for a cash transfer programme. Counties like Nakuru, Nandi, and Garissa that have adopted PBB, present this kind of information in majority of their programmes. This kind of information is often also captured in supporting documents to the County Executive Budget Proposal. 32

34 4.0 Dissemination of Preliminary Results For impact on county budget transparency, the project and its preliminary results and recommendations were disseminated in each of the ten counties from 23rd February to 3rd March, However, due to logistical challenges, we held dissemination of both Nakuru and Nandi separate results in one venue, Nakuru County on 26th February, We targeted a similar set of target audience but the actual attendance varied from one county to another. What was common in all the counties is that there was a good representation of CSOs but the number of county government officials and county assembly members varied significantly. However, media coverage was in four out of the ten counties, with radio comparatively the most used medium. For further details on dissemination, see Annexe 3. 33

35 5.0 Conclusions and Recommendations There are a number of actions that can be taken based on COBI findings to improve county budget transparency in Kenya. These include: County governments should publish and publicize CBEs, County Appropriation and Finance Act and YERs already produced for internal use at no extra cost. County governments provide the CFSP document online. As such, they should note that the simplest step to advancing budget transparency is to also post the CBEs online just like they do for CFSP and what Nakuru did; after all, for Mombasa and Taita Taveta, this document is already produced for internal use. In addition, the public and the media should remind some of the counties, like Kisumu, that it is a legal requirement that these documents are availed to the public as soon as they are tabled in county assemblies as some of the officers have the notion that these documents can only be availed to the public upon approval by county assemblies. County governments should develop and publish Citizen Budget. Counties have an opportunity to partner with CSOs (e.g., Nakuru) or they can take the initiative. Besides they should consult with the public on the content. County governments have an opportunity to increasing county budget transparency by producing Citizen Budget singly or through partnership with civil society organizations. There is a clear evidence of how the county of Nakuru partnered with CEDGG, a CSO that developed CB based on the county s CBEs. It is also important to note that this document must be released on time, which is as soon as the CEBs or accompanied with CBEs for it to be useful. Similarly, county governments can also learn from the national government through the Treasury that has now for the last three financial years produced their version of Citizen Budget. Counties can improve comprehensiveness of CBEs by focusing on providing information on revenue, expenditure and other fiscal information, including disaggregated information, multiyear projections, and so on. As assumption that counties publish CBEs, its assessment as well assessment of published CFSP show that, altogether, counties provide on average less than half of the desired level of information. Specifically, information or details on revenue, expenditure, debt and for other related issues such as contingent liabilities is either missing or minimal, and therefore counties have a mandate to capture this information to, not only enable the citizens effectively engage in the budget process, but to overall improve budget transparency. Counties should adopt PBB and in turn improve information on budget narrative, non-financial performance targeting, etc., to allow for effective budget analysis. County assemblies, the public and the media should put pressure on counties to ape Nakuru, Nandi and Garissa that have adopted PBB. This has an effect on improving complementary and non-financial information. 34

36 Budgets are poor on outputs and outcomes, need to provide link between counties policies, spending and macroeconomic policy. Counties should establish CBEFs in a transparent and integral manner as a way of formalizing public consultations and inclusivity. Establishment of formalized mechanisms for public consultations, as provided for in the law, including platforms such as CBEFs is critical, alongside availability of budget information, for effective public consultation and participation. It is only three counties, Taita Taveta, Machakos, and Kajiado that have so far established CBEFs; and as such, the other counties should follow suit. Citizens, media and members of county assemblies should exercise their oversight roles to ensure that counties uphold transparency and integrity in composition and constitution of these forums. 35

37 References 1. Fjeldstad, O. H., Decentralization and corruption: A literature review. Ulstein Anti-Corruption Research Centre (U4). Chr. Michelsen Institute, Bergen (July). 2. Kroth, V., Subnational Budget Transparency: An analysis of Ten Pilot Studies. International Budget Partnership-Measuring Subnational Transparency, Participation and Accountability. Open Budget Initiative. 3. Republic of Kenya. The Constitution of Kenya, Republic of Kenya. The Public Finance Management Act, Centre for Governance and Development and National Tax payers Association, Citizens Guide to Public Procurement Procedures for Constituency Development Funds 6. Republic of Kenya (2012) County Government Act, Republic of Kenya (2011) Urban Areas and City Act, Open Budget Reports (2010 and 2012) via budget-initiative/open-budget-survey/ 9. General reports via Republic of Kenya (2014) County Government of Kisumu Budget Estimates Fiscal Year 2014/ Republic of Kenya (2014) County County Government of Machakos Budget Estimates Fiscal Year 2014/ Republic of Kenya (2014) County County Government of Kajiado Budget Estimates Fiscal Year 2014/ Republic of Kenya (2014) County County Government of Nyeri Budget Estimates Fiscal Year 2014/ Republic of Kenya (2014) County County Government of Garissa Budget Estimates Fiscal Year 2014/ Republic of Kenya (2014) County County Government of Mombasa Budget Estimates Fiscal Year 2014/ Republic of Kenya (2014) County County Government of Nandi Budget Estimates Fiscal Year 2014/ Republic of Kenya (2014) County County Government of Taita Taveta Budget Estimates Fiscal Year2014/ Republic of Kenya (2014) County County Government of Nakuru Budget Estimates Fiscal Year 2014/ Republic of Kenya (2014) County County Government of Bungoma Budget Estimates Fiscal Year 2014/ Republic of Kenya (2014) County Republic of Kenya (2014) County Fiscal Strategy Papers for various counties including: Taita Taveta, Kisumu, Bungoma, Nyeri, Machakos and Nandi 36

38 Annexe Annexe 1: Summary of Questionnaire and Calculation of COBI SNOBS Questionnaire and Computation of COBI The results for each county are based on 70 questions in the SNOBS questionnaire completed by enumerators and peer-reviewed by IEA. Each question has multiple-choice answers/responses (a)-(d) and an additional answer (e) for questions that are not applicable. Response (a)describe a situation that represent best practice regarding type of budget information or practice; response (b) describe, second best; response (c) is meeting minimal efforts or standards; and response (d) is where standards are not met at all. There are, however, a few questions with three multiple-choice responses; for example, question 12 on whether the County Budget Estimates, as the Executive Budget Proposal, or any supporting documentation, presents estimates for the aggregate level of expenditure for a multi-year period (at least two years beyond the budget year). Response (a) says Yes, multi estimates of aggregate expenditure are presented generally assesses whether standards are met; response (b) says No, multi estimates of aggregate expenditure are not presented generally assesses whether standards are not met; and the third response (c) says Not applicable/other checks whether the question is applicable. Furthermore, each question provides a section for the enumerator or researcher to make citations of the law; budget document referred to or interview with county government officials in answering the question as evidence for the response selected. In follow up to this, is a section for comments which the enumerator can use to provide further explanation or justification to support their response. Beneath these sections, is one for peer reviewers comments to concur or contest, with clear justification, the enumerator s selected response, and in turn what the reviewer thinks is the correct answer. Once the questionnaire is completed, the agreed upon responses after peer review are quantified to compute COBI which is a point scale. For each question, the numeric scores attached is as follows: Five response questions Three response questions Response Score Response Score a 100 a 100 b 67 b 0 c 33 c Not counted as part of aggregated score d 0 e Not counted as part of aggregated score Based on a simple average, the aggregate score, which forms the COBI, is generated by adding up all the scores for each of the 70 questions and computing the average score. 37

39 Research Process The six enumerators embarked on data collection and completion of SNOBS questionnaire between June and August 2014 in ten counties that were surveyed for this project. The first leg of the research process was conducted in six counties, an enumerator per county, and the remaining four counties were subsequently reviewed by four out of the six enumerators. This process was preceded by development of the survey questionnaire, selection and recruitment of survey team, their training and pre-test of survey tool as explained below. The timelines are also indicated. Development of Survey Questionnaire The IEA, held a brainstorming meeting on 30th and 31st January, 2014 to lay the ground for the actual survey in June. In this meeting, we engaged a consultant from Tanzania who was the lead consultant on a similar project in Tanzania in The aim of the meeting was for the consultant to advice on the design of the questionnaire and also shed light, based upon his experience on things to look out for during the implementation of this project. Upon presenting and sharing SNOBS results from Tanzania, and discussion that ensued, we took notes of things to factor into the design of the questionnaire, recruitment of enumerators, their training and preparation for the actual field work. Some of the salient issues the consultant shared were on the challenges to expect, and how they were addressed in Tanzania, including issues of quality control and how to leverage on dissemination as an advocacy strategy. We learned that the International Budget Partnership (IBP) has already designed a sub-national open budget survey questionnaire. Upon review of this questionnaire, it was agreed that it made sense to adopt and customize it to the context of the county budget process and system instead of developing one from the scratch. Survey Team - Selection and Recruitment The survey team was comprised of six individuals with diverse academic backgrounds and qualifications ranging from sociology, economics, project planning and management, business management, but more importantly with varying experience on research and engagement with civil society organizations. One of the enumerators is an IEA staff, a research assistant, and the other five enumerators were recruited mainly using networks and our coalition partners and from IEA databases of researchers that we have engaged in the past. This process started in March The recruitment was competitive, and by early April we had received a number of applications which we evaluated internally. Quality controls were applied to ensure that we selected a team of five enumerators with adequate experience to conduct this kind of survey. To reduce transaction costs, and to leverage on familiarity, we ensured that the enumerators recruited resided in at least one of the ten counties and that the second county in which they conducted the survey bordered their own. In this regard, we had the IEA research assistant conduct the SNOB survey in Bungoma and Kisumu, with three enumerators surveying, respectively, Kajiado and Machakos, Nakuru and Nandi, Mombasa and Taita Taveta, with the remaining two enumerators only had one county apiece, Garissa and Nyeri. The Public Finance Management team at the IEA conducted pre-field training and a mid-field feedback session as well as overall supervision and backstopping of the survey exercise. 38

40 Training of Survey Team The IEA, through the head of the Public Finance Management Programme, conducted the pre-field work training on 19th May. In attendance were the five enumerators that we recruited, and PFM team of three of which one of the IEA staff was enjoined as the sixth enumerator. The basis of the training was a draft questionnaire that was largely borrowed from the International Budget Partnership (IBP) as pioneers of a similar exercise in Brazil, Indonesia and Tanzania. The IEA team reviewed this questionnaire and added a column to each question that elaborated the question further and indicated the budget documents, local institution and the relevant legislation governing the particular question. Pre-test Field Work, Feedback and Review of Pre-test A week after training the researchers, the IEA invited them to pre-test the survey toolkit with an objective of using the experience thereof to inform revision of the survey if necessary, as well as give the researchers an opportunity to experience the actual survey process. The pre-test took place on 27th May and was preceded by a half day meeting to recap on the earlier training and get a feel from the researchers on any outstanding questions they may have had. The IEA through the PFM team divided the researchers into two groups and tasked them to select a few questions of their choice and to use them to engage county officials in Nairobi and the neighbouring Kiambu County. Given that the researchers were all converged in Nairobi, it made sense, at least logistically, to select these two counties for the pre-test and also due to their proximity. On completion of the pre-test, we gave the researchers an opportunity to give us feedback on their experience, challenges and any lessons they may have picked. From the sharing, the IEA noted that it was very important they identified the right officers to interview as was explained by the group that went to Kiambu, for example, the County Budget Officer who is perhaps more versed with issues surrounding the budget process and the documents produced than other county officers. The other issues that arose included the need for introduction letters and prior booking of appointments to avoid delays and disappointments. 39

41 Annexe 2: Synthesis of Field Reports This brief report captures some common issues noted by enumerators from the ten counties they visited between June and August As part of the experiences they encountered, this brief notes challenges with regard to accessing budget documents and in seeking appointment with relevant county officers for interviews. In the process of this experience, the enumerators noted a few positive or innovative plans that some counties had with regard to dissemination of budget information. In their attempt to gather useful information to enable them complete SNOBS questionnaire, the enumerators interviewed various county government officers and citizens. Before they conducted these interviews, the first thing that all enumerators did was to check whether the key budget documents were posted online in each respective county s website. Based on the information they found on the website, they made calls and send s to respective county government officers seeking information that was not available on the website. There are certain questions that needed responses and/or evidence from county officers. In certain other instances, enumerators needed to follow up on some questions for clarification with these officers. The officers interviewed vary from one county to another and also based on their availability, but largely most of those interviewed included: County Executive Member for Finance, County Finance Officer, County Budget Coordinator, Chief Accountant, and Director Supply Chain, among others. Further, most enumerators also interviewed members of County Assembly for specific Committees, such as the Budget and Appropriation Committee, as well as staff such as Clerk of the County Assembly. For some questions on public consultations in the budget process, enumerators interviewed in some counties few members of County Budget and Economic Forums, random citizens or participants in public hearings, and so on. On the issue of service deliver oversight, enumerators had to visit county government offices and at least random schools and health centres to confirm by observation whether there was information for the public on the services offered and facility level budget information, respectively. Challenges In conducting the survey, the major challenge was difficulty in accessing budget documents from the relevant county officials. Majority of enumerators concurred that, despite prior communication, it was challenging to meet with relevant county officers as most claimed to have busy schedules or in meetings out of the office. Related to the above challenges, majority of the county officers were suspicious of the whole exercise and hence reluctant to cooperate and avail information that was requested by enumerators. For example, as reported by the media of the tussle between the county assembly and the county executive on the content of the County Budget Estimates, and other budget related information, the SNOB survey was mistaken for witch-hunt. 40

42 There were cases of officers who were not cooperative. For example, officers dealing with procurement were not forthcoming with information, and in other cases the budget officers easily warded off citizens by directing them to their county websites. Further, enumerators noted that some members of county assembly did not show up all together despite prior appointment. At the lower/facility levels, some county officers expressed fear of releasing health and school facility level budget information without authority from their seniors. Some claimed that they knew nothing:-...i was only asked to show them land for construction of the ECD classes Head Teacher, Mwariki Primary School. Administrative obstacles, such as bureaucratic nature of county government operation, and logistical ones such as incomplete or missing resource centres in some counties, was also cited as a hindrance to accessing budget information. Other general challenges are as follows: From their experiences, some enumerators concluded that it is difficult for the ordinary person to access budget documents. Some enumerators noted that, although they ultimately obtained information that they had requested upon numerous follow up, this was rather too late to be useful for an exercise or purpose such as SNOBS. There was an observation made by some enumerators with regard to public consultation forums at the counties. Respondents, who attended public hearings, indicated that they are characterized by short notices, 2-3 days, and elite capture as mobilization is done through national newspapers and through MCAs. Positive Plans Despite the above-mentioned challenges, some enumerators noted a few positive things or plans. Although meeting with officials featured largely as a challenge for most enumerators, this was not the case in Mombasa and Taita Taveta, as the enumerators were not required to make prior appointments. All they needed to do was to walk into the county offices and request to see the officer they are interested in. For example, in both Nandi and Nakuru, there was confirmation of advanced plans to prepare Consolidated County Procurement Plans, but it was not clear whether this will be shared with the public. 41

43 Annex 3: Synthesis of Dissemination Report This report is in three parts. The first part gives a summary of the objectives of the dissemination of the preliminary findings of SNOBS in all the ten counties. The second part of the report is a table showing the general details of the dissemination with regard to date and venue of the dissemination in addition to the nature and type of the participants who attended the dissemination of the preliminary findings in each of the counties; while the last part captures all the salient issues that came up from the dissemination in all the counties. Objectives of the Dissemination Forums in All the Ten Counties. The dissemination of the preliminary findings of SNOBS in all the ten counties was aimed at: 1. Sharing the preliminary findings of the SNOBS in each county. 2. Providing the county government with the opportunity to interact with the preliminary findings of SNOBS and therefore interrogate the same. 3. Increased capacity of citizens in the counties to know their role and to participate in the budget making process. 4. Educating the civil society and stakeholders working in these counties on the county budget making process and the existing opportunities for public participation. 5. Improved county government public finance management systems. 6. Providing opportunity for networking and creating synergies among the CSOs working in these counties 42

44 Table on dissemination of Sub-National Open Budget Survey (SNOBS) preliminary findings in the ten counties County Date Venue No. of Participants Kisumu 25th February, 2015 The Royal City Hotel, Kisumu 59; largely CSOs, all County Chief Officers and media representatives Nakuru and Nandi 26th February, 2015 Waterbuck Hotel, Nakuru 60; majority were CSOs and FBOs Mombasa 23rd February, 2014 Castle Royal Hotel, Mombasa 70; majority were from CSOs, County Controller of Budget, media representatives, youth leaders, county officials from finance department Nyeri 27th February, 2015 Green Hills Hotel, Nyeri 68; largely CSOs, County Finance & Planning officials and 3 MCAs Bungoma 24th February, 2015 The County Comfort Hotel, Bungoma 73; majority were CSOs, FBOs and County Planning officials Machakos 27th February, 2015 Tea Tot hotel, Machakos 55; largely CSOs, FBOs, Chair County Budget & Appropriation Committee Kajiado 3rd March, 2015 Leleshwa Inn, Kitengela 28; majority were CSOs and FBOs Garissa 26th February, 2015 Nomad Palace Hotel, Garissa 60; majority were CSOs, 6 Members of County Budget & Appropriation Committee, County Finance & Economic Planning officials, CBEFs representative Taita Taveta 24th February, 2015 Lavender Garden Hotel, Wundanyi Note 71; largely CSOs, County Governor, Deputy Governor, 6 CEC members, and FBOs In all the ten counties, there was adequate representation from the county government, except in Kajiado, Nakuru and Nandi. However, it was only in Taita Taveta County that the Governor, his deputy and members of the cabinet attended the forum and actively engaged the people on the issues raised by the study. This is captured in the picture on the next page. 43

45 In the front row, the Governor of Taita Taveta, His Excellency John Mruttu (left) and his deputy Ms. Mary Ndegwa Kibuke (right). Others in the photo are cabinet members and members of public attending the dissemination forum at Lavender Garden Hotel, Wandunyi Town. Cross-cutting issues arising from the forums Below are the salient issues that came up from the dissemination forums in all the counties: Public Participation Representative, structured, standardized and sustainable public participation to be anchored in law strengthening citizen role in the process. Concerns on the absence of legislation and regulations on what is public participation and on what do the public participate in. Poor or limited public participation is attributed to the communication approach adopted by the counties. Public participation is left in the hands of ward administrators, most of whom are non-residents hence not in sync with the local trends and dynamics. There is need for the county to change mode of communication and embrace local mechanism. Organizers of public fora must be alive to the local dynamics (market and tea buying days) if effective public participation is to be achieved. Civic education should precede public participation. 44