KENYA TRANSITION INITIATIVE FINAL EVALUATION Final Evaluation Report May 2008 through October 2013 October 2013 This publication was produced for review by the United States Agency for International Development. It was prepared independently by Joel D. Barkan, Njuguna Ng’ethe and Jacqueline Klopp on behalf of The QED Group, LLC, under Task Order #62 of the Program Development Quickly II (PDQ-II) IQC, contract number DOT-I-00-08-00017-00. The author’s views expressed in this document do not necessarily reflect the views of USAID or the United States Government. ii [This page intentionally left blank.] U G A N D A SOUTH SUDAN SOM A L I A T ANZAN I A E T H I O P I A Nairobi Eldoret Isiolo Laisamis Garissa Banissa Buur Gaabo Luuq Shinyanga Jinja Iganga Homa Bay Kondoa Londiani Narok Lokori Arusha Nanyuki Embu Kitui Mwingi Marsabit Nguni Habaswein Girito Wajir Liboi Dif Busia Kotido Kitale Lodwar Magadi Thika Kibwezi Buna Bura Musoma Moroto Katakwi Tororo Singida Todenyang Nyahururu (Thomsons Falls) Maralal Baragoi Moshi Murang'a Taveta Mado Gashi Mombasa Malindi Garsen Tarbaj Hagadera Takaba Lamu El Wak El Beru Hagia Ramu Afmadow Kisii Tot Marigat Lorule Lokwa Kangole Gilgil Kajiado Nyeri Sultan Hamud Same Tanga Hola Bilis Qooqaani Mandera Garbahaarey Jilib Pallisa Migori Webuye Kakuma Kericho Nakuru Namanga Kisima Meru Mtito Andei Kinna Moyale Wete Kilifi Kolbio Dolo Odo Jamaame Nagishot Mbale Solai Sabarei Machakos Kumi Mkomazi Voi Mariakanii Baardheere Kismaayo Nzega Soroti Namalu Babati Lokichar Kelem Naivasha Banya Loiyangalani North Horr Archers Post Tsavo Bu'aale Lokichokio Butere Kisumu Î GEOGRAPHIC INFORMATION UNIT USAID/DCHA/OTI 0 25 50 75 100mi 0 40 80 120 km Î Nairobi Nairobi Kiambu Kajiado Nation -Wide LOC AT IO N S OF K T I A C TIVI T IE S 2 00 8-2 01 3 $800.00 - $50,000.00 $50,000.01 - $100,000.00 $100,000.01 - $250,000.00 $250,000.01 - $1,000,000.00 $1,000,000.01 - $4,000,000.00 Data as of : 6.14.2013 Map Release Date : 6.21.2013 # The boundaries and names used on this map do not imply official endorsement or acceptance by the U.S. Government. OTI Office Partner Office L EGE N D Sources: USAID, USGS, ESRI, GIST iv CONTENTS EXECUTIVE SUMMARY…………………………………………………………….….…vii INTRODUCTION: THE EVALUATION AND THE EVALUATION TEAM...…….….…1 I. COUNTRY CONTEXT AND U.S. POLICY OBJECTIVES ..................................... 3 1.1 Kenya’s Two Decade Struggle for Democracy .......................................... 3 1.2 U.S. Policy Objectives in Kenya ................................................................ 7 II. NATURE AND HISTORY OF THE OTI PROGRAM IN KENYA (KTI) ................... 9 2.1 The Distinctiveness of OTI and its Relationship to USAID ....................... 9 2.2 OTI in Kenya: The Kenya Transition Initiative ......................................... 11 2.3 KTI’s Modus Operandi .............................................................................. 13 2.4 Adjustments in Approach ......................................................................... 17 III. APPROACH AND METHODOLOGY OF EVALUATION ................................... 19 IV. FINDINGS, IMPACT, AND EFFECTIVENESS ................................................ ..22 4.1 Objectives and Strategy .......................................................................... . 22 4.2 Program Characteristics .......................................................................... 23 4.3 Monitoring and Evaluation, the Database ................................................ 25 4.4 Non-Traditional Partners—Rural CSOs and “Youth” .............................. 27 4.5 Criteria for Awarding Grants and Selecting Grantees ............................ 28 4.6 Major Accomplishments ......................................................................... ..29 4.7. The Importance of Rural Civil Society for Democratization……………...32 4.8 Maintaining a Low Profile ..................................................................... ….34 4.9 Explaining Achievements: High Quality Field Staff and Savvy Leadership………………………………………………………………..34 4.10 Comparing Implementing Partners .......................................................... 35 4.11 The Relationship Between KTI and its Grantees ..................................... 38 4.12 Short vs. Long Term Support for Democratic Transitions……………….38 4.13 The Risks of Engaging Democratic Transitions ...................................... 39 4.14 Promoting Innovation .............................................................................. 40 4.15 The Legacy of Grants and Grantees ....................................................... 40 4.16 Social Capital and its Significance for Peace and Democratization ..... 41 V. DILEMMAS AND LESSONS LEARNED………………………………………….....43 5.1 Measuring and Attributing Impact……………………………………………..43 5.2 The Dilemma of OTI's Short-Term MO vs. the Long-Term Time Frames for Transitions to Democracy…………………………………………………..45 5.3 The Dilemma and Myth that OTI "Does Not Do" Sustainable Development……………………………………………………………………….46 5.4 The Dilemma and Myth that OTI Does Not "Build Capacity" ……….…….47 5.5 The Challenge of "Scaling Up": OTI's/KTI's Relationship to the USAID Mission's Democracy, Human Rights and Governance Program………..47 5.6 OTI's Dilemma of Being an "Agency Within an Agency" and the Challenge of Maintaining Intra and Inter Agency Harmony……………….48 5.7 Getting the Personnel Right…………………………………………………….51 5.8 The Grant Identification Process is Critical for Program Success………51 v 5.9 OTI's Objectives vs Grantee's ObjectIves ……………………………...…...52 5.10 The Choice of the Implementing Partner Makes a Difference……..…...52 5.11 The Limitations of "In Kind" Grants…………………………………………52 5.12 The Dilemma of Addressing Human Rights Violations vs. the Need to Protect Grantees Combating Such Violations…………….53 5.13 Devising an Appropriate Exit Strategy……………………………….…….53 5.14 The Dilemma of Evaluation: To Evaluate or Not Evaluate……………....54 5.15 The Dilemma of Attribution: To Brand or Not to Brand………………….55 VI. RECOMMENDATIONS ..................................................................................... 56 REFERENCES ....................................................................................................... 61 ANNEX I: PEOPLE INTERVIEWED FOR EVALUATION .................................... 63 ANNEX II: KTI GRANT PROJECTS REVIEWED, GRANT STATISTICS………....68 ANNEX III: FINDINGS FROM SURVEY OF GRANTEES…………………………….71 ANNEX IV: THE “BRIDGE” TO USAID’S DEVOLUTION SUPPORT PROGRAM...83 vi ACRONYMS AHADI Agile and Harmonized Assistance for Devolved Institutions (USAID) CHEM Chemonics Inc. CoP Chief of Party of USAID’s Implementing Partner CORD Coalition of Reform and Democracy CSO Bureau of Conflict Stabilization Operations, U.S. Department of State CSOs Civil Society Organizations CR OTI Country Representative DAI Development Alternatives Inc. DCHA USAID Bureau of Democracy, Conflict and Humanitarian Assistance DCR OTI Deputy Country Representative DG Democracy and Governance DRG Democracy, Rights and Governance GA Grants Assistant GM Grants Manager IEBC Independent Elections and Boundaries Commission IP Implementing Partner of USAID/OTI KANU Kenya African National Union KTI The Kenyan Transition Initiative program of USAID/OTI KTI-1 The first phase of KTI (implementing partner: DAI) KTI-2 The second phase of KTI (implementing partner: Chemonics) ODM Orange Democratic Movement OTI USAID Office of Transition Initiatives PA Procurement Assistant PEV Post-Election Violence PDO Project Development Officer PNU Party of National Unity PPR Program Performance Review PS Procurement Specialist SOW Scope of Work TNA The Party of National Unity URP United Republican Party vii EXECUTIVE SUMMARY Purpose and Methodology This evaluation of the Kenya Transition Initiative (KTI), OTI’s program in Kenya, is the second and final report of a two part exercise that began August, 2012 and concluded in at the end of October, 2013. The purpose of the first or “mid-term” evaluation conducted in September, 2012 and reported in (Barkan, Klopp and Ng’ethe 2012), was to assess the nature and impact of the program after roughly four years of operations and recommend improvements for its final year. It was based on a review of the principal program documents including internal performance reviews, and 99 hour-long interviews with the country representative and deputy representative of the program, staff of the implementing partner (IP), USAID and US embassy officials, and 43 recipients of KTI grants since its commencement of operations in May, 2008. The purpose of this second and final evaluation is different. While it also assesses the nature and impact of KTI operations, especially during the period following the first review, its primary concern has been to identify and consider the critical dilemmas and lessons learned from the KTI experience that are generic to OTI operations worldwide. As such the primary audience for this report is OTI itself, particularly its senior management, and more broadly USAID as OTI occupies a unique space and function within the Agency. The report is based on an additional 49 interviews conducted in Kenya in June 2013, and on the findings from a telephone survey of 178 KTI grantees conducted in February, 2013 immediately before national elections held in March. A complete list of all interviewed for both the mid-term and final evaluations is found in Annex I. Complete results from the survey are reported in Annex III Country Context and US Foreign Policy Objectives The anchor state of Eastern Africa, Kenya is the key to regional peace and stability and the prospects for democratic governance. For this reason the United States has devoted considerable attention and resources to Kenya since the country’s independence in 1963. The United States’ embassy in Kenya is its largest in Africa. US assistance programs to Kenya total over one billion dollars a year and are also the largest to any one African country. US military assistance including support via ANISOM is among the highest on the continent. Since 1992, the United States, via USAID, has also been the largest bilateral provider of assistance to achieve democratic governance, including support for the implementation of Kenya’s landmark constitution promulgated in 2010. OTI’s involvement and approach in Kenya must be considered in this context. That involvement was initially addressed to mitigating the effects of the 2007 post-election violence (PEV) that polarized the country and resulted in the deaths of 1,100 Kenyans and the displacement of 600,000 from their homes in January 2008. Over time, and especially after the passage of the new constitution, the program shifted to supporting democratic governance by strengthening civil society in areas of the country where it had been historically weak and key institutions of democratic governance, including the National Assembly (i.e. parliament), magistrate courts, and viii the Commission for Implementing the Constitution (CIC). As KTI winds up its operations during the second half of 2013, most of its efforts—now funded mainly by USAID Kenya and the UK Department for International Development rather than by OTI alone—are devoted to supporting Kenya’s new system of devolved governance provided by the new constitution. KTI’s Modus Operandi KTI’s support for these objectives followed the OTI model. (1) It has consisted mainly of a large number of small grants to nascent civil society organizations. By the middle of October, 2013, KTI had made 745 grants averaging just under $42,700 and totaling $31.8 million. Approximately one third of the grantee organizations supported by KTI received two or more grants. Sixty-four percent or just under two-thirds were made to civil society organizations. Nearly all were provided on an “in-kind” basis to avoid financial mismanagement by the recipient organizations. (2) At the same time, KTI operated in a highly entrepreneurial, flexible and opportunistic manner. By making a large number of small in-kind grants, KTI was also prepared to take risks and provide needed support quickly, sometimes within a period as short as two weeks. (3) As in other countries where OTI seeks to mitigate conflict in a post-conflict situation, KTI was able to operate in a politically charged post-conflict environment, because of the nature of its approach. Principal Findings KTI was successful in mounting its programs in three respects: First, the activities and institutions supported addressed the fundamental needs for which the program was established in Kenya—the need to mitigate and/or prevent further conflict, and the need to support democratization via the strengthening of key democratic institutions. A high proportion of grants were made in the most conflict prone areas of Kenya. Most of the organizations that received grants were supported by an international aid agency for the first time. Second, the OTI model of implementing a program via a large number of small “in-kind” grants was successful in that the activities supported by these grants were pursued as intended with very few failures. By providing support in this manner, KTI was able to reduce the failure rate to less than 6 percent of all grants. Grantees that performed at high levels were often provided with a second or third grant thus sustaining the work of “star performers.” Third, a program of the type pursued by KTI depends on key personnel in leadership positions supported by its local field staff operating at the grassroots. Both must be highly reliable yet entrepreneurial individuals who are prepared to take risks. Both must also be highly knowledgeable of local political conditions. KTI succeeded on all counts. The country representative and deputy representative were hard working, committed to the program and highly knowledgeable about Kenya, its politics, people and needs. They were also good managers in supervising subordinate personnel, including the chief of party (CoP) of the implementing partner and IP staff. KTI’s local field staff was composed mainly of young Kenyan professionals between the ages of 25 and 30 who are among the “best and the brightest” of their generation, and who performed well. Staff members serving in the capacity of project ix development officers (PDOs) and grants managers (GM) are critical for the process of identifying grantees, developing grant activities, and implementing grant activities. The program was also successful from the subjective perspective of KTI’s grantees. Results from the survey of 178 grantees indicate that KTI support was highly appreciated—86.5 of the grantees judged the activities funded as “Very Successful;” 91.8 percent said that the grants they received “strengthened” their organization and “built its capacity;” 98.8 percent said that they would “work with KTI again.” KTI staff were also given high marks—86.5 percent described KTI’s Project Development Officers (PDOs) as “Very Effective” in making their organization’s activities better than would have been the situation had there been no PDO and 80.9 percent rated the Grants Manager (GM) assigned to their grant as “Very Effective” (see Annex III). Notwithstanding these assessments, the evaluation team was unable to measure with precision the extent to which KTI’s program actually mitigated or prevented conflict. Having undertaken this evaluation four years after the start of the program, the evaluation team could not assess activities or their impact on a “before” and “after” basis in comparison with a control group of organizations that did not receive KTI support. Moreover, on the big question of whether KTI mitigated or prevented conflict arising during the run-up to and after Kenya’s national elections held in March, 2013, the evaluation team identified six other factors of why the elections were peaceful (see discussion in Section 5.1 on measuring and attributing impact). KTI nonetheless made a significant contribution to the development of social capital in Kenya, a critical requisite for peace and democratic governance at three levels: (1) by the recruitment and on the job experiences afforded to its Kenyan staff, most of whom are destined for leadership roles as their careers evolve; (2) by the strengthening of civil society organizations in conflict prone areas where previously there were few; (3) by giving rise to networks of these organizations. Though an intangible product, social capital is widely recognized by social scientists as a cornerstone of democracy (Tocqueville 1969, Putnam 2000) and thus peace. During its five and a half years of existence, KTI was carried out in the field by two implementing partners (IP) each with its own and distinctive approach. Development Alternatives Inc. (DAI) served as the IP from May, 2008 to May, 2011 when it was succeeded by Chemonics Inc. DAI was very fast identifying grantees and generous in the amount of funds devoted to carry out the activities specified in each grant. By contrast, Chemonics was more methodical and thus slower in selecting grantees and implementing their programs for funding. During the first fifteen months Chemonics served as IP, there was also a noticeable drop in the average amount of each grant as the figure dropped from $49,499. to $31,804. That figure has since risen to $45,301. The choice of IP makes a difference. Notwithstanding the recruitment of its excellent staff, and the success of its programs, the evaluation team found that there were few formal criteria for the recruitment of key staff including senior management, the chief of party (CoP) of the implementing partner, PDOs or GMs. Nor were there clear criteria for determining which grant proposals should be funded. Instead the probability of failure was primarily reduced through an intense and iterative process of project review known as rolling assessments. x One of KTI’s great strengths, consistent with OTI’s mandate, is that it moved quickly and entrepreneurially to take advantage of opportunities that presented themselves to both senior and junior KTI staff. However, operating in this highly independent manner sometimes gave rise to tensions within the USAID mission, especially between KTI and the mission’s program for democracy and governance, and between KTI’s leadership and senior personnel at the US embassy. Senior officials at the USAID mission to Kenya and the embassy praised KTI for its speed, flexibility and creativity, but also felt that the program had sometimes left them “outside the loop.” Because KTI mounted its programs in a post-conflict context, it sometimes and unintentionally funded activities that exposed KTI grantees to threats and physical risks. Although the number of grantees exposed was extremely small, its occurrence begs the question of how future OTI programs should minimize risks of this type, and how when such risks do occur they should be dealt with. Dilemmas and Lessons Learned There are a number of dilemmas generic to OTI operations worldwide that defy easy answers and on which we encourage OTI to further reflect. For a more detailed discussion, see Section V of the main report.  As its programs shift from those whose purpose is post–conflict mitigation to those which support the establishment of democratic institutions as a means to conflict prevention, OTI country programs often shift from short-term interventions to those whose focus is long-term and concerned with issues of sustainable development. Most OTI country programs last far longer than their intended period of two to three years, yet this raises issues of how these autonomous programs are to operate smoothly vis-à-vis or within USAID mission’s programs.  Measuring and attributing impact. As indicated above, it is very difficult to attribute and measure the impact of fast disbursing programs like KTI, because other equally powerful factors may be present.  By design and necessity, OTI operates on a rapid response and at times “under the radar” basis independent of the USAID mission’s democracy and governance program, and independent the embassy’s political section. However, because of (1) the long-term implications of these programs, including the need to support institution building, and (2) its highly political nature, OTI country programs must nonetheless work smoothly with USAID mission and embassy. This in turn requires a delicate balance between maintaining a critical measure of programmatic independence and coordination with other elements of the USG mission as a whole. Such a balance is essential to preserve OTI’s uniqueness and value—for itself, for USAID, and for the embassy which needs a rapid response unit as part of its “tool kit.” xi  The dilemma of OTI’s short-term modus operandi vs. the long-term time frames for transitions to democracy. OTI’s stated mission is that it “does not do sustainable development,” but because it is frequently drawn into efforts to build or strengthen key democratic institution, its programs take on a long-term dimension. How to manage this dimension in conjunction with the USAID mission is a challenge all OTI programs need to address.  Getting the personnel right is essential for program success. In the absence of extensive formal criteria for recruiting key personnel or for determining grantees, the burden of program success falls heavily on the judgments of OTI staff in the field from the country director on down to the field officers hired by the implementing partner.  The choice and approach of the implementing partner (IP) makes a real difference in how OTI programs are carried out in the field. Not all IPs are alike.  There is a delicate balance between mounting programs that address human rights violations in post-conflict situations and putting grantees at risk. Grants that address such violations must include measures that protect grantees.  Because OTI programs often shift from short-term interventions to long, greater and more systematic thought must be given to devising an appropriate exit strategy for each OTI country program.  To evaluate or not evaluate? OTI programs have historically been subject to minimal systematic evaluation by independent teams of outside evaluators, yet without such evaluations the importance of OTI’s work would never be known. Evaluations are also essential for any independent sub-agency that in essence is an “agency within an agency.” Principal Recommendations  OTI should continue its basic approach of mounting rapid, short term and opportunistic interventions to mitigate and prevent violence in post-conflict societies, while at the same time taking an opportunistic approach to strengthening key democratic institutions which is a long-term project.  OTI needs to rethink the time-frames of its programs. Not all programs are short-term, especially if OTI is to consolidate the gains obtained from the initial interventions it has supported.  Notwithstanding the inherent value of its approach, OTI needs to strengthen and “beef up” its evaluation procedures. Evaluations must occur at timely intervals so that “before and after” comparisons can be made of conditions in countries where OTI maintains operations, and the impact of these operations can be better assessed. Country programs that are extended beyond three years should also be evaluated at mid-term. Evaluations should also be xii compared on a cross-national basis to determine what works and what does not on a global basis.  OTI needs to develop or more clearly articulate its guidelines for the recruitment of key personnel as well as its guidelines for reviewing proposals for its grants.  The issue of grantee sustainability should be addressed especially when OTI provides support to institutions where the building of capacity is a long-term project.  OTI country programs should work with the USAID mission to develop a clear exit strategy at least a year before shut down to ensure that the assets built up over the course of the program are transferred to the USAID mission to the extent possible, and that the “handoff” to the mission is smooth.  Shielding grantees from physical harm. OTI should develop a clear human rights protection policy to ensure that any grantees put at physical risk because of OTI’s work will have clear procedures, personnel and networks that can be quickly activated to address their needs.  It is essential that the OTI country representative keep all relevant USG mission officials “in the loop.”  In each country where OTI has mounted a country program, consideration should be given to establishing an OTI alumni forum or association so that USAID can track personnel supported through the program.  As a general policy, OTI should strive to ensure that local host country experts are found at all levels of management in its programs.  OTI country programs should create a small advisory board of host country nationals to serve as a sounding board for the country representative, deputy rep, and chief of party of the implementing partner.  The country database used to track all grants should be improved to facilitate quick statistical analyses of the program as a whole. 1 KTI FINAL EVALUATION INTRODUCTION The Evaluation and the Evaluation Team The evaluation reported in this paper was a year-long exercise with two sets of objectives: First, to review the content and operations of the Kenya Transition Initiative (KTI), USAID/OTI’s program in Kenya, one year before its scheduled end, to assess the impact of the program and to recommend improvements during the final year of its existence. Second, to draw from the evaluation of this country program a series of generic “lessons learned,” questions and recommendations for OTI Washington as it moves forward to mount future programs elsewhere. Given this dual purpose, field investigations for the evaluation were, from the outset, divided into two parts. Part 1 was designed as a “mid-term” (or “three quarter term”) assessment conducted by the evaluation team in September 2012 after KTI had been in operation for roughly four years. It reviewed all aspects of the KTI program except its sub-program in the Eastleigh area of Nairobi from its inception in May 2008 through the period of the field work. Part 2 of the field investigations considered the remainder of the program from October 2012 through June 2013 when the evaluation team returned to Kenya for a second period of fieldwork. The evaluation team also considered KTI’s ongoing activities to lay the groundwork for USAID/Kenya’s AHADI program to support devolved government in Kenya that will continue through the end of 2013. Though not part of the KTI program “proper,” it constitutes KTI’s “handover” to USAID/Kenya as KTI exits the country, and provides several lessons learned for the conclusion of OTI programs generally. From the beginning, Parts 1 and 2 of the field investigations were intended to yield two different reports—a “mid-term” evaluation based on the first period of field investigations, and a final report that would supersede the first report by updating the findings from the initial field investigations with the findings from the second. The purpose and audience of the two reports were also different. The first was addressed primarily to KTI’s leadership and staff for the purpose of adjusting its approach during the final year of the program. By contrast, the second and final report is addressed to a broader and Washington based audience, that is to say, OTI and USAID generally, as it speaks to issues identified as common or likely to arise in most if not all of OTI’s programs around the world. The first period of field investigations on which both the first and this report are based consisted of extensive meetings and interviews conducted by the evaluation team with KTI staff as well as with a random sample of 40 grantees. These meetings took place over a period of five weeks beginning on August 27, 2012. The team also interviewed key members of USAID Kenya, the then US Charge d’Affaires (and now US ambassador) in Kenya, and prominent Kenyans knowledgeable about political conditions in the country especially the multifaceted process of democratic reform. The second period of field investigations in Kenya began on June 17, 2013 2 and lasted two weeks followed by additional interviews conducted by phone or in person in Washington during the second week of July, 2013. The information obtained through these meetings is also supplemented by a questionnaire based telephone survey of KTI grantees conducted during the second half of February 2013 by enumerators hired by KTI for this purpose. A full description of the methodology employed for the evaluation is found in Section III of this report. A complete list of those interviewed during both phases of the field investigations is found in Annex I. Selected findings from the survey of grantees are included in the discussion of the principal findings of the evaluation in Section IV. Complete results from the survey in the form of the answers to all question included in the questionnaire are reported in Annex III. The evaluation was conducted by a team of three political scientists with nearly a century of combined experience as researchers of Kenya’s political economy and related issues and as consultants or employees of the major aid agencies including DfID, USAID, and the World Bank. The team was led by Joel D. Barkan, Professor Emeritus of Political Science at the University of Iowa and currently Senior Associate at the Africa Program of the Center for Strategic and International Studies in Washington, DC. Njuguna Ng’ethe, Associate Professor and former director of the Institute of Development Studies at the University of Nairobi, and Jacqueline M. Klopp, Associate Research Scholar at the Earth Institute, Columbia University, rounded out the team. The team would like to express its appreciation for the support it received throughout the evaluation exercise from John Langlois, the country director of KTI, Galeeb Kachra, the deputy country director, Megan German, the Kenya Program Manager at USAID/OTI in Washington and Francis O’Malley, Administrative Program Specialist for the OTI Africa team. The team also wishes to acknowledge the excellent logistic support provided throughout the evaluation exercise by Elizabeth Gabster of the QED Group, LLC. Notwithstanding this assistance, the views expressed in this evaluation report are those of the co-authors alone, and do not represent any official position of the organizations with which they are affiliated or of the U.S. Agency of International Development. 3 I. COUNTRY CONTEXT AND U.S. POLICY OBJECTIVES 1.1 Kenya’s Two Decade Struggle for Democracy For more than two decades, Kenya has been engaged in a protracted struggle to become a prosperous democratic state. The struggle began in 1990 with protests against a decade long period of political and economic decline experienced under Kenya’s second president, Daniel arap Moi. These protests were led by a small number of political leaders who broke with Moi and supported by emergent civil society organizations (CSOs) of the time. They were part of the broader resistance to authoritarian rule that swept across Africa at the end of the Cold War. In 1978 Daniel arap Moi succeeded Jomo Kenyatta, Kenya’s first president, upon his death. During the 1980s, Moi’s rule became increasingly repressive and corrupt as he turned Kenya into a de jure one-party state. Like Kenyatta before him, Moi maintained himself in power by disbursing increasing amounts of patronage coupled with the jailing of those who refused to be bought. The Moi presidency was a quintessential example of neo-patrimonial or “big man” rule—the dominant form of governance across Africa during this period. From the mid￾1980s to 1990 per capita income in Kenya declined at an average rate of 1.0 percent a year. Social service delivery likewise declined. Four objectives were the basis of the early protests against Moi. (1) To roll back the repression of the 1980s by opening up a measure of political space so that civil society and a free press could function in a political system that had heretofore been closed. (2) To replace the existing system of one party rule with a multiparty system in which rival political parties could compete for power via elections that were administratively competent, “free, and fair.” (3) To reduce the power of what had become an imperial presidency and hold the government accountable to the governed—that is to say, to transform Kenya into a modern democratic state. (4) To reverse Kenya’s economic decline through the pursuit of prudent macro￾economic policies, a reduction of corruption, the rebuilding of the civil service, and the reestablishment of the rule of law. To realize these objectives, the reformers also demanded a new constitution. The protesters’ demands were echoed by key members of the international community, particularly the United States, which had become increasingly concerned by Moi’s repressive rule, corruption, and declining incomes for most members of Kenyan society. With the end of the Cold War and the end of apartheid in South Africa, they decided to force a change from the status quo. In November 1991, Kenya’s principal donors suspended $250 million in economic assistance to pressure Moi to address their concerns. By the end of the year, the President responded by securing Parliament’s repeal of the article in the constitution that defined Kenya as a one-party state. This opened the door for the establishment of several opposition political parties, and the holding of Kenya’s first multiparty elections in 26 years in December, 1992. Since then, multiparty elections for the presidency and for members of parliament have been held every five years—in 1997, 2002, 2007, and in March 2013. 4 In 1992 and 1997, President Moi retained power though neither he nor his party won a majority of the vote.1 He was succeeded in 2002 by Mwai Kibaki, the leading opposition candidate in the two preceding elections who finally replaced Moi when the president was barred by the Kenyan constitution from seeking a third elected term.2 Kibaki defeated Uhuru Kenyatta, son of Kenya’s first president, Jomo Kenyatta, by winning nearly 60 percent of the popular vote. He was reelected in 2007 under circumstances that many, including domestic and international observers, regarded as fraudulent because of irregularities in the tabulation and reporting of the vote. Though Kibaki and his chief challenger, Raila Odinga, each received approximately 45 percent of vote, a news blackout during the final day of the count and Kibaki’s hasty taking of the oath of office behind closed doors before the results could be verified, triggered widespread outbreaks of violence across Kenya. The violence began on December 29, 2007 and lasted through January 2008. Analysts disaggregate the violence into at least two broad types: mass killings and the displacement of people from their homes by the use of militias and excessive force by the police against unarmed civilans. Militias linked to followers of Raila Odinga, who believed their candidate had been cheated of victory, attacked the supporters of Kibaki in and around Eldoret, Kisumu, and elsewhere in the western Rift Valley causing mass displacements, rape and killings. After approximately one week, militias linked to followers of Kibaki attacked supporters of Odinga, mainly in Naivasha and Nakuru using similar tactics of ethnic targeted violence caused more killings and displacements. At the same time, police used live ammunition on unarmed civilians especially those protesting the election results in Kisumu. Attacks and protests also occurred in the Kibera and other slum areas of Nairobi as the capital city became divided into zones. The violence was mainly interethnic in character because the political parties contesting the election were coalitions of different ethnic groups. Campaigning was rife with hate speech that emphasized the overlap between ethnic identity and party affiliation. Supporters of Odinga were predominantly Luos and Kalenjins while supporters of Kibaki were mainly Kikuyus. In the course of the violence 1,133 Kenyans died and nearly 600,000 were displaced from their homes. Peace was ultimately restored on February 28, 2008 when Kibaki and Odinga, the principal adversaries in the crisis, signed the National Accord. The National Accord was brokered by former UN Secretary General Kofi Anan supported by the African Union’s Panel of Eminent African Personalities. It provided for the formation of a coalition government that would be co-led by Mwai Kibaki who was retained as President, and Raila Odinga, who was appointed to the newly created office of Prime Minister. Each would control half the portfolios in an expanded government of 36 ministries. The National Accord also committed the signatories to four main agenda items known as the Agenda Four reforms. The Accord also provided for the establishment of several special commissions to realize the 1 Moi won the presidency and his party, KANU, won a majority of the seats in the National Assembly because under Kenya’s “first past the post system,” the candidate with the largest number of votes, even if only a plurality, wins the seat. In both elections, the opposition split its votes among four parties and there was substantial state-instigated violence during the period before the polls. 2 The provision for term limits was inserted by Moi himself when the constitution was amended in 1991 to allow for multiparty politics. He did so mainly to appease the donors, but in 2002 Kenyans and the donors held him to the rule, and he reluctantly stepped down from office. 5 reforms and to prevent future violence by addressing the structural issues, especially the distribution of land, ailing the country. A commission of experts was also established to draft a new constitution. Despite the violence that followed the 2007 elections, Kenya has made significant progress towards achieving the original objectives of the early protests against Moi. Indeed, the Kenya of today bears little resemblance to the country of two decades ago, and it stands on the cusp of becoming a genuine democracy. The drafting and passage in a national referendum of a new constitution in August 2010 is arguably the greatest achievement to date. The new constitution puts an end to the imperial presidency by establishing a more equitable balance of power between the executive and the legislative and judicial branches of government. The new constitution created a bicameral parliament consisting of an expanded National Assembly and a Senate. The constitution also provides for a modest albeit significant devolution of power to 47 county level governments that will assume responsibility for social service delivery in such key areas as health, agriculture and roads at the local level. Though fraught with complexity, devolution holds out the prospect of providing a measure of power and resources to the members of Kenya’s smaller ethnic groups while breaking up the political cohesiveness and power bases of the large. The new constitution also contains important safe guards against human rights abuse, a new and independent electoral commission, gender equity, and provisions to raise the level of integrity among public officials both elected and non-elected. Whether these and other provisions of the new constitution will be fully implemented as Kenya moves forward following the 2013 elections is the big unanswered question in the country today. A new generation of political leaders has assumed office at all levels of government from the presidency on down. But will they fully embrace the new constitution or govern Kenya like their predecessors who embraced authoritarian rule? The 2013 elections in which Uhuru Kenyatta and William Ruto defeated Raila Odinga and Kalonza Msyoka to become President and Deputy President were surprisingly peaceful given Kenya’s history of electoral violence. The elections were the most complex in Kenya’s history as voters chose 418 members of the new bicameral Parliament,3 governors of the 47 new county governments and over 500 members of the new county assemblies. As new governmental structures are being established for the first time and old ones revamped, the incumbents of these structures are testing the limits of their authority. One consequence of this situation is that Kenya’s politics are currently passing through an extremely fluid stage that is likely to last for at least a year, perhaps more. Kenya’s emerging democracy is thus a fragile work in progress as the country is being pulled in two directions. On one side, is a group of prominent political leaders seeking to maintain aspects of the status quo by mobilizing the electorate on the basis of its ethnic identifications and by blocking full implementation of the 2010 Constitution. On the other side are leaders trying to consolidate the new democratic order, especially the new system of devolved government. Still others position themselves in between or take no position. 3 Of this number, voters directly elected 337 members of the National Assembly, and 47 members of the Senate. The remainder of seats in both chambers are allocated to nominees of political parties on the basis of proportional representation. 6 Rather than advance the implementation of the emerging democratic political order, the 2013 elections has exacerbated the fragility of Kenya’s political system in the near term. Uhuru Kenyatta’s margin of victory over Raila Odinga was decisive (50 percent to 43 percent), but the elections were clouded by breakdowns in their administration by the new Independent Elections and Boundaries Commission (IEBC) that were similar to those that occurred in 2007 under the former electoral commission which the IEBC replaced. The failure of a sophisticated computerized system for reporting the results in real time preceded by the failure in the computerized system for matching individual voters with voters in the official register of voters triggered allegations of fraud and undermined the legitimacy of both the elections and the Commission (Long, Gibson, Kanyinga and Ferre 2013; Barkan 2013). Odinga challenged the results in a petition to the new Supreme Court. However, the court upheld Kenyatta’s victory in a unanimous decision. Although the elections were relatively peaceful and Odinga accepted the Court’s decision, his belief and the belief amongst his supporters that the election was stolen persists months after the poll. As a result, Kenyan society remains deeply divided. On one side are the supporters of President Kenyatta, the acknowledged leader of Kenya’s largest ethnic group, the Kikuyu (17.2 percent of the country’s population),4 and his running mate, William Ruto, the most prominent politician amongst Kenya’s second largest group, the Kalenjin (12.9 percent). Their attitude is that notwithstanding the problems of the IEBC, Kenyatta and Ruto won the recent election fair and square, that the elections were peaceful, and that Kenya is ready to move forward towards become a middle income country. It is therefore time for all Kenyans, including the losers in the election, to accept the outcome and move on. By contrast, the Odinga’s supporters and members of the ethnic groups that backed his candidacy—the Luo (10.5 percent of the population), the Kamba (10.1 percent), the peoples on the Indian Ocean coast (approximately 10 percent), and the majority of the Luhya (12.8 percent)—believe that their future is bleak, and that the new government is not fully committed to the implementation of the 2010 constitution. Having failed to win power at the national level, including control of both houses of parliament, they seek refuge within the new tier of county level government. The division between the two camps is a classic example of the jubilant majority verses an aggrieved minority, exacerbated by the fact that an overwhelming majority of the members of the ethnic groups that comprise one coalition are opposed by similar majorities of those that comprise the other. For Uhuru Kenyatta, the challenge of governing a divided country is daunting. He must govern under a new constitutional order which he did not fully embrace, and that provides the president with less power than his predecessors. He must also hold on to his supporters and Ruto’s while minimizing the feelings of grievance shared by his opponents. 4 All percentages are calculated from Kenya Census 2009, and suggest that the Kikuyu share of Kenya’s population may have dropped by up to five percentage points since the preceding census a decade earlier. 7 The Kenyatta-Ruto coalition, called Jubilee, is itself fragile as it is led by two of the principal protagonists in the 2007 elections and their respective supporters who carried out extreme acts of violence against each other six years ago. The Jubilee coalition came together as a political strategy by its leaders to deal with their respective indictments by the International Criminal Court (ICC) for their alleged roles in the post-election violence. How their cases play out before the ICC coupled with the reality that the Deputy President aspires to become president himself, complicates the challenge of governing for Jubilee. The outcomes of these issues and their ripple effects are also unknown as the Kenya Transition Initiative winds up its program. 1.2 U.S. Policy Objectives in Kenya The United States has supported Kenya’s transition to democracy since the early protests against authoritarian rule two decades ago. At the end of the Cold War, the United States shifted its policy objectives in Africa from one of containing the influence of the former Soviet Union to one that emphasized democratic governance and the maintenance of peace and stability across the continent. Because Kenya is the “anchor state” of Eastern Africa, the United States has long regarded Kenya as a focal point of its Africa policy. Put simply, as Kenya went, so goes the greater eastern Africa region. The result was that during the Cold War, the U.S. regarded Kenya as a strategic partner in the policy of containment, and was uncritical of authoritarian rule. The end of the Cold War marked a strategic shift in approach. U.S. policy makers now embraced the view that democratic governance was not only consistent with American values, but critical for the attainment of regional peace and security. Indeed, the protests against authoritarian rule that swept across Africa during the early 1990s suggest that its continuation was a threat to political stability. Democratic governance was also viewed as important for economic development, a third U.S. policy goal. During the past decade, however, the focus of U.S. Africa policy has shifted again to one that balances support for democratization in those countries where the prospects for democracy are promising with support for regimes that take a strong stand against terrorism and share US security concerns. The most recent articulation of the importance of democracy promotion in U.S. Africa policy tempered by a concern for maintaining security and increasing American trade and investment across the continent was presented by President Obama in his speech at the University of Cape Town on June 30th, two days after the conclusion of the second round of field investigations for this evaluation (Obama 2013). In addition to playing a leading diplomatic role in mobilizing the international community to support the end of authoritarian rule in Kenya, the United States has been active in joining other members of the international community to promote democratic governance since 1992.5 This has taken the form of providing financial and technical assistance to strengthen civil society, improve the quality of elections, and strengthen key democratic institutions, 5 The United States was less active in its emphasis on democratization from 1994 to 1996 and in 2011-2012 as a result of the orientation of its then ambassadors in Nairobi. 8 particularly the National Assembly. This support has been most effective in achieving its objectives when it has been led by the U.S. ambassador to Kenya in close coordination with the USAID mission. To succeed, programs to strengthen democratic governance require both the political leadership provided by the embassy and the operational capacity provided by USAID—whether by the main USAID mission or by USAID’s Office of Transition Initiatives (OTI) or a combination of the two. U.S. democracy promotion efforts in Kenya have arguably been most effective when it has been mounted on a multilateral basis.6 Launched in 2008 immediately after the signing of the National Accord that established the grand coalition government of former president Mwai Kibaki and former prime minister Raila Odinga, the OTI Program in Kenya known as the Kenya Transition Initiative (KTI) is best understood and judged in the context of long term U.S. efforts to support democratic governance in the country. KTI program was launched in the wake of the post-election violence (PEV) that followed the 2007 elections, and like OTI programs elsewhere was intended to be quick-disbursing and addressed to the crisis that had engulfed the country. However, the program soon morphed into supporting Kenya’s long term struggle to become a modern democratic state. It is therefore important to remember that KTI is not the first USAID effort to support democratization in Kenya, but rather one that brought a unique approach to an area of development assistance that had been pursued for more than 15 years. The dawn of the 21st Century brought a second major theme in U.S. foreign policy in the post￾Cold War period—the war against terror. Following the bombings of the American embassies in Kenya and Tanzania in 1998, and especially after the attacks on the World Trade Center and Pentagon on September 11, 2001, Kenya has become a major focus of U.S. counter￾terrorist efforts in Eastern Africa. Kenya’s invasion of southern Somalia in 2011 to counter the al Qaeda affiliate, al-Shabab, is widely believed to have occurred with tacit US support as part of their common fight against international terrorism. How this impacts perceptions of the United States in Kenya, and what tensions may arise between U.S. democracy and human right promotion efforts, and counter-terrorism assistance, is an open question that needs to be explored. KTI is involved in some instances in addressing the blowback against the U.S. on the Kenya Coast where several Muslim preachers have been assassinated, and in places like Nairobi’s Eastleigh area where a large population of Somali refugees is skeptical of U.S. intentions. Though consideration of counter-terrorism programs in Kenya are beyond the scope of this report, U.S. programs of democracy assistance and conflict prevention, including KTI, cannot ignore their impact and how they are perceived by the Kenyan public or groups where KTI operates including Eastleigh and Mombasa. 6 Coordination amongst donors via the Democracy and Governance Donors’ Group (DGDG), a contact group of “like minded” donors, began in 1992 and has continued ever since although the modalities and name of the group have changed over the years. Though some countries participate irregularly, the core members of the group includes the United States, the United Kingdom, Germany, the Netherlands, Denmark, Sweden, Norway and Canada. The frequency and intensity of the meetings and activities of the DGDG tends to ebb and flow with Kenya’s election cycles. 9 II. THE NATURE AND HISTORY OF THE OTI PROGRAM IN KENYA (KTI) 2.1 The Distinctiveness of OTI and its Relationship to USAID The Office of Transitions Initiatives (OTI) was established in 1994 as a separate office within USAID to provide rapid, flexible, and short term aid to countries experiencing conflict and crisis. The purpose of OTI was to complement existing programs at USAID by bridging the operational gap between disaster relief, which is typically rapid in its response to crises but short term in duration, and long term development assistance, which is often unacceptably slow in starting and overly bureaucratized and rule-bound (Lawson 2009). In contrast OTI was established to be “creative, fast and risk taking.”7 OTI staff, both in Washington and the field, repeatedly emphasized this distinction and culture with some pride. The development of OTI’s “can do” culture, however, results in stated and unstated tensions between the Office and the rest of USAID, especially at the mission level. OTI staff regard themselves as effective because they are fast off the mark, while non-OTI staff elsewhere at USAID missions often regard OTI’s country level programs as a “Lone Ranger”—effective and well￾intentioned and often very useful, but an independent animal from the rest of the mission. Indeed, OTI is, in many ways, “an agency within an agency.” These distinctions in modes of operation and organizational culture are reinforced by four additional realities. First, OTI exists and operates under unique rules to act quickly to disburse assistance rapidly in a way that the rest of USAID cannot. Second, approximately 90 percent of OTI staff are personal service contractors (PSCs) whereas most American staff at USAID are career foreign service officers. Third, OTI’s country level representatives, staff and implementing partners report directly to the OTI headquarters in Washington, and not via the USAID mission director. That said, OTI programs do operate in country “at the pleasure of” the mission. Fourth, OTI’s country level programs are arguably more costly to implement, because of their “in-kind” disbursements than comparable sized programs run by the USAID missions. The merits and demerits of the differences between OTI and the rest of USAID are beyond the scope of this evaluation, but the differences are real and impact on this evaluation in two respects. First, it is important that the OTI program in Kenya be evaluated on its own terms; that is to say, from the perspective of the purpose and mode of operation of the OTI. To do otherwise, would be to apply inappropriate criteria for determining the degree of success or failure as well as the lessons learned. Indeed, before commencing this evaluation, OTI Washington devoted considerable effort to brief the members of the evaluation team on OTI’s purpose and overall approach in the countries where it works. Second, and because OTI is often successful in achieving its objectives, the question arises—as it did early in the course of this evaluation—of what becomes of OTI’s work and product once its short term program in Kenya (and elsewhere) shuts down and OTI’s staff depart? Are they simply abandoned, or are they taken over by the USAID mission, and if so, what issues does OTI’s legacy pose? 7 As Marian Lawson notes “OTI’s legislative authority is based on the disaster relief provision of the Foreign Assistance Act of 1961, which allows for broad executive discretion”(Lawson 2009). 10 Although these questions were not specified in the SOW for this analysis, the evaluators concluded that the evaluation report would be incomplete if they were not considered. A discussion of these questions is found in Sections V and VI below. Because they are targeted to countries in crisis, programs mounted by OTI are also often explicitly political in their objectives in that they are intended to support the establishment or consolidation of democratic rule in countries of high strategic importance to the United States. OTI’s strength is its ability to respond rapidly to the shifting political dynamics of whatever country it is operating in by devising, revising, and implementing interventions that advance two overarching U.S. policy goals—political stability and democratic governance. In particular, OTI’s ability to rapidly disburse small “in-kind” grants to civil society organizations that would ordinarily not qualify for financial support via USAID’s traditional programs allows for quick and concerted action as needed.8 OTI’s ability to disburse medium to large in-kind grants through the direct procurement of needed equipment and supplies enables it to also assist governmental entities such as special commissions and the judiciary in a timely manner. This approach, however, also carries some clear risks. For example, there can be the risk of a lack of transparency and accountability when money is disbursed quickly via OTI contractors if they do not keep good records. Other risks lay in the inadvertent undermining of opponents of authoritarian rule who might be branded by local critics or the press as “American agents”, destabilization, capture, or manipulation by various political actors as occurred in Venezuela (Lawson 2009). Finally, and because OTI’s use of grants “in-kind” are also more expensive than conventional USAID programs, their short to medium term nature beg the question of the sustainability of OTI’s programs after OTI withdraws.9 Will, for example, USAID assume responsibility for the continuation of KTI activities after the close of its program, and if so, under what criteria? Will some initiatives be taken over by other donors, and if so, under what criteria? These questions suggest the need for a more coherent exit strategy as KTI and OTI programs elsewhere move towards completion. Serious coordination problems potentially arise from the fact that while the local OTI program is expected to work closely with the USAID mission, it is not directly accountable to the mission but to the main OTI office in Washington, DC. The extent to which coordination 8 USAID’s traditional form of support is to provide funds and/or technical assistance to local organizations or other institutions including government agencies, research centers and universities that are qualified legal entities capable of managing the funds. To become a qualified entity, the organization must be legally registered and established a rigorous system of financial management approved by USAID. USAID often assists organizations, particularly civil society organizations, to establish such systems, but the entire process of qualifying for a concessional loan or grant takes upwards of a year. OTI bypasses this process entirely by making its grants “in-kind,” that is to say, by procuring and paying for needed items itself. The approach enables fast disbursements to organizations that would not ordinarily qualify in the near term for assistance from USAID. The approach is particularly useful when seeking to provide assistance quickly to new and in some cases informal CSOs operating in the rural areas, but the burden of work shifts from the recipient to the provider, in this case KTI’s implementing partner. 9 OTI programs typically last two to four years. Most USAID programs are, by contrast, of much longer duration including some that last a decade or more. 11 between OTI and the rest of the USAID mission occurs is thus a function of the personalities and working styles of the individuals involved, particularly the country representative (CR) and deputy country representative (DCR) of the OTI program and the director of the USAID mission and director of its democracy and governance program for USAID. Given its political objectives, OTI programs must also develop a good working relationship with the U.S. Ambassador and political section at the American Embassy. 2.2 OTI in Kenya: The Kenya Transition Initiative Given its purpose to mount fast disbursing programs in post-conflict societies, OTI Washington responded to the 2007/2008 post-election violence (PEV) in Kenya by initiating a dialogue with the USAID Mission and the U.S. Embassy in Nairobi about how OTI might operate in Kenya. From a foreign policy perspective, the U.S. had a clear interest in Kenya’s s recovery and reform. The National Accord signed on February 28, 2008, created sufficient stability and a “window of opportunity” by the ambitious reform agenda specified by the Accord. After negotiations and discussions within USAID, OTI moved forward with a field assessment in mid-March 2008 that confirmed the opportunity for an OTI program in the country. The March assessment identified “increased exercise by the Kenyan state and non-state actors of their capacity to (1) promote and enable broad-based recovery from the 2007/2008 election-related violence and (2) address the underlying causes of instability as key objectives for the program” (USAID/DCHA/OTI 2008). It recommended that $5.5 million be allocated for a program with the following breakdown: grants for activities to help the recovery from violence ($1.5 million), grants to address deeper structural problems in the country ($2.5 million), and operations ($1.5 million). Subsequently, USAID/DCHA successfully notified Congress of its intent to use OTI funds. The Kenya Transitions Initiative (KTI) program started in May 2008, only two and a half months after the assessment and thus an example of how rapidly OTI can translate an idea for a program into reality. Development Alternatives Inc. (DAI) was selected as the implementing partner with Getu Reta as its chief of party (CoP).10 After some internal delay, the OTI/KTI Country Representative (CR) and his deputy (DCR) were also in place. By September, three field offices were established in Nairobi, Eldoret, and Kisumu, the urban centers of the “hot spot” areas that bore the brunt of the post-election violence. DAI served as the implementing partner (IP) for the first three years of KTI until May 2011 when it was succeeded by Chemonics Inc. after the program was extended through the 2013 national election. As per USAID rules, the extension of the program resulted in the contract for the IP being rebid for the second round. Chemonics brought in a new CoP and staff though a small number of staff were retained from the previous IP. The first period when DAI was the IP is now referred to as KTI-1 while the current period under Chemonics is known as KTI-2. 10 Although the evaluation team exchanged emails with Mr. Reta, he was in Ethiopia and unfortunately died in early 2013 before he could be interviewed by phone for this evaluation. 12 Besides the change of implementing partner, KTI-2 involved a shift of its field office in Kisumu to Kericho, deepening its presence in the conflict prone northwestern Rift Valley. KTI also opened an office in Mombasa in June 2012 in response to serious problems on Kenya’s coast, including the rise of a secessionist movement (the Mombasa Republican Council) threatening to boycott and possibly disrupt the 2013 elections, terrorism, and the related internationalized “war on terror” against al-Shabab. KTI also started a program in Eastleigh where large community of Somali immigrants have come to live, resulting in a complex set of issues tied to Kenya’s long term stability. While this Eastleigh program is outside the purview of this evaluation, because it continues through 2014, it is indicative of how KTI has evolved and adapted to the very difficult and unpredictable political situation in Kenya. The program in Eastleigh is also the subject of a separate evaluation. Although the implementing strategy had been determined by May 2008 it could not be fully brought online until a cadre of talented Kenyan professionals for each office was hired. OTI/KTI operates using a personnel-intensive approach that delegates considerable authority to the field—to the country representative (CR) and deputy country representative (DCR) who in turn delegate to the local staff who run the program’s field offices. The approach gives KTI considerable freedom to move quickly, experiment, adapt and create networks and synergies among grantees and their activities. However, to succeed, the program must recruit talented people with the right temperament and skills for the job. The CR and DCR as well as the heads of the program’s field offices and supporting personnel must be smart and knowledgeable about the country’s political economy and confident in their work. They must also be imaginative and think strategically as well as possessing insight, political acumen, and high integrity. The vetting of prospective staff is a critically important task, as is the fostering of a collaborative and open team dynamic after they are hired. The KTI CR described the approach as “flexible, short term, and serendipitous” (Interview with John Langlois, August 27, 2012). Not all experimental grants work but borrowing from the model of venture capital, OTI Washington takes the view that if it can create productive and especially scalable activities from the successes and learn from the failures, it will still be more productive and likely to have a greater impact than if it took no or fewer risks. At an orientation meeting for two members of the evaluation team in June 2012, OTI Director Rob Jenkins noted that the OTI aims for a success rate of 80 percent but is happy with 60 percent given the payoffs obtained. Like venture capital, there is a clear trade-off between risk and the anticipated “big” payoff. This perspective, however, begs the question of how “success” is defined and measured, and how the risks of failures are minimized or contained. In respect to OTI’s main program in Kenya, the basic approach is that the grants KTI provides to recipient organizations are numerous, creative, and disbursed quickly, but small in terms of their dollar value.11 In other words, the risks KTI takes are modest in the course of successive iterations of disbursements 11 From approximately May 2008 to May 2011 when the KTI program was implemented by Development Alternatives Incorporated (DAI), the program made 237 grants at an average cost of $49,449. Since then when Chemonics Inc. (CHEM) became the implementing partner, KTI has made 230 Grants at an average disbursement of $31,104. 13 because the program sows many seeds, and, depending on which bear fruit, it often provides a second and third round of grants to those organizations that mounted a successful micro￾project with its first grant. Using this strategy, the financial risk is minimized while the ability to multiply the successes is clear. That said, the determination of what constitutes a successful program is made on a case-by-case basis by key staff involved with its implementation. There are apparently no standard criteria for re-funding other than the condition that the grantee fully implemented its initial grant for the purpose intended and a rough judgment by KTI staff of its impact and whether the activity should be repeated, or whether the grantee should be funded again for a different exercise. The diversity of KTI initiatives may make it impossible to apply a single or common set of standards to evaluate a potpourri of qualitatively different interventions. The risk to recipient organizations, however, can be high. Some of KTI’s most creative projects are those that have “pushed the envelope” and the boundaries of political permissibility in that they have been perceived as threatening by opponents of democratic reforms.12 The result is that in some cases, these opponents have threatened the physical safety of KTI grantees and, in some instances, KTI staff. This reality raises serious questions about the fallout of the failures that do occur as well as other issues such as security concerns for grantees engaged in politically sensitive work.13 What responsibility and obligations, if any, does KTI and USAID as an agency assume when it makes grant or otherwise assists a Kenyan organization that is subsequently put at risk by the activities funded—especially when these activities have their intended effect? Is the responsibility merely that of providing recipients with a written warning or a disclaimer before the grant is formally made that use of the grant may put them at risk? Does KTI incur an obligation to provide protection or redress, and if so, in what form and how much? These are thorny issues to which there is no clear answer, but issues which must nonetheless be addressed as discussed further below. The OTI/KTI approach of providing assistance via a large number of small grants also raises issues of accountability. How, for example, does KTI monitor and evaluate these activities particularly when their purpose is to effect political change? “Success” is difficult to measure and attribute with precision. 2.3 KTI’s Modus Operandi KTI also functions somewhat like a mini foundation similar to the German party foundations or the governance program of the Ford Foundation in that the ratio of the number and quality of staff to the dollar cost of the activities funded is much higher than that of USAID generally and the other major bilateral assistance agencies working in the governance field. In addition to the CR and DCR, and senior staff of the implementing partner, KTI has one or two Kenyan professionals in the key position of Program Development Officer (PDO) based at each of its 12 One such an activity would be the series of radio programs on the new constitution and produced and broadcast by the Kalenjin language station, KASS-FM. 13 For example, one grantee ultimately felt that he had to leave Kenya to protect his personal safety. 14 four (now three) field offices. Each of the PDOs is in turn supported by one or two grants managers (GM) and grants assistants (GA) as well as a two person team of procurement specialists based in Nairobi.14 The role of the PDO is to both solicit and receive proposals for potential activities and grants from local civil society organizations and or talented individuals who have the skills and interest to mount programs that advance KTI’s objectives in Kenya. These officers look for new and innovative ideas for activities or programs that address the primary objectives of KTI, namely to prevent a repeat of post-election violence that followed the 2007 elections, to prevent or mitigate violence in the run-up to the elections held in March 2013, to support the full implementation of the National Accord and democratic reforms generally. At the time of the first field investigation for this evaluation in September 2012, KTI had made just over 500 small grants in support of these objectives. By the time the team returned in June 2013, the total number of grants had risen to nearly 800 by the end of October when this evaluation was submitted to USAID. Approximately 40 percent were in the general area of conflict prevention and/or conflict mitigation.15 The rest were in a wide range of areas supporting various aspects of the National Accord and Kenya’s transition to democratic rule. Most of these grants were made to small civil society organizations or informal groups of a few individuals located in the “hot spot” areas that had experienced electoral violence in the past. These grants averaged just over $42,700 each. They were usually made on an “in-kind” basis; that is to say, the recipient organizations rarely received money. Rather they were given a figure in Kenyan shillings against which KTI would procure needed supplies, salaries, consultants and/or equipment, or pay for other items up to the amount of the grant to underwrite some specified activity. A significant portion of each grant, particularly when the recipient organization was organizing a conference, workshop, or performance to which a large number of people were invited, was typically spent on the transportation and feeding of participants. In such cases KTI procures the food and disburses transport allowances directly to the participants via MPESA rather than depositing funds in the account of the grantee so that it could make the payments. A full list of the grants and grantees reviewed for this evaluation, as well as a statistical analysis of all grants is found in Annex II of this report. Consistent with the emphasis on conflict mitigation a high proportion of grants were concentrated in the Western Rift Valley which bore the brunt of the post-election violence (see map on page iii). Beginning in June 2012, KTI opened an office in Mombasa to address the rising potential for violence along the Indian Ocean coast. 14 It was noted that a majority are young, highly qualified women. The Eldoret and Kericho offices also have a Professional Development Specialist who works with the PDOs and consultants and technical advisors are also hired at different times. 15 Of the 797 grants made by KTI as of October 16, 2013, 132 or 16.6 percent were classified by KTI as falling in the “conflict mitigation” sector. However, large numbers of grants classified in other sectors such as “youth engagement,” “media,” “civic action” and “advocacy” were also addressed to conflict mitigation and conflict prevention while 39.5 percent of the grantees who participated in the telephone survey conducted at the end of February, 2013 reported that their grants were in the conflict mitigation/conflict prevention field. The precise figure is difficult to calculate, but clearly grants in this area constitute the largest single category of grants with grants designed to strengthen democratic institutions constituting the second largest category of roughly a fifth of all grants made. For details see summary of grants by sector, Annex II at bottom of page 70 and Annex III, Table 14 which reports grantee’s descriptions of the grants they received. 15 In addition to its main portfolio of small grants, KTI made a small number of large grants ranging from $100,000 to nearly $850,000 to support “targets of opportunity” that emerged at short notice, but which were viewed as particularly worthy given the objectives of the program. These grants constituted less than 8 percent of the total (see Annex III, Table 5). They included a grant of $120,000 to computerize the court record system for the magistrate court in Eldoret, a similar grant for the magistrate court in Mombasa for $113,841, two grants totaling $149,700 to provide offices and equipment for the Constitution Implementation Commission plus five related grants totaling $1,048,548 to establish the Parliamentary Media Center and purchase of broadcast equipment to enable the National Assembly to televise its daily proceedings. The provision of such equipment complemented the decade long program to strengthen the legislatures initiated by USAID and later co-funded by USAID and DfID. All four of these “one-off” exercises were deemed as highly successful by their recipients. KTI has also made grants to respond strategically to Kenya’s changing political circumstances, including its decision to lay the groundwork for USAID’s support of devolved government during the final eight months of its program.16 Most of the small projects are unsolicited by the CR, DCR, or PDOs, but a high proportion of these are modified in the course of the dialogue between the PDOs and potential grantees. PDOs often refer to this process as “panel beating” a proposal into shape. A significant proportion of grantees interviewed by the evaluation team, perhaps as many as a quarter, expressed some unhappiness with the process as they felt that they had been forced to overly modify the programs for which they had sought funds. However, when we probed further to determine the extent of “panel beating” in the survey of grantees, conducted in February, 2013, only 5.2 percent of those interviewed agreed with the statement that “the activities funded by the grant largely reflected what KTI wanted to do and less of what we wanted to do” while 70.3 percent agreed with the statement that “the activities funded . . . reflected what our organization wanted to achieve.” However, nearly a quarter said “it was a mixture of both. (see Annex III, Table 21).” KTI’s program of small grants also revealed new partners, ideas, or critical follow-up activities. PDOs were always on the lookout for new ideas and to develop extensive networks and flows of local information. They try to match good ideas with people and organizations that have the capacity to implement them. In this way KTI actively nurtured the development needed civil society activity in selected rural areas outside the realm of most Nairobi-centric aid programs. In the process of working closely with grantees PDOs gain intimate knowledge of local conditions in their areas. One page concept papers are entered into the KTI database and then forwarded to Nairobi where the entire KTI leadership—the CR, DCR, CoP, and other 16 KTI was originally scheduled to wind up its program in July, 2013, but was extended six months to provide interim support for AHADI, a major and multi-year effort to build capacity among the new county level governments that came into being following the election of county governors and county assemblies for the first time in the elections held in March, 2013. KTI’s ability to provide timely support during the long multi￾month run-up to AHADI is another example of its ability to move fast when opportunities arise as well as being more nimble than its parent organization. 16 PDOs—read and vet the proposals. The review process normally begins within a week via a meeting or telecom to approve or modify the proposal. KTI’s leadership and PDOs also meet every two to three months to conduct “rolling assessments” of a dozen or more proposals in which they are compared and discussed against KTI’s strategic political concerns. The “rolling assessments” which normally take place at one of KTI’s field offices, facilitate a measure of peer learning amongst KTI staff who attend. The PDO who developed the concept paper with a prospective grantee is expected to defend his or her proposals in front of the entire group before it is given a “yellow light” or tentative approval by the group to develop the idea further. The proposal is then revised, and a budget developed under the guidance of a grants manager (GM) and grants assistant (GA). Once the proposal is fully developed it is “blue lighted” after which those who participated in the assessment must submit final comments within 48 hours. The proposals are then fine-tuned one more time after which they are given final clearance by the CR and DCR to proceed, and “green lighted”. This iterative process of vetting weeds out or modifies proposals that are likely to fail. Indeed, over the life of the program, only 46 or 5.8 percent of the 791 grants disbursed were canceled and regarded as failures—a remarkable achievement given the risk involved in many of the projects funded. Once a proposal is “green lighted,” the grants manager takes over its implementation to ensure that grant activities occur as planned. GMs often visit grantees to provide advice and support, while the PDO works to document and evaluate the activities funded. All critical information as well as thoughts and reflections are entered onto the database so that there is a clear “paper trail” of the process from the initial concept through execution of the activities or equipment funded by the grant. Proposals for costly activities where the amount to be risked is greater, are subjected to one additional iteration of review. If the amount of the proposed grant is over $100,000 the proposal is forwarded to OTI Washington for final scrutiny and approval. The vetting process from the initial writing of the concept paper to disbursement can take anywhere from a few days to two months depending on the complexity of the project and the issues that arise at each stage. This is significantly faster than a typical USAID project where the time frame often takes six months to a year or more. As noted previously, the process is also time consuming, and personnel intensive with respect to staff time relative to the dollar cost of the grants, but the tradeoff is that disbursements are rapid and most grants achieve their stated purpose. Because nearly all grants are made “in-kind,” each office has a Procurement Specialist (PS) and Assistant (PA) to deliver what is needed for the funded activities to take place, and to ensure that USAID’s procurement rules are followed. Together with the GMs the procurement staff must also communicate the rules to all grantees to avoid misunderstandings. This system does allow USAID to support grassroots groups or movements without registration for important activities that would not be possible under a traditional USAID program. It also frees up time for grantees to focus on the actual work. Reliance on in-kind grants seems to work particularly well for new groups that are under resourced, and for umbrella organizations or movements that might be more prone to conflicts over resources. On the other hand, the use of this mechanism limits the opportunities for grantees to develop their capacity for financial management. There is a clear tradeoff. The reliance on in-kind 17 grants can also be cumbersome for organizations with well-developed institutional structures and procedures for managing their finances. 2.4 Adjustments in Approach As the situation in Kenya returned to an uneasy normalcy by the end of 2008, KTI’s initial objectives were reformulated in August 2009 to focus more on structural reforms and accountability and less on PEV recovery.17 The shift was also encouraged by the annual internal program performance review (PPR) which suggested that KTI be provided with an additional $5 million in funding for FY2010 and another $5 million for FY2011 contingent on its adoption of a “coherent and useful evaluative system,” “an operational scaling up,” and “a refinement of its strategic vision and objectives to address the imperatives of Kenya’s evolving political environment in year 2” (USAID/OTI/Kenya 2009). KTI thus reformulated its goals to include the following: First, the enabling of “public institutions to undertake reforms and to manage instability and uncertainty.” Second, the mobilization of “the public, youth and key change agents to organize for change and to demand accountability and reform.” The two goals reflect a two-pronged “insider-outsider” approach of working with government reformers on the one hand, and creating civil society pressures for continued change including a broader transformation of citizens’ attitudes towards governance on the other. KTI took a leading role among U.S. government agencies in supporting the peaceful referendum of the new constitution held on August 4, 2010, by assisting civic education and mobilization of the Kenyan public.18 In 2011, KTI added a third objective that focuses on fostering identity and self-confidence among “at risk youth” in the predominantly Somali area of Eastleigh in Nairobi.19 KTI spent its last two years in Kenya focused on sustaining key reforms of state institutions, Eastleigh, the 2013 elections, and recently (beginning in May 2012) xenophobia and related issues in the Coast province. KTI has also addressed the threats to stability arising from the cases of Uhuru Kenyatta and William Ruto now before the International Criminal Court (ICC). The need for continued efforts on peace-building and entrenching the constitution, especially during the run-up to the national elections held in March 2013, prompted USAID/OTI to extend the KTI program through the elections. Though originally scheduled to close its doors in July 2013, the program was extended through the end of the year to lay the groundwork for USAID’s intention to support Kenya’s new system of devolved government at the county level beginning in early 2014. While this work was funded by the USAID mission to Kenya and thus not formally part of the KTI program, it enabled KTI to make a smooth exit by building 17In practice, these are intertwined and a shift away from recovery from PEV can lead to some gaps such as less attention on solving displacement and reconciliation issues in any deep and concerted way (See Klopp, Githinji, and Karuoya 2010). 18 The constitution was approved by two thirds majority and promulgated on August 27, 2010. Since then its provisions and their implementation have become the standard for future reforms. 19 This evaluation was not required to focus on this additional objective since this was the subject of a separate evaluation 18 upon its earlier work in conflict prone areas of Kenya, and setting up the mission to “carry the baton” after KTI is gone. As with traditional USAID programs, the success or failure of KTI depends on the quality and skill of its implementing partner (IP), or contractor. Although all grants disbursed under KTI are reviewed and approved the by the CR and DCR, the IP plays the lead role in the day to day operations as it is responsible for keeping this grant disbursal system and the funded activities working. A key challenge is the careful management of systems of accountability including monitoring and evaluation. How this is done matters a great deal, especially given broader concerns about OTI’s accountability and transparency (Lawson 2009). As noted earlier, Development Alternatives Inc. (DAI) was the IP for KTI from May 2008 until May 2011. A number of issues with respect to management style and approach arose during this period, including the manner in which grants were disbursed to grantees. A comparison of the modus operandi of the two IPs and their strengths and weakness is found in Section IV.10 below. 19 III. APPROACH AND METHODOLOGY This evaluation aims to provide some needed learning from the KTI experience and OTI generally for the broader debates on the different modalities for delivering U.S. foreign assistance, and their role in advancing U.S. foreign policy objectives. It also aims to bolster monitoring and evaluation (M&E) efforts at OTI which, in respect to Kenya have been regarded as the “Achilles’ heel” of the program and “the sources of one of its biggest criticisms” (USAID/OTI/Kenya 2011). In this regard, the unconventional work that KTI pursued and the political dynamics it sought to influence, raise both philosophic and policy issues for OTI and USAID that are relevant beyond Kenya. As specified by the SOW and noted in the introduction, this evaluation was conceived from the outset as a two part exercise involving two separate periods of field investigations in Kenya and two write-ups of the findings from those inquiries. Part I was a mid-term evaluation for KTI which reviewed the program from its inception in May 2008 through September, 2012. The mid-term evaluation is reported in (Barkan, Klopp and Ng’ethe 2013), and was based on field investigations conducted by the evaluation team from August 27th through September 28, 2012. Part II is the final evaluation based on the first round of field investigations in Kenya supplemented by a second conducted from June 17th through June 28, 2013 as well as by follow up interviews in Washington in July. The first round of field investigations consisted of interviews by the evaluation team with all senior staff and key subordinate staff working for USAID/KTI or its implementing partners. These interviews were conducted in Nairobi, and during site visits to KTI’s field offices in Eldoret, Kerichio and Mombasa. Interviews were also conducted with 41 leaders of a random sample of 37 organizations that had received grants from KTI.20 Because of considerations of distance (and hence travel time) from each of the four field offices visited by the team to the location of the grantee, not all grantees included in the random sample were reviewed for this evaluation. The team also met with senior officials of USAID/Kenya, U.S. embassy officials, and prominent Kenyans knowledgeable about the KTI program and/or Kenya’s evolving political situation at the time. The first round of field investigations were preceded by meetings with relevant USAID/OTI staff in Washington, including a “boot camp” organized by USAID/OTI for members of the evaluation team to familiarize them with the nature and purpose of OTI as a “sub-agency” within USAID and its modus operandi in the countries where it works. The team also read relevant USAID/OTI and KTI program documents prior to starting its field investigations, and was provided with a copy of the KTI data base consisting of a detailed record of each grant made during the course of the program. The random sample of grants 20 Interviews were also conducted with the leaders of six recipient organizations which the country director and deputy country director of KTI asked the evaluation team to meet during the first round of field investigations in September, 2012. One of these organizations was also included in the random sample of the grants selected for study while the others were not. Although the director and deputy directly clearly wanted the team to review these grants because they regarded them as among the most successful of the KTI program, the evaluation team found their experiences with KTI to be similar to the grantees chosen randomly. 20 and thus grantees considered for this evaluation were drawn from the data base as it existed in September, 2012. The second round of field investigations consisted of return interviews with all senior staff and selected key subordinate staff of USAID/KTI and Chemonics Inc. While interviews were again conducted with a small number of grantees, this was not a major aspect of the second round, because the evaluation team had designed a phone survey of all grantees which was conducted during the last two weeks of February, 2013. The purpose of the survey was to supplement the “face to face” interviews conducted in September, 2012. Although enumerators for the survey were only able to locate and interview 178 present or former grantees out of a total of 309, 21 the survey provided the evaluation team with a much larger data base of grantee views than the interviews conducted during first period of field investigations. Principal findings from the survey are reported in the next section of this report. A complete presentation of all findings from the survey is found in Annex III. Unless indicated, all interviews were conducted on a confidential basis to obtain candid reflections on the interviewees’ experiences with KTI. Nearly all of the interviews took an hour or more to complete. During the course of the periods of field work in Kenya, the members of the evaluation team also met with a significant number of prominent Kenyans not directly involved with USAID or KTI to better understand the current political situation in the country. Considerable time was devoted to this effort to better understand the likely challenges facing Kenya between the first and second round of field investigations and after the 2013 elections. A complete list of all persons interviewed in both the first and second phases of the field interviews is provided in Annex I. The evaluation team expresses the appreciation to those individuals who made themselves available to the team and answered the many questions it posed. To understand the method and limitations of the sampling process it is important to remember that the grant, not the grantee or individual recipient of a grant, was the unit of analysis for the selection process. Once a grant was identified for review, the team then interviewed one and often two members of the organization that received the grant. More than one third of these individuals and their recipient organizations had received two or more grants from KTI (See Annex III, Table 2).22 Interviews with these individuals thus explored their entire experience with KTI and were not limited to their experience with the grant included in the sample. While this provided more detailed information for the team, such information may have caused the team to focus disproportionately on grants and grantees who had been successful rather than those that had received only one grant or failed. That said, some grantees who had received two or more grants had received grants under both KTI-1 and KTI-2. These grantees were able to compare their experiences between the different implementing partners engaged 21 This is a response rate of 53 percent which is regarded as quite high for phone surveys, and which was achieved by requiring the survey enumerators to make at least three attempts to contact each grantee. Most of the attrition was the result of grantees changing their mobile phone numbers and email addresses from those recorded in the data base of grants. 22 As of October 16, 2013 149 of 426 grantees or 35.0 percent received two or more grants. 21 for first and second phases of the program. Interviews with grantees also lasted an hour or more and thus provided a rich narrative of their experience with KTI. Before turning to the findings of the evaluation, a word must be said about the extent to which the evaluation team was able to “grade” the performance of the program and ascribe “success” or “failure.” Before the team interviewed KTI staff and grantees, two realities were apparent that would limit the “judgment” of the evaluation. First, that it would be very difficult to measure the impact of either phase of the program or the program overall. This is because the evaluation commenced four years after the start of KTI making it impossible to conduct a pair, or series of assessments, on a “before” and “after” basis. No benchmark or “baseline” measures of conditions in Kenya prior to KTI were available to the evaluation team other than the initial and qualitative assessment conducted by OTI of the situation existing in Kenya immediately before the start of the program in 2008. Nor, apart from the periodic and qualitative reviews of the program’s progress, was there a second “mid-term” benchmark of measures for the period immediately before the start of KTI-2. The result is that while this assessment is more systematic in its approach than previous reviews, it is nonetheless what it is—a systematic, but qualitative determination of the extent to which KTI-1 and KTI-2 and the activities undertaken in each have addressed the challenges and objectives posed in the original project documents. We are, therefore, able to assess whether such activities and other interventions occurred, whether the procedures of KTI generally raised the quality and likely impact of these activities, and their perceived impact as articulated by KTI staff, KTI grantees, and outside observers. We are also able to make a general assessment of whether from the perspective of KTI staff and grantees, the interventions mounted by KTI succeeded or failed. And, we can make some observations of the likely legacy of the program after KTI ends. But we cannot quantitatively measure the extent of success or failure. A second and more frustrating reality and limitation, is that the evaluation team cannot establish a causal relationship between KTIs efforts in Kenya and the presence of peace and stability or the absence of conflict and violence in those areas of the country where KTI has made its grants over the past five years. A major (though not the only) goal of the program was to address the post-election violence of early 2008 and prevent a repeat of such violence over the course of the program including the current run-up to the 2013 elections. We can thus make a reasonable judgment of whether the activities undertaken under the program were more or less likely to prevent or mitigate such conflict, but we cannot say definitively whether they did so or failed to do so. For example, the presence of peace, albeit tenuous, in the areas within which KTI mounted its efforts in Uasin Gishu County cannot be attributed to specifically to KTI, because these efforts as they constituted only one set of several different interventions by KTI and other actors addressed to this goal. Conversely, the incidents of violence that occurred in Eastleigh, Garissa, and Tana River during the run-up to the 2013 elections, cannot be attributed to the presence or absence of KTI. KTI has been active in the first, but not in the second and third. Indeed, as discussed in Section IV, there are at least seven explanations for why the level of violence that accompanied the 2013 elections was both substantially lower and localized compared to what occurred in 2008. It is important for readers, particularly senior administrators at OTI and elsewhere at USAID to appreciate these constraints before considering the remainder of this analysis. 22 IV. FINDINGS, IMPACT, AND EFFECTIVENESS 4.1 Objectives and Strategy Overall, the KTI program and its components were aligned with both its specific and overarching objectives. Consistent with the modus operandi of OTI programs generally, KTI was flexible, nimble and responsive in seizing critical windows of opportunity. It moved fast. In most cases, the grants were targeted to stabilizing the local situation and conflict mitigation though not all activities did so directly. Some were directed towards bolstering Kenya’s democratic reforms, including support to the judiciary, parliament, and the Commission for the Implementation of the Constitution (CIC), though these too arguably contributed to or ultimately will contribute to political stability and a more peaceful environment. From the above it is evident that KTI made a concerted effort to frame its objectives, strategy, and program characteristics in a manner intended to create convergence of all three. The program reflected all the features of OTI. It has been thoughtful, field-driven, innovative, self-critical and has been working with non-traditional partners of the donor community, especially emergent civil society organizations (CSOs) in conflict-prone areas. Unlike most donor programs that support civil society, KTI reached deep into the rural areas where civil society is weakest rather than simply pour more resources into the “cottage industry” of CSOs that lives off the donors in Nairobi, but does not project a “rural reach.” Our assessment of KTI’s overall and specific objectives is that they were informed by the right diagnosis. From the beginning to the present, the program was opportunistic and pointed in the right direction but also more ambitious in its goals than a single short term program of its nature could accomplish. Supporting stability with unity in Kenya is an enormous challenge, and is in many ways “the Kenya political problem.” Creating a modern and democratic political system is an even more daunting task though it is not uniquely Kenyan. KTI realized this and, therefore only sought “to contribute” to the overall effort. With regard to these two broad objectives—maintaining peace and stability and advancing the process of democratization—the evaluation team found that at both grassroots and national levels, KTI has largely met these by following its stated strategy; executing a program that reflects the desired program characteristics; and closely matching the program interventions with the OTI approach. Critical to the conceptualization and implementation of the program has been the identification and taking advantage of unanticipated targets of opportunity. To cite but four examples: the decision to fund the broadcasting of proceedings from the National Assembly; the decision to support the computerization of court records by the magistrate courts in Eldoret and Mombasa; the decision to support the work of the Constitution Implementation Commission; and the decision to close out the KTI program by laying the groundwork for USAID’s future efforts to build up the administrative capacity of the new county level governments, the key to devolution in Kenya. There are many others. 23 4.2 Program Characteristics At the national level, four related windows of opportunity have presented themselves and KTI developed interventions to take advantage of all four. These are (1) the 2007/8 post-election violence (PEV) which arguably posed the greatest threat to Kenya’s stability since independence; (2) the national referendum in 2010 which resulted in the adoption of a new constitution that greatly enhances the prospects for a successful democratic transition; (3) the pre-trial hearings of the International Criminal Court (ICC) that resulted in the indictment of three well known leaders and one media personality, which still has the potential to trigger violence similar to the PEV23; and (4) the 2013 national and local elections which risked the prospect of national disintegration, but which ended peacefully despite numerous flaws in their administration by the independent electoral and boundaries commission (IEBC). KTI moved into the country fast soon after the 2008 PEV to take advantage of the first window, and has correctly modified its program to address the others ever since. The result is that when taken all together, KTI’s diverse program of in-kind grants reflect OTI’s intended operational characteristics well. In some cases, KTI has taken advantage of local opportunities to initiate discussions with a potential grantee. As previously noted, KTI usually moved fast with respect to identifying and funding a worthy intervention. The fast speed was reflected in the average two to four weeks lapsed time between the identification of an activity and the disbursement of in-kind support, particularly during KTI-1. The relevance is evidenced not only in the composition of the entire program portfolio as shown in the data base, but also in the timeliness of the grant disbursements and the initiation and implementation of the funded activity on the ground. With one or two exceptions of grants that were cancelled, all activities examined for this evaluation were deemed as timely and relevant by the grantees. Not a single grantee regretted having undertaken an activity on the basis of bad timing or irrelevance although some KTI-2 grantees expressed frustration that the review process dragged on over a longer than necessary. In our telephone survey of grantees, 98.8 percent of those responding said they would work with KTI again (Annex III, Table 24) As noted above, more than one third of the grantees received two or more grants. Some received as many as seven. When asked why this happened, recipients of multiple grants said that the extensions were the result of the activities funded by the first grant going well, and giving rise to new activities conceived by the grantee and/or KTI that required funding and deemed consistent with KTI’s goals. The CR and DCR concurred with this observation. Most grantees interviewed would like further funding. As suggested above, we therefore conclude that the grantees that received additional grants were those deemed particularly “successful” or worthy by KTI on the basis of their performance with their initial funding.. Perhaps most important, this approach reflects KTI’s position that the grantees are not mere recipients of funds but CSOs contracted to implement KTI’s intended activities as well as their own. It is a partnership beneficial to both sides. As discussed in Section II, KTI, consistent with OTI practice elsewhere, provided most of its grants (78.7 percent of the grantees surveyed) on an “in-kind” basis (Annex III, Table 20a). 23Because the accused claim their indictments are part of a broader Western conspiracy against Africans, it is possible that violence will also be targeted against foreigners. 24 We found this form of support to work well for the very reasons intended as the mechanism enabled KTI to support emerging CSOs and related organizations that were not “the usual” recipients of donor support as well as being organizations in areas of Kenya prone to inter￾ethnic conflict, particularly the western Rift Valley. Disbursements through in-kind grants reduced the accounting load of grantees, increased the level of interaction between grantees and KTI staff, and enabled KTI to utilize resources efficiently and flexibly.24 The KTI program has demonstrated flexibility in a number of ways. First, funding ideas have come from both the potential grantees and KTI staff. This does, however, require a high level of interpersonal skills on the part of KTI staff and especially its implementing partner. Such skills build mutual trust between KTI and grantees. Neither side perceives the other as pushing its own agenda though in reality KTI has contracted the “grantees” to implement its program. A minority of grantees, however, did express the view that their original ideas had been modified or “panel beaten” to suit KTI goals. That said, the evaluation team found that the majority of funding ideas were initiated by the grantees, usually in response to urgent local circumstances, and that the activities ultimately funded by the grants reflected their objectives. In response to the question “Did you or your organization approach KTI for funding or did KTI approach you?,” 73.0 percent of the respondents in our telephone survey indicated that they had approached KTI (Annex III, Table 17), while 70.3 percent reported that the activities funded “largely reflected what our organization wanted to achieve. Twenty-four percent reported “it was a mixture of both” (Annex III, Table 21). Second, the two sides negotiate the shaping and final budget aspects of all grants though KTI’s grants managers obviously have more say in the content of the final grant agreement. In a few cases, this led to grantees refusing to sign the proposed grant until their demands were met by KTI. Some of the reasons for this unhappiness are discussed below where we compare the contrasting operating styles of the two implementing partners (IPs). Third, KTI can request a grantee to undertake an added activity or a new activity if new circumstances develop such as threat of violence in an otherwise peaceful area. Fourth, grants can be amended or extended if new circumstances warrant though the scope for modification seems to have narrowed during KTI-2. This helps the grantee and the activity it is carrying out to maintain real time relevance, which is crucial for preventing violence. The evaluation sought to find out whether the program is field driven as intended. We concluded that notwithstanding these caveats, it was. This has been achieved mainly through sustained interaction between the CR, DCR and CoP and the managers of the IP in the field, particularly the Program Development Officers (PDOs) and the grant managers (GMs). Key to this are the rolling assessments discussed above and the iterative review process underlying the development of each grant. 24 Grantees are also relieved of the audits associated with conventional USAID programs, as it is the IP not the grantee that is subject to USG audit. 25 4.3 Monitoring and Evaluation, the Database In addition to the rolling assessments and “yellow” through “green light” reviews discussed in Section II, the top KTI management, PDOs, and grant managers, interact regularly with the grantees through field visits. The implementation is, therefore, labor intensive which in turn provides a highly personalized and continuous monitoring and evaluation system that is quite different from traditional M&E. Put differently, PDOs and grants managers know “their” grantees well. A related characteristic of this monitoring is that KTI personnel, especially the PDOs, are always “scouting the field” for possible areas of intervention. The program is also field driven in another more traditional sense: the field officers are required to regularly file detailed field generated data that are then entered into the data base, including an extensive online accessible record of each grant. Though this field driver is increasingly common in donor programs, we were struck by the level of detailed information down to the mobile phone number and location of each grantee. Indeed, field meetings by the evaluation team with the random sample of grantees selected for contact confirmed that PDOs and/or grants managers from the relevant KTI field office had visited grantees in their home areas which were sometimes quite remote, and “grassroots” in terms of the level of Kenyan society at which they operated. PDOs and GMs clearly made considerable efforts to meet grantees in their own communities. Good as the database is, it could be further improved and perhaps drive the field process more effectively if it were designed to facilitate quantitative and comparative analyses of the data overall. For example, grants, their purpose and characteristics, and the characteristics of grantees could be assigned numerical codes to facilitate quick comparisons, including statistical analyses, of different types of grant. The current database is rich in narrative descriptions of what each grant entails, but this information could be analyzed more effectively and yield better lessons learned, if it were possible to sort grants into various categories based on their characteristics and then compared. For example, are large grants more effective in respect to meeting their objectives or some other criteria than small? What about grants given to organizations, headed by women compared to those headed by men, and so forth? All grants are assigned the geographic coordinates of the location of their activities which is a major advance. However, until the geographic locations of other data such as the locations of political violence is also included in the database, it is impossible to test for a relationship between KTI interventions and the decline of violence. The same goes for the geographic locations of other relevant data. It should also be easier to compute the percentage of grants of different types, their average budgets, etc. Grant activities could also be assigned a numerical grade at each stage so that the trajectory of individual grants or groups of grants could be ascertained. Put differently, as currently designed the database contains much in the way of useful and detailed information, but not in a way that is user friendly to conduct macro-level analyses of the entire database and monitor the evolving “big picture” of the KTI program from thirty thousand feet. Instead, a description of each grant is considered one grant at a time. Entering the same data contained in the data base via one of the standard social 26 science statistical packages such as SPSS25 and/or a popular GIS package such as ArcView would facilitate such analysis. Has the program been self-critical and if so how? The program has provided a forum for self￾criticism primarily through the rolling assessment meetings. The evaluators attended one such meeting and were struck by the level of candor exhibited at the meeting, the informal and free atmosphere, and the focused nature of the discussion. However, the agenda was too packed which meant that the discussion of some issues was rushed and some project proposals not considered. In addition to the rolling assessments, KTI has conducted its own annual internal evaluation known as the Program Performance Review (PPR). Discussions with the top KTI management suggested that some lessons have been learned from these internal evaluations. However, the present study is the first externally led and multi-part evaluation of an OTI country program.26 A major outcome of the program’s “field centeredness” is its high level of responsiveness to local changing conditions. The timing of the program’s entry into Kenya was itself in response to broad political events including the trauma of the 2007/2008 PEV. The alignment and realignment of program objectives was also in response to new circumstances, the details of which are constantly unfolding. At its best, responsiveness anticipates what is likely to happen and prepares to respond quickly. Though formal scenario building was not a feature of the program, informal scenario building was found to be part of routine discussions among KTI personnel: What if the courts had barred Uhuru Kenyatta and William Ruto from running for president and deputy president before the Marc, 2013 elections? What if the Coast region had erupted into full scale violence? What if the 2013 elections had completely collapsed? What if the Kenya government fully implements the governmental structures (particularly devolved government at the county level) as prescribed by the 2010 constitution)? Beyond these big scenarios, it was evident during the rolling assessment that the PDOs and the grant officers are encouraged to “think scenarios,” that is to say, how the grant proposals they are considering mesh with the changing political realities. The program has also been responsive in that it has developed new activities in the perceived future trouble spots such as the Coast Province and the Eastleigh area of Nairobi. A notable feature of the program is that it has been politically entrepreneurial as manifested by taking calculated political risks. Though many of the funded activities are not overtly political, they all have a political component insofar as they seek to empower hitherto 25 Statistical Package for the Social Sciences. STATA would be another example. 26There have other been external evaluations of parts of KTI-1 and KTI-2 including the “peace caravans” and of the “road shows.” There have also been internal reviews and stock taking of lessons learned, most notably the KTI Legacy Report (USAID/OTI 2013a) which sought to chronicle the major accomplishments of the program over its entire five year period and Lessons Learned from the Kenya Transition Initiative (USAID/OTI 2013b). Neither could be considered an evaluation in the formal sense, and the evaluation team regarded the first as an exercise in self-congratulatory public relations. By contrast the lessons learned exercise and resultant document was a thoughtful and useful analysis that identified several issues similar to those by the evaluation team in its mid-term assessment and which are explored further in Section V of this report. 27 powerless groups; educate various groups on their rights; provide voice forums for the voiceless; cultivate alternative leadership; and persuade groups to conceive alternative forms of political affiliation and association. In a country where the political leaders mostly mobilize the electorate on the basis of their ethnic identities, one of the most significant aspects of the KTI program is its attempts to inculcate a sense of national and public good and rational calculation within what is often a zero-sum struggle for power. Grantee programs financed by KTI seek to teach people how to identify and pursue their self-interests in a manner that strengthens a democratic constitutional order, toleration of difference, rule of law and a more level playing field for most citizens. All this is “political” even when it is not explicitly labeled such. Some of KTI’s activities, however, have been overtly political and, therefore, politically risky. Youth programs intended to develop alternative leadership best exemplify this type of activity, with the most notable example being the high profile Yes Youth Can campaign which annoyed both former President Kibaki and former Prime Minister Odinga.27 Activities that have taken on powerful local political leaders and their patronage networks have also entailed taking calculated risks. The evaluation team encountered a number of cases where it regarded the risk as too high, because the activities funded put grantees’ in physical danger. Whether the activities were low or high risk, it was evident that KTI fully understood their political nature. In other words, the evaluation team did not encounter a situation in which KTI “found itself” without having decided to “knowingly put itself there.” While political entrepreneurship is an unavoidable (and desirable) feature of the program, it still requires considerable “political savvy” and analytical skills to execute successfully, and there is always a risk of failure with real and serious costs. Being “right” 80 percent of the time, is not good enough when lives or the physical safety of grantees become an issue. 4.4 Non-Traditional Partners—Rural Civil Society Organizations (CSOs) and “Youth” KTI has succeeded in working with non-traditional partners to implement its objectives and activities. KTI has also worked with well-known CSOs as well as government bodies, including district peace forums, the National Assembly, and the Commission for the Implementation of the Constitution (CIC). However, the largest number of grants were made to less well-known rural-based CSOs rather than to the better known Nairobi based groups normally favored by international donors, partly on account of their articulate leadership and urban base. KTI has generally given much greater emphasis to the grassroots level on the correct assumption that creating bottom-up demand for reforms is the best assurance that the reform agenda will not be derailed. At the same time, this two-level “top-down/center-out” and “bottom-up” approach by the donors and Kenyan CSOs generally has created space for reform discourse and mutual strengthening/comforting between the two levels. The two approaches are complementary, and dictated by a political environment that is still not entirely comfortable with the reform agenda. 27 Though not considered to be a program initiated by KTI or part of its core program, KTI provided start-up funds to launch the campaign. 28 The complementary nature of this two-level approach was illustrated by several cases where grassroots organizations supported by KTI were working with national bodies such as the National Cohesion and Integration Commission and the District Peace Forums. This has, among others, broadened the voice for reconciliation though actual reconciliation is yet to take place. It has also given grassroots agents demanding reform an avenue for voicing the underlying causes of instability, something they are more knowledgeable about than the national agents based at the center (i.e. in Nairobi). Indeed, the evaluation team was often struck by the depth of knowledge of the local agents and grantees of the political problems of the country, especially their manifestation in the local areas in which they worked. Some, if not most, have found a way of getting this knowledge amplified through the national reform agents. A review of the program portfolio reveals that not all grants have targeted immediate stabilization needs. Grants targeting post-PEV reconciliation self-evidently target stabilization. Others target longer term issues that are the underlying cause of instability and, therefore need addressing under the Agenda Four reforms of the National Accord. We also found that nearly all grantees have some “youth relevant” activities, regardless of whether their grant specifically targeted the youth. In all cases, grantees are deeply concerned about ”the youth problem” be it youth unemployment or the tendency of the youth to be “bought” (i.e. manipulated) by the political leaders.28 They want to see the youth play a bigger role in driving the reform agenda because the youth are seen as the main beneficiaries of reforms and the main threat to stability in the absence of reforms. Given that Kenyan youth are beginning to demand accountability and acceleration of the pace of reform, KTI has done well to target the youth reform agenda both as an immediate stabilization measure and as long term agenda.29 4.5 Criteria for Awarding Grants and Selecting Grantees Critical to the effectiveness of KTI program is the criteria for selecting grantees. We probed this issue and concluded that no serious mistakes were made during this process. On the other hand, the evaluation team noted that there was no list of specific criteria guiding the selection of grantees. Perhaps this reflects the need for flexibility to identify unique targets of opportunity? KTI’s broad objectives are obviously an important consideration. However, the selection can be quite dynamic depending on field conditions, and opportunities of the moment. Much depends on the individual judgments of the PDOs, the CoP, the CR and DCR. At the rolling assessment attended by the evaluators, the selection process was based primarily on a series of pragmatic issues confronting the project rather than on program 28 The meaning of the term “youth” in the Kenyan context does not mean the very young or adolescents, but rather young adults from the ages of roughly 16 to 35. Many of these have entered the labor market, but their futures are bleak because of limited education or skills and also high rates of unemployment in the country. 29 Notwithstanding its targeting of “youth,” KTI has not, apart from initial start-up support, been involved in the Youth Bunge program mounted by the USAID mission to Kenya with encouragement from the U.S. Embassy. 29 objectives alone. The major considerations included the objectives and merits of the proposed activity, for example, its relevance for conflict prevention; whether the concept for the project had been initiated by KTI, whether the prospective grantee was the most appropriate organization to carry out the proposed activity, whether there were previous grantees working on the same issue in the same geographic location, etc. Because there was no specific list of selection criteria, the PDOs had to be good in “selling” the merits of each proposed project against the suggested criteria raised during the assessment. That said, and as discussed in Section II, the process of group discussion and iterative review plus the fact that the CR and DCR sign off on every grant meant that all proposals were thoroughly vetted before the development of the specific grant budget which itself entailed an additional cycle of review. Overall the system worked in that there were very few cases of KTI having funded the wrong activity or grantee. 4.6 Major Accomplishments In summary, the major achievements of the KTI program are six: (1) Consistent with OTI’s purpose, KTI was timely—it entered the Kenya at the right time and fast off the mark, to a degree not possible by other offices comprising USAID/Kenya. (2) The objectives and strategy were based on the correct diagnosis of the local political environment, and the process of implementation has been adjusted to reflect evolving circumstances since the program commenced five years ago. (3) Notwithstanding its emphasis on quick short term interventions, KTI has recognized that the causes of instability are both short and long term, and adjusted its program accordingly. As the program progressed, KTI also recognized that as the prospects for instability decreased, the demand for political reforms rose, and that KTI could (and ultimately did) contribute to the process of democratization in the country. (4) KTI correctly channeled support to a new form of civil society that is rural based and located in areas of Kenya that are most prone to conflict. As such KTI also nurtured the development of civil society in areas typically neglected by USAID and other donors. (5) KTI managed to operate an inherently political program in a charged political context without attracting open political attention and criticism that would most certainly have undermined the program.30 While some individual grantees were put at excessive risk, the program itself was never threatened. Lastly, (6) and most significantly, KTI made a significant contribution to creation of Kenya’s social capital and organizational capacity with the result that KTI has most likely made a lasting and long term contribution to Kenya’s democratic transition. As both the U.S. ambassador to Kenya, and the USAID mission director remarked in separate meetings with the evaluation team, “if OTI/KTI did not exist, we would have to invent it.” From the above narrative it is evident that the KTI objectives have been highly relevant to the specific political context of the country since 2007. Following the 2007/8 PEV, the most pressing problem was to contain the violence and then prevent the outbreak of further violence. The epicenters of the violence were the Rift Valley and Nyanza, and this is where KTI concentrated its initial activities in Kenya. KTI moved in unnoticed by the public to 30 One reason for avoiding unwanted attention is that unlike USAID missions generally, KTI has not been required to brand its support as assistance from USAID or the American people. By taking a lower public profile, it has arguably been more effective than had this not been the approach. 30 work with existing local groups. And in June 2012, the destabilizing potential of the deep grievances around land and economic and social inequality on the Coast and the related rise of the secessionist Mombasa Republican Council (MRC), led KTI to establish a program there. Since the beginning of the program, KTI has periodically adjusted its interventions to the changing political landscape in Kenya, and taken advantage of the different windows of opportunity and challenges presented. As the post-election violence that shook Kenya in 2008 abated and the country shifted into the Agenda Four reforms including the referendum on the new constitution, KTI adjusted accordingly into reform support and advocacy. During its final year of operations as the political scene was dominated by ICC debate, the 2013 elections, and the establishment of new structures of government after their conclusion, KTI has again adjusted to the changing realities on the ground. KTI has recognized that the causes of instability in Kenya are both short term and long term, social, economic and political and that these causes are interlinked. The immediate trigger of the PEV was the bungled 2007 elections but the underlying causes of the PEV, especially in the Rift Valley are perceived as historical injustices over land ownership as well as a sense of grievance at being displaced from power. The politicians and other leaders in the region exploited these perceived historical injustices to mobilize the youth for short term political gains, a pattern that is repeating itself in the Coast Province. In Nairobi, the initial trigger was the elections, but as in the rural areas where income inequalities are stark the pattern of violence also involved socio-economic dimensions. KTI has strived to address both short term and long term causes of instability while at the same time recognizing that most of the long term causes as well as institutional reforms are beyond its formal mandate. KTI has therefore, concentrated the largest portion of its early grants on short term stabilization and conflict mitigation activities—an approach described by some of our interviewees as prophylactic. At the same time KTI, beginning in 2009, commenced a partial shift to sensitize the public to the dangers inherent in not addressing the underlying long term causes of instability, which are rooted in authoritarian governance that deepens problems arising from the unequal distribution of resources rather than solving them. KTI has also supported long term institutional reforms by making a small number of large grants to Parliament, the judiciary and the Commission for the Implementation of the Constitution (CIC). In seeking to strengthen these institutions, these grants arguably reduce the prospects for instability albeit over the long run. Has instability decreased in the program areas? According to those interviewed the answer is “yes,” but this is the answer one would expect from both KTI staff and grantees. More objectively, cases of actual reported violence did decrease over the life of the program in the areas where KTI operated, though this assessment is based mainly on reports in the media. But as discussed in Section III, it is very difficult to attribute this decline specifically to the KTI program, or to any other donor. Furthermore, even in those areas where violence declined, tensions between former protagonists remain high, and a small incident can trigger inter-ethnic confrontation. As one senior Kenya government administrator put it, “in these areas, no crime is committed by an individual as everything is seen as ethnic group crime.” 31 One of the most salutary features of Kenya today is the notable increase in the public demand for good governance and accountability. This is best exemplified by the numerous legal suits against the executive branch, including court challenges of presidential decisions. There is an increase in public demand for reform and accountability which suggests that both society and public institutions have acquired new capacity to assert themselves. However, it is difficult to attribute this “new voice” and awareness to any one program, whether Kenyan or initiated by the international community including KTI. Indeed, to the extent that one can suggest attribution, the rising public demand for accountability is mainly the result of a new generation of Kenyans that experienced new freedoms during the 2000s, and which has higher expectations of government. The ratification of Kenya’s new constitution in 2010 that embodies these new expectations has also emboldened the public to openly seek better governance, and changed the public narrative. A critical outcome of the constitution is the reform of the judiciary currently under way led by the Chief Justice. This has given the public hope of a “fair hearing” against public institutions, including the hitherto “untouchable executive” (although this did diminish in some quarters with the Supreme Court’s ruling over the election). With reference to the probable effects of the KTI program, one can say that the program has probably had a positive effect on demands for public accountability in some specific localities, but this is a subjective judgment, and the absence of any “before” and “after” surveys of public opinion. Moreover, the rising public demand for better governance is one that, as discussed in Section I, is part of the overall process of democratization in Kenya that has been underway for more than twenty years. The constitution has created a plethora of new institutions all charged with some responsibility for the reform agenda. These institutions are supported by the state with only minor support from the donor community. They include some agencies supported by KTI including the CIC and the National Cohesion and Integration Commission (NCIC). KTI has also supported older albeit previously weak institutions such as the National Assembly which have begun to assert themselves as a check on executive power in recent years. Those institutions that have received KTI support have certainly increased their capacity to undertake reforms, most notably parliament with the KTI grant to facilitate live broadcasting, 31 and the law courts in Eldoret, Kapsabet and Mombasa which established the first computerized court case record system in Kenya with KTI support.32 About 40 percent of KTI grants have, however been made to small CSOs, which means KTI’s direct impact on core public institutions has been promising “pilot projects,” but limited in scope.33 31 Though timely and highly valued, KTI support to the National Assembly must be viewed as a supplement to the long term program of support by USAID and DfID that began in 2001 and for which the continuing implementing partner has been the State University of New York. That said, KTI’s support to parliament is yet another example of the program’s opportunistic approach. It identified a need which in this case the main USAID/DfID program could not meet, and filled the gap. 32 KTI provided additional support to the magistrate courts in Kericho and Mombasa before winding up. 33 For example, KTI’s grants to computerize the case record system at the magistrate courts in Eldoret and Kapsabit have had a profound demonstration effect of what is possible in the area of judicial reform. They have not, however, transformed the judiciary. 32 Have some of KTI’s programs been more effective than others? This question, together with the question on program impact can only be answered conclusively with data from a formal survey. Most programs were judged by those we interviewed—the PDOs who made the grants and the grantees who received them—to be effective in that they did what they were supposed to do, and because there is some output to show for it. Results from our survey indicates that an overwhelming majority of grantees (86.5 percent) considered the activities funded by KTI to be “very successful” in achieving the activities’ objectives (Annex III, Table 22). These self-assessments are obviously not free of potential bias. However, we concur that most of the programs strengthened civil society in geographic areas where it was previously weak or non-existent. This is especially true of the Rift Valley where development of civil society has lagged behind most other areas of Kenya except the northeast. And even in Nyanza Province where civil society has traditionally been stronger compared to the Rift Valley, KTI provided a new direction focusing on major national themes emerging after the 2007/8 PEV. The capacity to focus is, therefore, another measure of effectiveness and in this sense too, the program has been effective. Using this measure alone, perhaps the most effective program activities were those that forced the civil society to focus directly on prevention of political violence. This is still the dominant issue requiring attention. All those interviewed agreed that the KTI program made some contribution by focusing on the problem, though the extent of the contribution is difficult to measure. 4.7 The Importance of Rural Civil Society for Democratization Has KTI achieved the goal of contributing to Kenya’s transition to democracy? Again, this is a difficult question to answer with precision, but the answer is “yes.” The longer and more nuanced answer is “possibly yes.” KTI moved into Kenya at a time when country’s transition to democracy was facing its greatest threat. If the PEV was not contained when it was, the prospects for democracy would have vanished, perhaps for a decade or more. By recognizing the long term threat and engaging quickly, KTI helped save the prospects for democracy. As noted above, KTI then recognized the significant milestones and adjusted its program accordingly. All the milestones—the 2010 referendum on the new constitution, the ICC cases, the rise of xenophobia across the country (by Christians and members of Kenya’s largest ethnic groups against Muslims and Somali-Kenyan citizens as well as the fear of being infiltrated by fundamentalist Islam; against “up country” Kenyans by residents on the Coast, etc.)—have major implications for the future of democracy in Kenya. KTI is engaged in all of them. As discussed above, KTI has invested heavily in supporting a new kind of civil society in the rural areas and this in our view could turn out to be KTI’s greatest contribution to the democratic transition in Kenya. This is because (1) the transition to democracy depends on the demand for it from civil society and Kenyan citizens generally. This has been the pattern in the past—in Kenya and elsewhere—and will continue to be the future pattern. A democratic constitution alone cannot deliver democracy unless it is supported by a high and sustained level of demand for reform. KTI has contributed towards creating such demand, though the level of the contribution is difficult to assess. (2) Demand for democratic reforms 33 has historically come from urban-based, well known and well-funded CSOs led by urban middle class elite. This group has been vocal but lacked extensive linkages or “reach” into the rural areas. Such linkages are critical for a successful transition to democracy, because half of Kenya’s population still resides in the countryside, and because of the limited understanding of modern democracy in these areas. Conversely, the weakness or absence of linkages with the rural areas has made it easy for anti-reform forces to disorganize urban based CSOs. They are often seen as part of a donor driven “cottage industry” and this has made it relatively easy to discredit them. By funding a new and rural based civil society, KTI has begun to alter the architecture of civil society in Kenya. The appropriateness of KTI’s investment in supporting the development of non-traditional CSOs in the rural areas is confirmed by results from our telephone survey. First, the grantees are a talented group. Seventy-one percent are university graduates, those who have received some post-graduate education or those who hold post-graduate degrees (Annex III, Table 5). They are also “cosmopolitans.” Ninety-two percent reported that they had lived one of Kenya’s major urban areas (Annex III, Table 12), while an astonishingly high proportion (88.0 percent) stated that they used the Internet once or more than once a day (Annex III, Table 13a). They are, therefore in a position to understand and articulate democratic ideas and to provide the necessary leadership. They are also a highly energetic and entrepreneurial group. Three-quarters reported that they were or had been involved with other civil society organizations in addition to the organization funded by KTI (Annex III, Table 11), thus bringing considerable networks into their activities. As noted previously, 73.0 percent indicated that they had initiated the contact with KTI, and 48.8 percent reported that their organizations have received grants from other organizations in addition to KTI (Annex III, Table 15). That said, 75.6 percent of those who participated in the survey were male (Annex III, Table 6). Kenya’s transition to democracy will nevertheless take time, indeed longer than most donors and Kenyans are prepared to admit. The emergence of rural CSOs will expand the demand, but the most prominent members of the current political class will not respond quickly. Most oppose or are indifferent towards reform, because it is in their interests to do so. This means the cost of the transition is sure to rise over time, in money and probably in lives. The pockets of resistance remain strong. In some cases there might be good economic and structural reasons to slow down the pace of reforms, especially devolution. 34 4.8 Maintaining a Low Profile One of KTIs achievements which is likely to be missed by a casual observer is that the program has managed to operate for five years without attracting more attention from political leaders threatened by reform. Given that the program is about political change, and given the high level of resistance to reform, one must wonder how KTI has managed to escape more open political attack. It is true that the Yes Youth Can program did attract negative comments from both the President and the Prime Minister, but KTI was only peripherally involved with this high profile program. As previously noted a small number of grantees have been threatened, with one or two forced into exile until the situation surrounding their cases cooled down. But on the whole, the program has run smoothly without overt negative attention. The principal reason why KTI has not attracted attention is that unlike most USAID programs, KTI has not been required to publicize or “brand” grantee activities as programs mounted by support “from the American people.” While the relationship of KTI to USAID is never denied, it is not advertised either. Most grantees know that support from KTI comes from the United States government, and are appreciative of this support. It is the public that is usually unaware of the source of grantee funding as few official acknowledgements are given. 4.9 Explaining Achievements: High Quality Field Staff and Savvy Leadership KTI’s achievements can be attributed to several factors in addition to those discussed above such as identifying the right strategy, targeting funded activities to an appropriate audience, choosing effective partners, sound field orientation grounded in a knowledge of local conditions, self-criticism, etc. While all these have contributed to the achievements, two others need highlighting. First, the quality of key staff, and second, the quality of the programs field agents. The high quality of PDOs, grants managers, and procurement officers has already been discussed. Our survey confirmed our initial impression that these officers are highly effective on the ground with a large majority of the grantees (86.5 percent) indicating that the PDOs were “very effective” in helping make the activities of the organizations better than they would have been had there been no PDO (Annex III, Table 18b). Furthermore, the assistance received by the PDOs was rated as either “excellent” or “very good” by an equally large majority (91.2 percent) of the grantee organizations (Annex III, Table 18c). The story is the same for the grants managers (GMs). Here too, a large majority (78.5 percent) of grantees rated them as “very effective” in arranging for the in-kind support for your organization (Annex III, Table 19c).” GMs received a similarly high rating of “excellent” or “very good” by a combined majority (87.7 percent) for the overall assistance received from the GMs (Annex III, Table 19d). The evaluation team was also struck by the management style, commitment and political savvy of the CR and DCR.34 The CR was often described as the one who provides strategic 34 It is worth noting that the team was composed entirely of three academics committed to being independent and critical thinkers, an MO that sometimes borders on the cynical. 35 direction, and always asks strategy-related questions. The DCR, on the other hand, was described as almost obsessed with operational details. “You cannot get the slightest incorrect detail past him,” we often heard. However, when probed whether the DCR had a tendency to “micro manage” the answer was always “no”—indeed, quite the contrary. Most regarded the DCR as always responsive to whatever query or email they sent his way and that he was thoroughly knowledgeable about their work. He was regarded as helpful, but also a manager who kept staff “on their toes.” The CR and DCR clearly have different skills and operating styles that complement each other. Both were regarded as highly supportive of staff in their own way. Both are also highly committed to, indeed enthusiastic about their work. We were told and observed that they get on well. Together, they have provided high caliber leadership to KTI. 4.10 Comparing Implementing Partners It is clear that KTI’s two implementing partners came with different modes of operation that have impacted on both grants and grantees. Development Alternatives Inc. (DAI), the implementing party for KTI-1 (May 2008 to May 2011), was perceived by both PDOs and grantees as being faster and less rigorous in its review process of grant proposals, and in the implementation of grants approved. During KTI-1 the time between the initial discussion between the IP and prospective grantee and the disbursement of funds or in-kind support was 2-4 weeks. Some of these fast disbursements can be attributed to the fact that DAI started its program soon after the post-election violence when the situation required immediate action. But rapid disbursals also reflected DAI’s corporate dynamics and culture. More emphasis was placed on being in the field and getting activities launched quickly. Large sums of money would be wired to the personal accounts of KTI staff members who would withdraw the cash and physically deliver it to grantees, sometimes in remote places. Although there were no known incidents of fraud or robbery, the procedures were certainly vulnerable to such. Staff members who carried large sums of money also felt unsafe. Grantees who have worked with both implementing partners also found DAI to more generous with travel and other allowances for grantees than its successor. Indeed, a comparison of the average grant budgets for both IPs indicates that while the average grant under DAI was $49,449, the average grant disbursed by Chemonics through August, 2012 was $31,804—nearly 36 percent less.35 In addition, some grantees complained about a lack of flexibility when dealing with the current IP when attempting to move funds from one category of their grant budget to another even though they were well within the spending limit of the grant. These grantees regarded the inflexibility of Chemonics to be unnecessarily rigid and undermined the effectiveness of the grant. The current system while better in terms of accountability than under DAI was also slower through the end of 2012, and a number of grantees complained of the delays.36 At the time 35 A full comparison of the average grant budgets by field office is presented at the end of Annex II. For each field office, disbursements by Chemonics were substantially lower than by DAI. 36 This was also noted in the evaluation of KTI’s support for local FM radio stations. 36 the evaluation team conducted its first round of field investigations in September, 2012, the average time elapsed from the first discussion between the IP’s field staff and the prospective grantee to the actual disbursal of funds had risen to about two months, sometimes more. The evaluation team concluded that while there was a tradeoff between speed and rigor that steps could be taken to speed up the process, and so recommended in its interim report.37 When the team returned for its second round of field investigations in June, 2013 it found that the time period for evaluating grant proposals and making disbursements had been substantially reduced since December, 2012. The change of command also resulted in an increase in the average amount per grant. As indicated in Annex II, the average grant made by Chemonics from September, 2012 thru October, 2013 rose to $45,301, a level very similar to the average amounts per grant under DAI. Chemonics also tried to address the problems arising from KTI staff moving around with large amounts of cash by introducing M-Pesa to make payments to grantees. While this was a good idea, there were many initial hiccups that resulted in delays in payments and thus some tension between grantees and participants in their activities.38 Not surprisingly, grantees that had worked with both IPs reflected more positively about their experience with the first IP (DAI) than with the second (Chemonics). The comparison of the different operating procedures of the IPs also revealed what appears to be a tradeoff between speed in disbursing grants, and accountability. DAI did not work with grantees to set up rules and systems and at times it was probably not possible to track some of the money disbursed. Further, DAI appeared to put little effort in collecting data and documentation overall. By contrast, when Chemonics came in for KTI-2 (May 2011 to present), it tightened reporting and implemented greater cost controls. For those grantees who had worked with DAI, this was somewhat of a shock. Chemonics rarely disbursed cash grants and instead focused almost solely on “in-kind” contributions with strict and careful budgeting. However, in some cases this came across to many as excessive “penny pinching” especially as far as transportation and meal costs are concerned. Finally, some PDOs and grant managers appreciated the increased rigor they were expected by Chemonics to communicate to grantees. They felt that the clear rules insulated them from misunderstandings and certain pressures that can arise with grantees. This resulted in grants of higher quality. On the other hand, they also acknowledged the trade-off between rigor and getting the job done. The procedures and restrictions KTI staff themselves are required to follow to be reimbursed for travel expenses to the field have also increased under Chemonics. This has sometimes increased the time PDOs and GMs need to manage their grants, and discouraged staff from 37 The CR and DRC concurred with this recommendation. 38 The biggest complaint by participants at activities organized by grantees was over the payment of travel allowances. Payment via M-pesa often resulted in participants not receiving their allowances until one or two days after the event thereby creating a hardship for some in securing transport to return home from the event. 37 visiting grantees as frequently as staff did under DAI. Not surprisingly, some staff were not happy with these rules. This situation also points to the need for more flexibility on the part of Chemonics in a way that would not endanger the admirable aspects of its system. OTI might also develop guidelines and lessons learned that it can share with all of its implementing partners in future. When the evaluation team returned to Kenya in June, 2013 it found that Chemonics had responded to its recommendation to speed up the process of reviewing grant proposals and making disbursements without sacrificing careful scrutiny of the process. The team was pleased with the improvement in the rate of disbursements which was due largely to the arrival of a new chief of party (CoP). The CR explained the improvement more in terms of the contrasts in management style between the former and current CoP, rather than of the overarching approach of the IP. Whatever the reality, the point is that the choice of procedures by the IP and how they are interpreted by its key personnel make a difference. Both must be properly calibrated if OTI’s modus operandi of fast project reviews and disbursements is to be maintained. One procedure followed by both DAI and Chemonics was particularly appreciated by grantees. They have been permitted to retain the equipment and materials provided as in￾kind assistance after the end of their KTI grant to use as they see fit. Some of this equipment was expensive or certainly beyond the means of grantee, and included video cameras, TV monitors, projectors, photocopiers, computers and printers. In all cases such equipment continued to be used after the close out of the grant to sustain its operations whether or not the organization had received a follow-on grant from KTI. The evaluation team identified three additional issues arising from the way Chemonics dealt with its staff. First, as is typical for a program in its final year of operation, there are the inevitable wind down problems, including the anxieties of staff who need to find work after KTI. Should they be looking for employment now, and possibly depart the program in its final months, or should they stay to the very end? In this regard, there needs to be some creative thinking on the part of KTI about how to ensure that highly trained staff are considered for other USAID/Kenya positions, and also given incentives, including performance bonuses or a severance gratuity, to stay on the job until the very end of the period. We noted that KTI is making efforts to retain staff within the USAID/Kenya mission by providing them with details of possible job opportunities and encouraging them to apply. This was appreciated by the staff who intend to remain on the job. Second, some concern was raised about salaries which some staff felt were not competitive or which varied greatly amongst staff with the same job descriptions and duties. KTI searches for highly motivated staff who are less interested in monetary gain and more interested in the nature of the work. Nevertheless, for morale, adequate salaries and adherence to the principle of equal pay for equal work are important. We were not in a position to determine how serious the issue of unequal salaries really was, but it was raised in a number of interviews with KTI staff. Third, we also observed that all the higher level managers (finance, operations, and regional program managers) are ex-patriots. While some had very considerable Kenyan experience, so too had most Kenyan staff. It was not clear if this is a deliberate policy or simply happened, but the message sent to several bright young Kenyans working for KTI was problematic. This 38 apparent mobility ceiling for Kenyan staff bothered a number who found this discouraging. They perceived that higher level positions appeared to be reserved for non-Kenyans, a questionable practice given the duties Kenyans perform at the USAID mission in Nairobi, and for other implementing partners of USAID in Kenya.39 4.11 The Relationship Between KTI and Its Grantees This evaluation has concluded that the KTI program is field driven at the level of implementation. Data from the survey supports this conclusion. As noted above, 70.3 percent of the respondents indicated that the activities they were implementing with funding from KTI reflected what their organization originally wanted to do when they sought the grant from KTI. Only 5.2 percent indicated that the activities only reflected what KTI wanted to do (Annex III, Table 21). The absence of KTI imposition during implementation is perhaps reinforced by the fact that the KTI field officers have been easy and flexible in their relationship with the grantees. The PDOs were said to be “flexible or easy” to work with by 82.5 percent of the grantees (Annex III, Table 18a), while the Grant Managers were said be the same by an equally impressive 75.7 percent (Annex III, Table 19a). However, the process of generating the proposal and taking it through the various stages of review suggests that there is substantial input from the center. Though proposals come from the field through the PDOs, KTI management—the CR, DCR, CoP and the PDOs themselves—all have the opportunity to adjust, rework, and “panel beat” proposals into something quite different, or at least not identical to, the original proposal in order to get the proposal to advance KTI’s objectives and approach. This then raises the question of who between KTI and the grantees is actually driving the process? KTI’s response to this issue was that KTI should rightly drive the relationship because the grantees are in essence implementing KTI’s objectives as contracted parties. A proposal will, therefore be changed or panel beaten only when its objectives are fuzzy or the scope is too wide or too limited, among others. Proposals will also be panel beaten to be consistent with U.S. policy objectives or internal USAID and OTI rules. In such cases a potential grantee is free to renegotiate the proposal or turn down the funding altogether. Consistent with the findings from our survey, only a small number of grantees have turned down funding. The overwhelming majority— more than 90 per cent—did not. It should also be stressed that what emerges from the extensive review process are grants that support a complementarity of interests and objectives—not only those of KTI, but also of the grantees. Indeed, as noted above in Section 4.2 nearly all grantees said they would work with KTI again. 4.12 Short vs. Long term Support for Democratic Transitions The Kenya Transition Initiative is fundamentally about supporting a democratic transition. However, a profound tension exists between KTI’s rapid action/short term approach and the need for sustained engagement in supporting a transition to democratic constitutional rule. 39 For example, the Chief of Party of the implementing partner for USAID’s program to strengthen the Kenya National Assembly has been a Kenyan since 2005. 39 Transitions tend to be protracted long term exercises that necessarily play out over multiple election cycles, and often suffer reversals or stasis (Carothers 2002). Further, even when institutions are established to enhance democratic politics, profound challenges persist in breathing life into these new institutions. Prominent Kenyan constitutional scholar Yash Ghai notes that creating a culture of constitutionalism where the intent, values and principles of the constitution animate society is a much more difficult and complex challenge (Ghai 2009). KTI was aware of this and funded many activities, organizations and networks that appeared to be building and spreading awareness of the constitutional reforms, their intent and purpose. Yet precisely because of the long timeline for transitions, KTI was also aware that these organizations and networks needed to endure and adapt past KTI program lifespan. Overall, building “capacity” was central to KTI’s efforts yet OTI’s internal cultural and rhetoric minimize the importance of this aspect of KTI’s work and contribution. KTI openly talked about influencing culture, a more difficult and complex task. This is why KTI supported visual arts, books, radio talk shows and theater that opened up questioning of elements of authoritarianism and their “banality” in Kenyan society. At another level, KTI with its “venture capital approach” modeled itself on less hierarchical and open technology innovation hubs and companies in search of good ideas from anyone, anywhere. As a result of this and also the style of the leadership, KTI itself appeared to operate in an open manner that supported internal criticism and discussion. This created a supportive environment for young professionals that encouraged their voice and intellect. Support of young Kenyan professionals in this way, thus, also inadvertently became part of the overall KTI agenda even if not explicitly articulated or theorized in this way. 4.13 The Risks of Engaging Democratic Transitions Another aspect of democratic transitions is that they invariably involve conflict and confrontation, because reformers must challenge concentrations of unaccountable power. Even as KTI tries to be low key and strengthen open modes of communication and dialogue to support reform, some actions and actors supported by KTI may be perceived as direct threats to powerful politicians. This is particularly true of the work around hate speech or the International Criminal Court. Indeed, as noted at the end of Section 4.3, a number of KTI grantees and employees felt physically threatened at different stages in the program, and we were cautioned by some Kenyan political analysts that the potential threat to grantees could extend beyond the lifetime of the KTI program. In addition, some KTI field staff deliberately used the program more to put themselves in the public eye for their own political advantage, eventually winning elections. While, at one level, this is a sign that citizens can respond positively to young politicians with a reform message, the downside reality is that these KTI associates are linked to powerful politicians within their political parties. At least one senior political analyst cautioned that these KTI protégées could feel pressure to reveal information about some of the more risky aspects of the KTI program and the grantees engaged in them. It is a potential risk. 40 4.14 Promoting Innovation One positive impact of KTI is its role as an incubator of innovation and ideas. KTI hired and gave considerable autonomy to a cadre of young, highly-trained, sophisticated professionals who appear to relish the role of developing new ideas and working with the grantees whether as a PDO, grants manager, or procurement specialist. The PDOs in particular have to defend ideas in the rolling assessments and in conversations with KTI senior management, which gives them opportunity to hone their analytical and persuasive capabilities. We also noted cases where particularly talented grantees are hired into KTI, and at least one case of mobility within KTI when a motivated and talented young grants assistant was promoted to grants manager. Yet another was a personable, energetic and savvy PDO who became part of the young African leaders initiative, and participated in President Obama’s teleconference with such leaders during his recent trip to Africa in June 2013. Another key feature of KTI’s incubator role was the large numbers of talented women in positions of responsibility. Overall, by utilizing and fostering the talents and skills of these young professionals KTI is helping to nurture Kenya’s human and social capital as discussed in Section 4.16 below. Furthermore, by positioning them as connectors within networks of grassroots reformers and activists across the country, KTI is having an unintended but nevertheless positive impact on Kenya’s civil society where many of these young professionals are likely to work at the end of the program. 4.15 Legacy of Grants and Grantees Overall, it is clear that a wide variety of grantees have gained capacity in terms of skills, networks, and physical infrastructure to act, innovate, and create impact, though the exact extent of these impacts are not directly measurable. The larger scale grantees such as large NGOs like the National Council of Churches (NCCK) and government entities such as parliament or the CIC gained critical support at critical times, which clearly enabled them to act more quickly and efficiently in implementing and pushing for reforms. For example, the head of the CIC Charles Nyachae called the assistance received for setting up the office “vital” since the Kenyan government procurement rules would have delayed their operation significantly. More typical grantees, smaller grassroots groups and movements, in part through their constant interactions with KTI staff at a local level, developed or honed the capacity to write proposals, network, organize ideas, and realize strategies. While the extensive back and forth with PDOs and grant managers could at times be frustrating and time consuming, overall many grantees found this an excellent and important learning experience. In addition, many of the grantees were continuing with projects and activities, either through another KTI grant, another granting agency, or their own resources. In a number of cases the physical infrastructure was critical to allowing grantees to engage in activities well past the actual sponsored activity. For example, the Hosana Empowerment group of actors and writers based in Kericho, which performed a number of plays at markets around the theme of “Nobody is Above the Law” received critical equipment (portable stage, wireless microphone, PA system) that allows them to continue to make plays and reach out to audiences. In Pumwani in Nairobi, we found that the Kamakunji Community Based 41 Organization Network (K-CBO Net) was still using tents, chairs, and computers donated by KTI. They were unclear whether they could rent out the chairs and tents in order to support their organization’s finances, which would have added to K-CBO Net’s sustainability through an improved financial situation.40 Overall, this infrastructure (and the training to use it) is another KTI legacy with an impact beyond the activities it financed. 4.16 Social Capital and its Significance for Peace and Democratization Perhaps most significant, and in conclusion, KTI has contributed to the formation of vital social capital in Kenya at several levels and in areas of the country where social capital has been extremely limited in supply compared to Nairobi and the former Central Province. As demonstrated in landmark studies of Italy and the United States by Robert D. Putnam (Putnam 1993, Putnam 2000), and by Alexis de Tocqueville more than a century before (Tocqueville 1969), social capital in the form of civil society organizations and the networks of between them, is essential for the establishment of democratic political systems, and the absence of violent conflict on which they depend. CSOs foster the cooperation between individuals for some common purpose, build trust among members, and in the process lay the groundwork for peace. Though it is an abstract phenomenon whose presence is difficult to measure, its formation is one key indicator of KTI’s success. KTI has contributed to the formation of social capital in Kenya by incubating current and future civil society leadership, creating capacity in grantee organizations and possibly through networks created by and amongst the grantees themselves. It has done this at two levels and possibly a third: First, by creating a cadre of highly skilled staff to implement KTI’s programs in the field. And second, by enhancing the skills and experience of the individual grantees whose programs it has financed. To the extent that grantees, both as individuals and their organizations, have come together in networks of cooperation, both formal and informal, to reinforce each other’s efforts, the KTI has also fostered the formation and spread of social capital at a third level. With respect to KTI staff, the evaluation team was highly impressed by the skills we observed how well they understood the nature of OTI/KTI and its mandate, why they are working there, and how well they carried out their duties. The PDOs, grant managers and procurement officers were all highly skilled professionals, highly motivated, analytical in approach and highly sensitive to the political nature of their working environment. The PDOs were all university graduates who fully understood that the KTI offered no long term employment, but who were and remain still highly committed to their jobs because they want to “make a difference” by advancing political reform. They are “highly wired” into the local political environment and have the skills to quickly see the conflict implications of minute local political changes. KTI’s grant managers were also highly motivated, and possessed skills that are in high demand, especially in the area of accounting and financial management. They will have no problem securing employment elsewhere after KTI closes down. Together with the PDOs, their skills can make or break the implementation of the program. They fully 40 This has since been clarified for them by KTI. 42 understand their roles and the political context within which KTI operates. KTI’s procurement officers are similarly highly motivated and fully understand their importance in an organization that provides support mostly in-kind. All three categories personnel evolved into a formidable asset for both KTI and Kenya. The CR, DCR and CoP of the implementing partner are to be commended for recruiting such a talented staff to work on the program. Thus, as KTI winds down in Kenya and its Kenyan staff move on to new positions, this asset is one that will continue to impact on the country in the years to come. Some will move to different positions at USAID, some may implement future programs of other donors, or Kenyan NGOs, and some may move into the private sector or government. Indeed, one PDO was elected to the new Senate in the 2013 elections. As KTI exits Kenya, USAID/Kenya and/or OTI/W should make a concerted effort to keep track of the whereabouts of these key staff. The second level of social capital formation has occurred among the grantees themselves. During the course of its inquiry, the evaluation team sought to determine the ways the grantees have benefitted from their association with the KTI program. They indicated that, first, they learned the need for organizational financial discipline, and are now utilizing the procedures demanded by the KTI when seeking support from other financiers. Second, they have learned the art of preparing grant proposals from both the PDOs and the grant managers. Third, they have learned how to contextualize their thinking in the larger questions facing the country rather than thinking solely about the discreet activities in which they are engaged. Fourth, they have learned to seek linkages between what their organizations are doing and political conditions in Kenya. Fifth, they have learned to maintain lines of communication with organizations that might no longer be funding them, including KTI. Sixth, they are now aware that a “non-traditional” donor organization like KTI exists. Finally the grantees, like the KTI staff, have developed immense analytical and predictive skills. Thus, whereas nearly everyone was predicting chaos during the March 2013 elections, the grantees proved to be good readers of Kenyan politics. Seventy percent of the respondents to our phone survey correctly predicted that there would be no violence during the election (Annex III, Table 26a). Directly related to the second level of social capital is the organizational capacity that KTI helped create in the CSO grantees. KTI does not see this as core to its mandate though it is arguably one of the most important legacies of the program. Rather, KTI has sought to work with organizations that have some capacity and credibility to achieve its objectives. This is especially true for those grantees that KTI approached for funding as contrasted with those that initiated contact themselves. A major assumption of KTI operations has been that most organizations do not have all the needed capacity even though some might have more capacity than others. On this basis KTI has been more than willing to engage in organizational capacity building so long as the capacity is required to implement a KTI activity. Indeed, our survey confirmed that grantees recognize and appreciate KTI’s role in organizational capacity development. A very large majority (91.8 percent) indicated that the support received from KTI has “strengthened their organization” in terms of building its capacity to carry out its activities (Annex III, Table 23a). 43 V. DILEMMAS AND LESSONS LEARNED The findings reported in the previous section led the evaluation team to identify a series of dilemmas and lessons learned for OTI as it reviews its operations around the world. We divide these into dilemmas posed by the nature of OTI’s mandate and modus operandi, that is to say, the logical inconsistencies and operational issues resulting from OTI’s programs in the field, and lessons learned from a review of one such program. While we cannot say with certainty that these dilemmas and lessons will be present in all countries where OTI has mounted programs, we are confident that they are not limited to the Kenyan case. We also recommend in Section VI that the exercise of this evaluation be replicated in at least two or three other countries to test for the presence of these and other dilemmas and lessons. We begin first with a discussion of the dilemmas, because there is no single policy or operational response to the existence of these conundrums, and then turn to a summary of the lessons learned. 5.1 Measuring and Attributing Impact One of the biggest dilemmas for the KTI program and OTI generally is whether it is possible to systematically and quantitatively measure the overall impact of the sum of very different short term interventions. Over the short lifespan of an individual intervention which usually unfolds over a period of one or two months, though in a few cases more, some impact was observed but only that which could be described, not measured, at the anecdotal level. Even then the impact was “still unfolding.” Taken together over the period of the program, the interventions might also yield individual stable impacts that are clearly observable, but because they are quite different in form and content, their cumulative impact would be difficult to measure. In the meantime, other actors including other donors introduce similar interventions and this complicates the problem. To which provider of each basket of interventions does one attribute the impact(s) observed? This then raises a basic question: Should KTI be seeking to measure and attribute impact given the nature and the duration of its interventions and the presence of other actors in the arena? This dilemma is best illustrated in trying to assess the impact of the KTI program on preventing violence during the March 2013 elections. There are at least seven overlapping explanations, possibly more, of why peace prevailed during the run-up to the elections, the elections themselves, and after of which the interventions by KTI are only one. (i) The “never again” psyche. According to this explanation, Kenyans and particularly key elites were mortally scared by the PEV of 2008 and had reached a consensus that it should never be repeated. The election contest was therefore framed—by elites, by the media and civil society—in a manner meant to remind each other and the Kenya public that violence could happen again if the political rhetoric, including hate speech, was allowed to get out of hand as happened in 2007/2008. 44 (ii) The Kenyan cases before the International Criminal Court (ICC) The indictment of six prominent cases by the ICC and the pending trials of three, contributed to the absence of violence in that all candidates and those who ran their campaigns were acutely aware that the ICC and the international community generally was watching. The indicted contestants from the Jubilee coalition did not want to make matters worse for themselves while CORD did not want to do anything that might attract the attention of the ICC. (iii) The Jubilee Alliance. The Kikuyu/Kalenjin alliance under the Jubilee banner was primarily an alliance of the those who perpetrated violence on each other in 2007/2008, and in the run-up to the elections in 1992 and 1997. As such the alliance made it possible to set aside the issues underlying the 2007/2008 violence even though the issues had not been resolved. If the 2007 protagonists had been on different sides in 2013, some form of violence would almost certainly have occurred. (iv) Role of civil society and media. Civil society, religious organizations and the media invested heavily in peace crusades and messages. This was particularly noticeable during the campaign and even more so during the highly tense vote-counting period when the country held its breathe and waited. One wrong message and violence would have broken out. That’s how close it was. (v) Donor’s role. All donors can claim some credit for also investing heavily in the campaign for peaceful elections. In this effort, USAID’s programs, including KTI, were among the most prominent, but still two among many. Though one can conceive of a research design and statistical methods such as regression analysis to attribute the presence of peace to individual donors, results would be muddled at best.41 (vi) The perceived effectiveness of key institutions established by the 2010 constitution. Kenya’s new constitution passed overwhelmingly in a referendum held in August 2010, created or commenced reforms of several key institutions which arguably contributed to the maintenance of peace during the elections. The new Supreme Court and the vetting of all judges in the existing judiciary led to the acceptance by the losing presidential candidate, Raila Odinga, of the Court’s decision to uphold the results announced by the Independent Elections and Boundaries Commission (IEBC). Acceptance of the outcome of the presidential election by Odinga’s supporters and by other prominent political leaders of his CORD coalition may also have been the result of Kenya’s new system of devolved government as the coalition won control of 24 of the 47 new county governments. Put differently, the creation of the new tier of devolved government achieved its purpose by dividing power so that the losers at the national level were not completely shut out of the political process as in the past. 41 For one example, one could attempt to “explain” the presence of peace both generally, and/or by specific geographic regions by using some common measure of donor contributions, most likely the total amount of money each spent on the elections. As previously noted, however, such a common measure that lumps different types of interventions together is unlikely to yield clear results. 45 (vii) Security forces’ preparedness. In marked contrast to 2007/2008, Kenya’s security forces were highly prepared for containing violence in 2013. Approximately 90,000 police were trained and deployed before the elections, more than twice the number deployed in 2007. The forces were better coordinated and pre-deployed to the most likely “hotspots.” The police also made clear that any protests would not be tolerated, and that the perpetrators of violence would be dealt with harshly (i.e. that they would be shot). While the possibility (indeed probability) of intimidation by the police cannot be ruled out, the reality is that it contributed to the absence of violence.42 These explanations are interlinked, and difficult to untangle. Together, they help explain why the elections were largely peaceful. KTI was a direct contributor to explanations (iv) and (v), but was not the only contributor. U.S. and other diplomatic efforts encouraged (i) and (ii). The dilemma is how to isolate KTI’s role and contribution given the presence of other players and their contributions. For this reason, the evaluation team cannot determine with precision the contribution of OTI/KTI in preventing violence in the 2013 elections much less in mitigating non-election conflict or preventing future political conflict. Measuring impact becomes even more problematic with regard to measurable impact on the prospects for long term reforms in Kenya. However, the evaluation team can state with a high degree of confidence that after making nearly 800 small grants, the program has contributed to Kenya’s transition to democracy in at least two ways. First, the activities funded by the grants were generally well conceived in terms of their purpose and target audience. The rate of failure, 5.8 percent, was also very low due in large part to the iterative process of developing and reviewing proposals for grants by KTI staff. Second, and perhaps more significant over the long term, the program contributed to the pool of social capital required for democracy at two levels as discussed in at length in Section IV.16. 5.2 The Dilemma of OTI’s Short term MO vs. the Long term Time Frames for Transitions to Democracy KTI came into Kenya after the 2007-2008 PEV which had pushed Kenya’s process of democratization into reverse. In this context, OTI’s approach was designed to preserve the possibility of a democratic transition rather than support the transition itself. However, soon after the country stabilized, KTI moved to address the structural conditions that entrench authoritarianism and impunity that in turn enable mass violence. These include control of media discourse by powerful politicians, weak civil society, especially in the rural areas, and a culture of impunity made possible by weak democratic institutions. KTI thus shifted its focus from one transition that was short term (violence to peace) and consistent with OTI’s modus operandi (MO) to another that was long term, but also consistent with OTI objectives—laying the groundwork for a democratic and constitutionally ruled society. The latter by its nature is 42 As Peter Anyang Nyongo, the Secretary-General of ODM told one member of the evaluation team, “The police presence was so great that it was impossible to protest the decision by the Supreme Court” that Uhuru Kenyatta had won the election. 46 fraught with uncertainty, reversals and possible stasis (Carothers 2002). 43 Moreover, supporting a democratic transition is an inherently long term exercise that clashes with KTI’s short term mandate. Yet supporting democratic transitions is the logical extension of KTI’s, and OTI’s organizational purpose of reestablishing political stability. Indeed, long term peace and the absence of violence are most likely to be guaranteed by democracy, because where democracy functions well, potential parties to conflict are accommodated by the political system and its configuration of institutions. Put differently, in conflict prone societies such as Kenya, which are marked by deep fault lines between the peoples of different ethnic and/or religious groups, democracy is the best potential mechanism to mediate conflict and achieve peace. OTI is inevitably faced by this dilemma: its terms of reference and mandate is to address specific short term issues and needs, but in doing so OTI is drawn into addressing the long term to sustain and consolidate the impact of their work. 5.3 The Dilemma and Myth that OTI “Does Not Do” Sustainable Development Closely related to the dilemma of short vs. long term interventions is the dilemma that OTI “does not do” sustainable development when in fact it does. Because OTI is meant to provide rapid and short term response to crises in post-conflict societies, senior personnel at both KTI and OTI state that the goal of “sustainable development” is not one with which they are primarily concerned. Perhaps for this reason, the SOW for this evaluation did not instruct the evaluation team to consider KTI’s interventions from the perspective of whether and how they contributed to sustainable development in Kenya (i.e. sustainable peace and security, democratization, etc.). Yet the question is inevitably raised of what happens to the successful efforts and grantee organizations supported by OTI programs? Are they (i) simply to be assisted and then orphaned with the view that the truly worthy and fittest will survive and move on to the next stage of their respective development? Or (ii) should there be some follow-on program of support that sustains the development of the most worthy efforts over a much longer period? (iii) Moreover, if OTI is frequently drawn into long term efforts to support democratization to consolidate the gains of its interventions, must it at least give some thought to the sustainability of its work? Indeed, OTI’s own internal assessment of the lessons learned from the KTI program state that the reality that OTI does not do sustainable development is a myth (USAID/DCHA/OTI 2013b) The dilemma and myth that OTI does not do sustainable development raises other questions that OTI must consider if not always address: If the purpose of OTI/KTI-like programs is to stabilize post-conflict situations and lay the ground for democratic governance, at what point must these programs or a successor program shift from stabilization to the building of key 43 Mass violence in Kenya has been cyclical peaking around election periods and depending on the nature of the ethnic coalitions. Thus, peace in one election cycle does not ensure peace the next time round unless there is a transformation of politics. Thus, KTI’s work on a transition to peace may not be done after one election. 47 institutions of countervailing power that are the basis of democratic governance—e.g. of rural civil society organizations in areas of a country where there are few or none, and of other key political institutions including the legislature and judiciary? As already noted, stabilization is a short term effort while institution building and democratization is a long term project. How best should OTI/KTI shift from the first to the second? The issue of sustainability was also raised by KTI grantees and took the following form: “We, the grantees, like what we are doing, we have developed a high appetite for the kind of activities we are implementing, we have acquired very useful political and other skills, but what will happen after KTI leaves? What other programs within USAID or other aid organizations follow the same or similar approach?” From a programming perspective, the issue for both KTI and the USAID mission is how and who can manage so many small grants with the same kind of personnel intensive methods and skills that KTI has developed. We suspect that this dilemma is not limited to KTI, but arises again and again with OTI operations worldwide. 5.4 The Dilemma and Myth that OTI Does Not “Build Capacity” Parallel to the myth that OTI does not do sustainable development is the myth that OTI does not and should not “build capacity,” that is to say, the technical competence of individuals and the organizations for which they work to perform critical tasks. Here again, the reluctance to engage in capacity building or more accurately, the reluctance to acknowledge that many of its grants contribute to capacity building, flows from the rhetoric that OTI limits its interventions to short term rapid responses to some crisis. But as discussed in section 4.16 (page 41), one of the most significant achievements and legacies of KTI, and the product of similar OTI programs of small grants to support emergent civil society organizations in other countries, is the development of social and human capital at three levels: (1) KTI’s own field staff, (2) the individual grantees and their organizations, (3) the informal networks amongst grantees and individuals. The development of social capital is a key ingredient of building capacity, particularly organizational capacity, and thus critical for the development of civil society and other organizations and institutions including government. As discussed above, the importance of building “social capital” for peace, democratization and the establishment of strong institutions of governance, etc. should not be underestimated. So too the building of capacity among such institutions. Indeed, 91.8 percent of the respondents in the phone survey reported that the capacity of their organization had been strengthened by the grant from KTI (Annex III, Table 23a. 5.5 The Challenge of “Scaling Up”: OTI’s/KTI’s Relationship to the USAID Mission’s Democracy, Human Rights and Governance Program Closely related to the issue of sustainability is the challenge of “scaling up.” Some of KTI’s most impressive interventions, for example, the court records management interventions in Eldoret, Kapsabet, and Mombasa and the land registries in Kilifi, Kitale, Nakuru, Kajiado, and Thika, appear to be experiments, albeit successful ones. At what point do these experiments 48 transition from being isolated pilot projects to mainstream programs? Is OTI/KTI going to be part of this transition and if so how? The danger here is that OTI/KTI could find itself engaging in many experiments of a “pilot” nature which are not initially designed as pilots with the long term objective of scaling up? Clearly OTI/KTI is very good in piloting rapid responses but is OTI/KTI also conscious of the pilot nature of many of its interventions? The answer to this question has implications for OTI/KTI’s choice of implementing partners. If an intervention is consciously conceived as a pilot, then the OTI/KTI implementing partner should ideally be in a position to scale up the pilot if and when it is deemed successful or to drop it if and when it is deemed unsuccessful. The issue of scaling up again begs the question of KTI’s, and by extension, OTI’s relationship to the USAID mission in Kenya and the other countries in which OTI works, especially mission programs to support democracy, human rights and governance (DRG). How should the proverbial “baton” be passed from the OTI country program to the mission? Ideally, traditional USAID programs build upon and begin where KTI/OTI programs end. Yet this does not seem to be how things actually work for a number of reasons including parallel programs and few coordination mechanisms between OTI and USAID, as well as the inflexible and bureaucratic operating mode of USAID proper. Should there be a higher level of “integration” between the two, and if so what does that mean—that OTI would from the outset work more closely with the mission’s DRG program? At what point and how does KTI shift from being a venture capital organization that supports CSO start-ups and other worthy opportunities to a more conventional program of democracy assistance? 5.6 OTI’s Dilemma of Being an “Agency Within an Agency” and Challenge of Maintaining Intra and Inter Agency Harmony With two exceptions, KTI was universally viewed by those interviewed by the evaluation team at USAID/Kenya and at the US Embassy in Nairobi as “very helpful” because of its ability to do what traditional aid mechanisms and offices within USAID cannot to do; namely to move rapidly and opportunistically in a crisis or into uncharted territory. Indeed, the need by USAID for OTI’s “guerrilla” like approach is the very reason why this “agency within the parent agency” was established two decades ago. This need and OTI’s capacity for rapid response is currently illustrated in Kenya by KTI’s bridging role to AHADI, USAID’s pending multi-year program to support political devolution by assisting the establishment of county level government. The implementation of devolved government in Kenya is a major component of implementing the 2010 Constitution, but remains a highly sensitive political issue because the recently elected central government of President Kenyatta and Deputy President Ruto is not enthusiastic about sharing power fully with the 47 new county level governments. These new county level governments need various types of assistance if they are to get off to a promising start, and survive, and several donor agencies including USAID have programs pending. USAID must therefore carefully lay the groundwork for whatever assistance it ultimately provides whether it be to the governors and executive departments of the new county governments, to the new county assemblies, to local civil society or to all three. Donor 49 agencies must also be mindful of the political dynamics within each county to which it is considering assistance and to the political dynamics between those counties and the central government. A full discussion of these considerations is provided in Annex IV of this evaluation, but suffice to say, these political considerations are fluid and complex, and understanding them requires personnel with experience in Kenya, not new hands.44 Moreover, because AHADI will be implemented by a contractor which will not be selected and establish its office in Kenya before the end of 2013, the KTI program was extended through the end of year to serve a need that was not or could not be met by the USAID mission. As the “bridge” to AHADI, KTI and its knowledgeable staff will help the mission identify the most promising opportunities for AHADI which the new program can pursue once it is up and running in 2014. Indeed, the evaluation team concluded that USAID/Kenya’s decision to extend KTI for this purpose was a smart use of KTI’s resources that also recognized the strengths of the program. The involvement of KTI in AHADI nonetheless also illustrates the tensions that sometime arise between OTI, the USAID mission and the US Embassy and the need for better integration, coordination, or simply communication between the three. As noted above, staff at USAID/Kenya and the embassy regard KTI as a “very helpful” and necessary resource. Indeed, as the chief of mission himself observed, “if KTI (and by implication OTI) did not exist, we would have to invent it.” A similar view was echoed by the director of the USAID mission. That said, there are periodic tensions resulting from three factors: First, the contrasting programs and approach of KTI on the one hand and the DRG office of USAID/Kenya on the other. Second, the contrasting personalities, respective expertise and leadership styles of those directing the two programs. Given its mission and organizational approach, it is not surprising that KTI is led by a CR and DCR who are highly entrepreneurial in outlook and politically savvy with a strong and detailed knowledge of Kenyan politics. Indeed, as discussed above, the characteristics of the CR and DCR as well as that of subordinate staff are precisely what KTI (and OTI programs generally) require, and why KTI’s record of performance has been high. On the USAID/Kenya/DRG side, the effort is led by a skilled manager who spends most of his time making the wheels of the bureaucracy turn rather than spending time in the field. This too is a necessary skill given the many moving parts and budget of the DRG program in Kenya. By contrast, the CR and DCR view themselves as a pair of freelance guerilla warriors. The programs, approach and value added of KTI and USAID/Kenya/DRG and those who lead them are complementary, but the tensions between them are also clear. Will these tensions in style and approach result in a creative portfolio of interventions in support of Kenya’s transition to democracy? Or will these tensions result in turf wars and/or coordination issues between the two programs? We see no reason why it cannot be the former. 44 The evaluation team felt strongly, that given the political minefields that USAID is likely to encounter and KTI’s role, that we would include a separate annex on devolution in this report. Though not required by the SOW, the members of the evaluation team think it important to share our thoughts on this final area of the KTI program in Kenya. 50 Consistent with the OTI ethos, the CR and DCR define their work in terms of identifying and taking quick advantage of opportunities that come before them. Consulting with or even informing the head of the DRG program or USAID mission director is not the first matter that crosses their minds. By contrast, the head of the DRG program (and his predecessor) is frustrated by the fact that he does not always know what OTI/KTI is doing in the field even though he appreciates the need for nimbleness and the outcomes of their work to date. He would be happier if there were greater consultation, coordination and “integration” between the two programs. A third dimension of the relationship is the independence of OTI/KTI as “an agency within an agency.” In addition to having a different purpose and approach, OTI/KTI formally reports to OTI Washington and not to the director of USAID/Kenya. In addition, the fact that both OTI/KTI and DRG programs mount interventions that are explicitly political in nature, means that neither can operate independently of the US Embassy, particularly the ambassador, the DCM and the political office. Thus here too, the evaluation team identified a tension. OTI’s mandate and approach is to work quickly and independently, but the need to maintain working “triangular relationship” between the OTI and its counterparts at the DRG office and embassy is essential. Maintaining such a triangular relationship depends more on informal rather formal communication. What is required is not the “integration” of the country OTI program into the DRG program at USAID, but rather that the CR and DCR be diligent in keeping the head of the DRG program and the ambassador “in the loop.” The OTI CR must be a good “schmoozer” rather than a loner to make the triangular relationship work. So too the head of the DRG program and the ambassador or someone he or she designates to follow the work of OTI.45 The challenge of keeping all parties “in the loop,” is compounded when a fourth actor, the Department of State’s Bureau of Conflict Stabilization Operations (CSO) arrives in country to mount its own initiatives. CSO is not involved in the current effort to scale up USG assistance for devolution, but did send a team to Kenya from August, 2012 through March, 2013 to address and mitigate the potential for electoral violence that was also a core focus of OTI/KTI. CSO also works in other conflict prone countries where OTI is present, and addresses similar issues of post-conflict stabilization and mitigation. If the Kenya experience is any guide, the level of consultation and coordination of interventions between OTI and CSO is sometimes not what it should be. CSO initially ignored KTI even though the latter began its work in Kenya four years prior to CSO’s arrival in the country, and had deployed a more experienced and knowledgeable team in the field. CSO eventually drew on KTI’s expertise and the relationship between the two improved. Given that CSO’s field staff reports to Washington and to the U.S. ambassador it is sometimes necessary for the latter to step in to resolve the problem of “too many cooks” spoiling the soup—which he did. This is also a generic issue that OTI might want to discuss directly with the Bureau in Washington. 45 The leader of the evaluation team has observed the relationships between USAID missions and US embassies for more than twenty years. USAID’s DG programs have been most successful where the USAID mission director and the head of the USAID/DG program have a close working relationship with the ambassador. In countries where OTI operates, the CR must join that conversation. It is the informal rather than the formal relationships that are key, but the prospects for such informal conversations are a function of the personalities and temperaments of the individuals concerned. 51 5.7 Getting the Personnel Right A major conclusion of this evaluation is that the effectiveness of the KTI program is to a very large extent dependent on “getting the personnel right” from top management, that is to say the CR, DCR and CoP down to the field operatives, particularly the PDOs and grant managers and procurement officers who are all Kenyan citizens and who are often residents or originally from the geographic areas where KTI works. The OTI/KTI approach is personnel intensive. Its success or failure turns on its ability to recruit the best individuals available. While recruitment is relatively easy to address for top management because one is looking for just a few individuals, it is more difficult when recruiting top level PDOs, grant managers, and procurement officers. With one or two exceptions, KTI via close consultations between the CR and DCR with the CoP of the implementing partner, got it right. The evaluation team was highly impressed by the skill level, knowledge of local political conditions, and high level of motivation expressed by the field staff interviewed—particularly the Kenyan nationals employed by the implementing partner. Almost all were just plain smart yet individuals with good skills at interpersonal relations—“the best and the brightest.” Notwithstanding the high level of personnel KTI recruited for its programs, a number of questions came up: What methods, if any, does KTI have to nurture the required qualities among the existing staff and to ensure that future staff have the skills and the right mentality? What type of a selection process do the CR, DCR and CoP employ when recruiting new staff? What is the nature of personnel reviews in terms of frequency and the criteria used? And more generally, and most important what guidelines, if any, does OTI provide to insure that the high quality of personnel recruitment by KTI is replicated elsewhere? OTI needs to address these questions. 5.8 The Grant Identification Process is Critical for Program Success The grant identification process is critical for effectiveness and impact of the program. The grant identification process is also important for creating program coherence and for ensuring that the activities reflect the program strategy. The collective and iterative approach adopted by the rolling assessment is one of the best practices of the KTI program and one that in the judgment of the evaluation team greatly reduces the risk of failure. The evaluation team was not in a position to determine whether the procedures used by KTI to identify potential grantees and develop grant proposals were those employed by OTI generally. If not, there is much to be learned from the KTI case. That said, these procedures are personnel intensive and dependent on a high level of judgment and knowledge of local conditions on the part of the staff of the implementing partner. As discussed in the previous section, this puts a premium on staff recruitment as the procedures followed are only as good as those implementing them. The recruitment of first-rate staff, however, may not always assure programmatic success. Selecting funding activities and targeting stabilization needs are closely related, yet the evaluation team noted that the criteria for both were not entirely clear. There are no formal guidelines or weighting system for 52 scoring the criteria that staff might use to make funding decisions. Moreover, such a system can only work if the criteria themselves are constant and well understood recognizing that KTI field staff must be allowed some discretion to identify the most promising grantees.. 5.9 OTI’s Objectives vs. Grantees’ Objectives Notwithstanding the benefits accrued from the iterative review process of grant proposals, the agendas and objectives of grantees beyond those that address KTI objectives cannot be ignored. KTI views grantees as “contractors” who are supported in order for KTI to achieve its objectives. The grantees’ other objectives are not their main concern. That said, KTI’s generally believes in the grantees it supports by “buying into” the grantee’s organization, and investing time and money in the group. KTI also needs to consider grantees’ broader objectives and agenda, because they directly or indirectly affect the implementation of the KTI supported activity. Put simply, grantees will be less inclined to implement grants that address KTI’s agenda if it does not provide support for their own. 5.10 The Choice of the Implementing Partner Makes a Difference The approach and impact of KTI-1 and KTI-2 (through the end of 2012) were quite different with clear tradeoffs between them. The challenge is how to achieve the best of both and whether that is in fact possible or realistic. The differential experience also raises the question of whether a short term five year program should be rebid in mid-stream, a process that seems wasteful to the evaluators because of the high transaction costs imposed. 5.11 The Limitations of “In-Kind” Grants Although the use of “in-kind” grants can be extraordinarily useful when assisting start up CSOs in the rural areas, exclusive reliance on such grants introduce certain rigidities and costs—especially the cost of required personnel on the part of OTI and the IP—that the programs initiated by KTI cannot be assumed by USAID/Kenya after KTI shuts down. Exclusive use of in-kind grants also stifles the development of financial management capacity on the part of grantees. Use of in-kind grants can cause delays in grant implementation if GMs and the IP generally are working on a high volume of grants. The mixture of cash and in-kind disbursement can produce effective implementation and create more connectedness between the center and the field among others. The in-kind model therefore requires flexible application and careful evaluation of each proposed grant in order to determine the optimal mix of cash and in-kind disbursement. That said, 75.2 percent of the respondents in the phone survey agreed with the proposition that “in-kind support made it easier to implement (the) project, while 75.7 percent said that their grants manager was “very flexible” in implementing the grant (Annex III, Table 19a and 20b). 53 5.12 The Dilemma of Addressing Human Rights Violations vs. the Need to Protect Grantees Combating Such Violations KTI and OTI generally pride themselves on acting rapidly, creatively, and opportunistically to exploit possibilities for reform that arise serendipitously over the course of events in the field. This is a real strength of OTI programs and their approach. However, in certain contexts, this rapid and opportunistic approach needs to be balanced by more measured, cautious mechanisms to assess and manage risk including review of certain categories of activities by legal and human rights experts. As discussed in Section 4.3 above, this is particularly true when support for human rights is at stake. By the very nature of OTI’s work, grantees and staff engage in activities that may, by challenging existing power hierarchies, be particularly high risk. Institutionalized mechanisms and procedures to address risk are therefore critical even for activities that are considered low risk but by a turn of events can end up to be high risk in unforeseen ways. Risk cannot be dealt with in an ad hoc way. In the KTI program, we came across two situations where we thought institutional mechanisms were needed. One involved a legal case around hate speech which may have benefitted from more rigorous legal review; hate speech cases in Kenya must follow careful procedures and culprits sometimes bring these cases to court to get them thrown out on procedural grounds. These “frivolous cases” make it appear to the public that the matter was solved or no evidence existed for the claims of hate speech when in fact that is not true. Another set of cases involved grantees engaged in projects that may be construed to have a link to the ICC. This is an ongoing concern as the ICC cases unfold and some grantees have been threatened as a result of their work supported by KTI. Here too, KTI and OTI generally must balance its efforts to halt or minimize human rights abuses with its need to protect grantees combating such violations. 5.13 Devising an Appropriate Exit Strategy If OTI’s modus operandi is to mount a large number of short term initiatives that nonetheless draw it into supporting democratic transitions that are long term (section 5.2), and if it is a myth that OTI can avoid issues of sustainable development (section 5.3), then the Office must give more thought to devising an appropriate exit strategy when it winds down its programs. As noted above, how does OTI “pass the baton” from its operations to those of the USAID mission? As the experience of KTI makes clear, the program cultivated considerable human capacity even if this was not its initial and primary goal. Within the networks between grantees, staff and participants in its activities a great deal of social capital and detailed local knowledge has accumulated about selected counties outside Nairobi. If the investment in these human resources, knowledge and networks are not utilized and leveraged by USAID\Kenya in the complex road ahead, it would be an enormous waste. This is particularly true as USAID engages in support for devolution, which is fraught with many potential pitfalls including complex local politics. Indeed, as discussed above, USAID/Kenya has found KTI extremely helpful in bridging the gap between KTI and the new AHADI program precisely because of 54 KTI’s flexibility, rapid reaction and its local networks and local knowledge. At a substantive level too, many of the issues around violence and authoritarianism, and the related challenges of implementing the 2010 Constitution—resource-sharing, ethnic intolerance and minority representation in politics—will play themselves out, possibly with even greater intensity, at the county level. That KTI will conclude its operations in Kenya by laying the groundwork for AHADI is a sensible way to wind down its operations, but whether the utilization of KTI in this manner was part of a conscious exit strategy as distinct from a serendipitous utilization its resources to address an obvious need was unclear to the evaluation team. An appropriate exit strategy would also include some plan for the continuing employment of KTI personnel, particularly those Kenyan nationals employed by the implementing partner. Although the prospects are good that the implementing partner ultimately selected for AHADI will hire some if not all KTI staff, there are no guarantees. In the meantime, KTI staff, both Kenyan and non-Kenyan are looking for opportunities elsewhere with the result that some may not serve until the end of the program. Indeed, a potential problem facing KTI as it nears the end of its term is that will lose key employees, a situation that could be highly disruptive to the conclusion of its operations. The evaluation team noted that as of the time of its second field investigations in June, 2013 that no policies were yet in place to retain key staff until the very end although a system of performance bonuses or gratuities would probably minimize the risk of early departures. The evaluation team further observed that a significant number of KTI’s Kenyan staff should be groomed for management roles when the implementing partners for AHADI and other USAID/DRG mission programs are selected. Finally, while it may be impossible for OTI to develop a clear exit strategy at the beginning, or even during the middle of a country program, such a strategy should be developed no later than a year before shutdown. . 5.14 The Dilemma of Evaluation: To Evaluate or Not Evaluate? A widespread belief within USAID, is that OTI has long resisted evaluations of its country programs on the grounds that it is impossible to evaluate a fast moving and ever changing target. Because most OTI programs consist of a large number of small and short term projects supported by in-kind grants, it is impossible to conduct a rigorous “before” and “after” measure of these programs accomplish. Moreover, because the focus and content of interventions is constantly in flux, any conclusions drawn from an initial baseline evaluation are unlikely to serve as good points for comparison with the findings from subsequent rounds of evaluation because the interventions considered in the later rounds may be very different in purpose and type than those that came before them. Put differently, one cannot compare apples and oranges so why try. The challenges faced by the evaluation team in measuring and attributing impact, also suggest a major lesson for OTI generally; namely that while the impact of its programs cannot be isolated and measured quantitatively, these programs can nonetheless be assessed far more thoroughly and at regular intervals than in the past to determine what works what does not, and why. Indeed, we hold up this evaluation report as an example. Moreover, by conducting such assessments at regular intervals, significant lessons are 55 learned. As noted previously, this evaluation is the first two part evaluation of an OTI country program by an independent team of evaluators engaged from outside the Office, and thus the first assessment of how an OTI program evolved over time. The two part approach also enabled the evaluation team to determine how KTI responded to the observations and recommendations contained in the first interim report. Finally, the exercise clarified what an evaluation of an OTI country program can and cannot accomplish within the parameters of OTI’s approach, and given the diverse types of interventions OTI programs often support. Notwithstanding these benefits, it must be remembered that the present evaluation did not meet strict methodological requirements to measure the “before” and “after” impact of the program’s interventions. Given the generic issues raised in this evaluation, the present exercise suggests that similar evaluations might be undertaken of OTI programs in other countries. Such replications would facilitate a comparative understanding of the strengths and weaknesses of OTI operations generally including the overall impact of those programs and lessons learned. For these reasons the present evaluation also points to several improvements that could and should be made in future evaluations of OTI country programs. These recommendations are discussed below in Section VI. 5.15 The Dilemma of Attribution: To Brand or Not to Brand? The final dilemma facing OTI is whether to “brand” or “not brand” the interventions that comprise its country programs. As noted above, KTI usually maintained a low profile in Kenya, particularly in conflict prone areas where any overt role on the part of an international donor agency to claim credit for the assistance it provides risks calling excessive attention to its work—and, in KTI’s case, to its many grantees. That said, the need for OTI to “tell its story”—within USAID and perhaps to skeptical members of Congress, pulls the Office in the opposite direction. In this regard, the evaluation team was struck by the many efforts to put a good “spin” on the narrative of what the grants in kind program had accomplished. For example, the recent KTI Legacy Report (USAID/DCHA/2013a) and other efforts at telling the KTI story were conducted mainly for promotional purposes rather than for analysis of the programs cited. This perceived need to “brand” KTI interventions appears to be driven mainly by requests by the parent Office to “tell the OTI story”, but such publicity is really for intra-USAID and domestic U.S. consumption than for citizens of the host country most of whom know that USAID has provided needed assistance. The perceived need to brand is analogous to the need to attribute positive outcomes specifically to inputs made by OTI when, as discussed above in Section 5.1, it is often impossible to measure such results. 56 VI. RECOMMENDATIONS As discussed in the Introduction, the primary audience for Part 2 of this evaluation, and thus for its recommendations is OTI and USAID generally rather than KTI. Because KTI is now is in its wind-up phase, the dilemmas and lessons learned discussed in the previous section and the recommendations they suggest are particularly relevant for OTI as it goes forward with programs similar to KTI in other countries. To the extent that KTI is “typical” of OTI operations generally, the evaluation team recommends that the lessons learned from the Kenyan case be considered carefully in Washington, perhaps in a group forum such as a staff seminar or staff retreat. We also suggest that OTI circulate this report, particularly Sections V and VI to its CRs, DCRs and implementing partners currently operating in other countries to determine the extent to which the findings presented in this evaluation are valid elsewhere— which they may not be. With this important caveat in mind: 1. OTI should continue its basic approach of mounting rapid, short term and opportunistic interventions to mitigate and prevent violence in post-conflict societies. OTI country programs should also continue to lay the institutional groundwork for transitions to democracy over the long term. The ability of OTI to support the development of civil society organizations in selected rural areas beyond the capital city is particularly noteworthy, as is its preferred mechanism of providing assistance through small “in kind” grants. Stated simply, OTI is capable of doing what its parent agency, USAID, cannot do—move quickly into situations where the window of opportunity is short and any delay in providing support degrades its value. For this reason, OTI is usually valued by the U.S. ambassador and by the USAID mission director as a unique and effective tool at their disposal. 2. Notwithstanding the inherent value of its approach, OTI needs to strengthen and “beef up” its procedures of evaluation. The evaluation team was surprised that the two part exercise it was asked to perform was a “first” for OTI when it should have been the norm. Regular and periodic evaluation by independent and external experts will not only maintain the high standards and creativity the team observed during its field investigations in Kenya, it will also shield the office from the perception that OTI seeks to avoid being evaluated, and, perhaps, justify the resources USAID allocates to the Office. Moreover, for the very reasons that OTI operations in the field do not easily lend themselves to quantitative measures of success or failure, it is important that the Office mount as rigorous a form of qualitative evaluation as possible—to both “tell the OTI story” in a credible manner, and to assess with greater precision the strengths and weaknesses and needed areas for improvement in its ongoing country programs.46 Thus, for example, it is important that every OTI country program engage a small team of two or three external evaluators to produce a clear written baseline assessment of conditions in the country at the start of the program including an initial judgment whether and how the program has commenced its operations. Part 1 of the evaluation of KTI did not commence until the fourth year of the 46 To be fully credible elsewhere within in USAID, and outside the Agency, it is important that evaluations be conducted by external evaluators. Though sometimes useful, narratives by OTI staff such as the recent “Legacy Report” (USAID/DCHA/OTI. 2013a)are likely to read and be perceived—as they were by the evaluation team—as self-serving efforts at branding and public relations. 57 program which is usually too late to improve operations. It could not provide a baseline assessment, and its value for the CR, DCR and implementing partner to make the recommended adjustments was limited, though the evaluation team noted that several of its recommendations were in fact carried out. 47 Without a baseline analysis, it was also impossible for the evaluation team to assess the true impact of KTI operations on a “before” verses “after” basis. As most OTI country programs commence for a period of two or three years, an evaluation of the program, similar to the mid-term evaluation of KTI should be conducted no more than two years after operations start. And because OTI country programs are usually extended and average over four years in length, a final evaluation similar to the present exercise be undertaken during the final six months of operations. Such evaluations should be conducted at roughly the same time as the Program Performance Reviews (PPRs) undertaken by OTI staff, and their findings compared with the findings and recommendations derived from those exercises. Perhaps most important, the evaluation team recommends that the director of OTI and his/her senior staff read, compare and discuss these evaluations to better appreciate the general lessons learned and recommendations they produce. Because conditions vary from one country to the next, some findings will be country specific while others generic, but the latter cannot be fully identified in a single country evaluation such as this study. We therefore also recommend that if a more rigorous series of “baseline,” “mid-term” and “final” evaluations of each country program cannot be mounted in the near term, that the next best course would be to replicate the two part approach followed for this exercise in two or three other countries.48 Only through such a comparative exercise can the broader lessons be learned, and OTI procedures and operations be adjusted accordingly. Finally, the evaluation team recommends that the findings from this proposed exercise of comparative evaluations be systematically shared with all of OTI’s implementing partners given the role they play in mounting OTI programs in the field. Indeed, any failure to share findings would undermine a major reason for strengthening OTI’s evaluation regime. 3. OTI needs to develop or better articulate its guidelines for its recruitment of key personnel as well as its guidelines for reviewing proposals for its grants. Because OTI’s approach is personnel intensive, the success or failure of its country programs turns on a combination of the quality and skills of key personnel, and on the procedures it follows when determining which interventions to support. Recruitment of key personnel, both senior management (the CR, DCR and CoP) and field staff (especially PDOs and grants 47 The evaluation team recognizes that its call for multi-part evaluations to better assess the impact of OTI programs is partially met by the current practice of the annual Program Performance Review (PPR) conducted by most OTI country programs. 48 Though it may seem obvious and self-serving, it is important that all evaluation teams be comprised of acknowledged experts on the country hosting the OTI program under review as well as individuals knowledgeable about USAID and donor policy generally in the areas of conflict mitigation, democracy and governance. All teams proposed for evaluations should be thoroughly vetted by OTI before the team is finalized for the exercise. 58 managers) needs to be undertaken against a clear set of criteria as does the process by which such personnel review and select grantees for funding. These criteria, however, do not appear to exist except in general terms. To maintain the high quality of programming of the type observed by the evaluation team in Kenya, OTI needs to be more specific and transparent in articulating what its selection criteria are to prospective grantees. 4. OTI needs to rethink the time-frames of its programs and/or to develop more flexible OTI type mechanisms within the main USAID system to address the continuous need for rapid action in countries in crises in combination with interventions that lay the groundwork for democratic rule. This recommendation stems from the observation that democratic transitions are long term processes where crises and violence are possible at any time and institutional support needs continuous and creative programming. The evaluation team noted that the KTI program went well beyond its intended two year lifespan, but that this was necessary and typical of OTI programs. Indeed, as noted above, U.S. ambassadors and USAID mission directors like to have access to OTI as part of their “toolkit.” OTI should therefore develop mechanisms or partnerships with USAID to both maintain its capacity to provide rapid short term support complemented by the capacity to support the process democratization which is an inherently long term process. 5. The issue of grantee sustainability must be addressed. The largest portion of OTI country programs often consists of support for new and weak civil society organizations. As the viability of these organizations is often critical for both conflict mitigation and long term progress towards democratization, the issue of what happens to these organizations after OTI completes its program needs to be addressed. To simply state that “OTI does not do sustainable development” and abandon what OTI has started is both a myth, and wasteful. OTI together with the Office of Democracy Human Rights and Governance at USAID in Washington, should therefore think through how, what is nurtured by the former is sustained by the latter, without mortgaging the opportunities for the latter to pursue its own agenda. 6. Consider Establishing an OTI Alumni Forum or Association. Having established a nimble organization to implement KTI which in turn made nearly 800 grants to nurture the development of civil society organizations in regions where civil society has been historically weak, OTI should develop an appropriate mechanism for keeping track of and engaging this valuable resource. As discussed in Section IV, KTI made a major contribution to the development of social capital in Kenya at two levels—its own staff of Kenyan nationals, and the leaders of the organizations that received KTI grants. The individuals involved constitute KTI’s greatest legacy in Kenya, and it would short sighted to for KTI to merely say “goodbye” at the conclusion of its program. Instead, and because OTI develops similar reservoirs of human capital in the other countries in which it works, OTI Washington should consider how it might establish an alumni forum or association to maintain contact with its current and former employees and grantees worldwide. Such an association could be similar to the association of former Fulbright Fellows, or the register of former grantees maintained by the National Endowment for Democracy. It would require some effort and modest funding by OTI Washington complemented by an effort in the field—by the USAID mission or U.S. embassy to periodically update the list of such 59 “alumni,” and bring together in an appropriate forum. Such an association could also play a “brokerage” role for USAID, other donor agencies, and other American NGOs such as PACT, to tap into what OTI has left behind. An alumni forum or similar mechanism could both sustain and multiply the fruits of the original country program. 7. Shielding grantees against physical harm. Because grantees and participants in OTI projects can face real personal risks in the course of their work, OTI should develop a clear human rights protection policy to ensure that any grantees or partners that may be put at physical risk because of OTI work will have clear procedures, personnel and networks that can be quickly activated in the case of need. This requires collaboration with the USAID mission and U.S. embassy which may need to deal with such cases after OTI has completed its program and departed the country where the problem has occurred. The team further recommends that OTI Washington engage an appropriate number of specialists knowledgeable about witness protection to advise CRs, DCRs and CoPs in this area, because the need by country teams for support in this area is likely to occur periodically in post-conflict societies where OTI concentrates its work. 8. OTI country programs should work with the USAID mission to develop a clear exit strategy at least a year before shut down to ensure that the rich local knowledge and networks built by OTI with considerable effort and expense are transferred to the mission to the extent possible, and that the “handoff” from OTI to the USAID mission is smooth. A sound exit strategy would include the scaling up the most successful and innovative projects pursed by OTI that fit into USAID/DRG’s priorities. It would also address the need for a continued focus on peace-building and conflict mitigation as this need is unlikely to terminate in most post-conflict societies with OTI’s departure. The exit plan should also include a plan for the monitoring of post-OTI human rights protection issues resulting from OTI activities. 9. Keeping all relevant USG mission officials “in the loop.” OTI CRs and DCRs need to establish closer working relationships with key personnel at the USAID mission, particularly the mission director and the director of the mission’s office of democracy, human rights and governance (DRG). Given the inherently political quality of OTI programs, CRs and DCRs must also maintain an effective liaison with the U.S. ambassador or his/her designate in the embassy’s political office. OTI also needs to develop an effective working relationship with the Department of State’s Bureau of Conflict Stabilization Operations (CSO) in those countries where the Bureau is also mounting programs of its own. OTI Washington should seek to establish a clear division of labor with CSO or procedures for determining such, and vice versa as there is a potential for overlapping initiatives, mixed signals to key host country actors, and missteps in the field. The problem is not of OTI’s making, but rather by the establishment of CSO to address issues previously addressed by OTI alone. The problem is compounded by the fact that CSO has comparatively few resources than OTI, and that its interventions and deployment of personnel are short lived. 10. As a general policy, OTI should strive to ensure that local host country experts are found at all levels of management in its programs. As discussed in several sections of this report, 60 the evaluation team was highly impressed by the skill level and motivation by Kenyan staff employed as PDOs and grants managers. However, the team also noted that no Kenyans were employed in positions higher than that of the PDO. This reality was raised by several PDOs and grants managers during the course of their meetings with the evaluation team. 11. Create a small advisory board of host country nationals for each OTI country program to serve as a sounding board for the CR, DCR and CoP. The role of this board would be to provide timely advice and guidance on politically sensitive projects including the provision of support for cases involving the violation of human rights. The board should include individuals respected for their impartiality and knowledge of the country’s politics as well as a legal expert and specialist in human rights protection. The establishment of such a board will reduce the downside risks of poorly thought out rapid response interventions.49 The advisory board would not review all interventions mounted by a typical OTI country program. Rather it would be used by senior management in those instances where the appropriateness of a particular intervention is in doubt or where problems have arisen from a recent intervention. 12. Improve the Database. The KTI database is a unique and highly valuable resource for the management of a program consisting of a large number of small interventions such as OTI’s. However, it is not set up to facilitate easy quantitative analysis of much of its data, because the data is often entered in a narrative rather than numeric form. Thought should be given to the development of a standard template for the databases for all OTI country programs that allow the user to consider different aspects of all grants and grantees on a variable by variable basis including the ability to quickly compute summary statistics (means, percentages, correlation coeficcients, etc.). Data of a qualitative or ordinal nature should be assigned codes so that grants with different characteristics (e.g. large grants vs. small; grants where the grantee is female rather than male) can be compared with respect to their level of effectiveness, the purpose of the grant, etc. etc.). Use of one of the standard social science software packages for statistical analysis such as SPSS, would make the development of an improved data base both inexpensive and easy. 49 For example, in Kenya, it would have been useful for KTI to have had access to such advice when mounting its rapid responses to hate speech. 61 REFERENCES Barkan, Joel D. 2013. “Kenya’s 2013 Elections: Technology is Not Democracy: Lessons from Kenya,” Journal of Democracy, vol. 24, no. 3 (July), 156-65. __________________, Jacqueline M. Klopp and Njuguna Ng’ethe. 2013. Kenya Transition Initiative Final Evaluation: Part 1, Mid-Evaluation Report, June 2008 through June, 2012. Washington: The QED Group LLC Barney, Joseph. 2009. “Kenyan corruption is brought to book.” Financial Times. http://www.ft.com/intl/cms/s/0/ef83ec12-754c-11de-9ed5- 00144feabdc0.html#axzz29g5cGHtF Carothers, Thomas. 2002. “The End of the Transition Paradigm” Journal of Democracy 13.1: 5-21 Ghai, Yash. 2009. “Decreeing and establishing a constitutional order: challenges facing Kenya” Paper prepared for the Kofi Annan Foundation. http://kofiannanfoundation.org/sites/default/files/Microsoft%20Word%20- %20Paper%20by%20constitutional%20expert%20Professor%20Yash%20Ghai.pdf Klopp, Jacqueline M., Patrick Githinji, and Keffa Karuoya. 2010. “Internal Displacement and Local Peacebuilding in Kenya: Challenges and Innovations.” United States Institute of Peace. http://www.usip.org/publications/internal-displacement-and-local-peacebuilding￾in-kenya Lawson, Marian. 2009. “USAID’s Office of Transitions Initiative After 15 Years: Issues for Congress.” Congressional Research Service. http://www.fas.org/sgp/crs/row/R40600.pdf Long, James D, Karuti Kanyinga, Karen Feere and Clark Gibson. 2013. “Kenya’s 2013 Elections: Choosing Peace over Democracy.” Journal of Democracy. 24, 3 (July), vol. 24, no. 3, 140-55. Obama, Barack. 2013 “Remarks by President Barack Obama at the University of Cape Town.” Washington: The White House, June 30, 2013. Putnam, Robert D. 2000. Bowling Alone: The Collapse and Revival of American Community. New York Simon and Schuster ______________. 1993. Making Democracy Work: Civic Traditions in Modern Italy. Princeton: Princeton University Press. Tocqueville, Alexis de. 1969. Democracy in America. Translated by George Lawreence. Edited by J.P. Mayer. New York: Anchor Books 62 USAID/DCHA/OTI. 2013a. KTI Legacy Report. (Currently available only on a flash memory obtainable from OTI. A website link is planned). July, 2013. USAID/DCHA/OTI. 2013b. Lessons Learned from the Kenya Transition Initiative (Written by Michael Fritz, Gretchen Murphy and Leah Werchick). February, 2013. USAID/DCHA/OTI. 2009a. “Office of Transition Initiatives (OTI): First Year Accomplishments.” Internal presentation to the US Mission. USAID/OTI/Kenya. 2009b “Program Performance Review.” August 27, 2009 USAID/DCHA/OTI. 2008. “Kenya Country Assessment.” Unpublished manuscript. 63 ANNEX I PEOPLE INTERVIEWED FOR EVALUATION Individuals interviewed during Phase I of field investigations (September, 2012) General (28) Michael Chege Policy Advisor, Ministry of Planning Mary Gachocho UN Habitat, Development Partners on Land Secretariat Albert Gachuka Chair, Kikuyu Elders Peace Initiative, Eldoret8 Nancy Gitau Director of Political Affairs, Office of the President and political advisor to Uhuru Kenyatta John Githongo CEO, Inuka Trust, Nairobi Karuti Kanyinga Associate Professor, Institute for Development Studies, University of Nairobi and Advisor to Kofi Annan Hussein Khalid Executive Director, Muslim Human Rights Institute (MUHURI), Mombasa Laura Walker Hudson CEO, Frontline SMS, Nairobi Maina Kiai Director, InformAction and former Chair, Kenya Human Rights Commission Cornelius Korir Bishop, Catholic Diocese for Eldoret Salim Lone Journalist and advisor to Prime Minister, Raila Odinga Keffa Magenyi Program Coordinator, IDPAC (Internal Displacement Policy and Advocacy Centre), Nakuru Kenneth Marende Speaker, National Assembly Kennedy Massime Executive Director, Centre for Democracy and Development and convener of ELOG (Electoral Observation Group) S.E. Migot-Adholla Board Member, Equity Bank and former Permanent Secretary, Ministry of Agriculture Harris Mule Consultant to the World Bank and Former Permanent Secretary, Ministry of Planning Willy Mutunga Chief Justice, Supreme Court of Kenya Ibrahim Mwathane Chairman, Land Development and Governance Institute David Ndii Independent consultant on governance and political economy Joel Ngugi Judge, High Court of Kenya and Head, Judiciary Transformation Secretariat Pheroze Nowrojee Prominent attorney and civil society activist Peter Anyang Nyongo Minister of Health Services Mary O’Hagan Country Representative, National Democratic Institute (NDI) John Okumu Professor Emeritus of Political Science and former Director, Center for Refugee Studies, Moi University John Sei Former chair, Kalenjin Council of Elders, Eldoret Cyprian Selebalo Land Advisor, UN Habitat, Development Partners on Land Secretariat S.O. Warfa Provincial Commissioner, Rift Valley Province, Nakuru 64 Michael Yard Chief of Party, International Foundation for Electoral Systems and advisor to the Independent Boundaries and Electoral Commission US Department of State/USAID/KTI Staff (12) Johnnie Carson Assistant Secretary of State for African Affairs Megan German USAID/OTI, Washington Robert Godec Charge d’Affaires and head of U.S. Ambassador to Kenya James Hope Acting Mission Director, USAID/Kenya Galeeb Kachra Deputy Director, USAID/OTI/KTI John Langois Country Representative, USAID/OTI/KTI Catie Lott Former Democracy and Governance Advisor, USAID/Kenya Mark Meassick Economic Growth Director, USAID/Kenya Patricia Ochieng Project Development Specialist, USAID and adviser to former Ambassador, Michael Rannenberger John Schroeder Lead Foreign Affairs Officer, Bureau of Conflict and Stabilization Operations, Department of State John Smith-Sreen Director, Office of Democracy and Governance, USAID/Kenya Chris Wendell Bureau of Conflict Stabilization Operations, Department of State and Chair, Interagency Taskforce on the 2013 Elections. Implementing Partners (DAI and Chemonics) Staff (16) Masoud Muhammed Ali Grants Manager, Chemonics, Mombasa John Bett Project Development Officer, Chemonics, Kericho Holly Flood Chief of Party, Chemonics Inc. Mary Kagunyi Project Development Officer, Chemonics, Nairobi Sam Kona Former Project Development Officer, DAI, Nairobi Selenne Korir Project Development Administrator, DAI, Eldoret Mark Lawler Regional Office Director, Chemonics, Eldoret Husna A. Mbarak Project Development Officer, Chemonics, Mombasa Margaret Muyanga Project Development Officer, Chemonics, Mombasa Samuel Muthui Senior Procurement Manager, Chemonics, Nairobi Zainob Ronoh Grants Manager, Chemonics, Kericho Stephen K. Sang Project Development Officer, Chemonics, Eldoret Mercy Shahale Project Development Officer, Chemonics, Eldoret Barbara Wasilwa Grants Assistant, Chemonics, Kericho Carole Watheka Senior Grants Manager, Chemonics, Nairobi Geofrey Waweru Procurement Specialist, Chemonics, Nairobi Grantees (Grant Recipients) (43) Mahmoud Abdulahi Secretary General, Council of Imams and Preachers, Mombasa 65 Dolphina Alego Majestrate, County Law Courts, Eldoret C.A. Anyanzwa Program Coordinator, Kamukunji Based Community Network, Nairobi Abdul Aziz Coordinator, Kericho Interfaith Network, Kerichio Patricia Bourne Inspector of Police, Peace Building Secretariat, Provincial Commissioner’s Office, Nakuru Bill Cheruiyot Script writer, Hosana Empowerment Group, Kericho Mary Chepkony Chair, Rural Women’s Peace Link, Eldoret Abdul Hamid Finance Manager, Kenjya Muslim Youth Alliance John Kebaso National Organizer, Community Communication, Sotik Stephen Kipligat Chauir, Hatua Moja, Eldoret Abdulahi Kiptanui Chair, Kericho Interfaith Network, Kericho; Chair Kipsigis Consultative Group, Kericho, and former town councilor, Kericho Municipality, 1969-2007 Mercy Kinua Coordinator, Peace Building Secretariat, District Commissioner’s Office, Nakuru Sally Kirui Project Coordinator, Sotik-Borabu Women’s Peace Drive, Sotik Denis Kogo Chair, Youth Revival Initiative, Eldoret Simon Korir Field Coordinator, Sotik-Borabu Women’s Peace Drive, Sotik Daisy Kosgei National Steering Peace Steering Committee, Eldoret Julius Lamaon CEO, KASS Media, Nairobi Kenneth Langat South Rift Valley Coordinator, House of Law and Justice, Kericho Daniel Limo Rural Initiative for Sustainable Development, Eldoret Fred Matiangi Former Chief of Party, SUNY/Kenya Charles Mbogo Chief Magistrate, County Law Courts, Eldoret Wanjiko Mbugua Chair, Governance Forum Kathurima Mburugu IT specialist, County Law Courts, Eldoret Philip Meli Head of Commercial, KASS Media Institute, Nairobi Kenneth Marende Speaker, National Assembly, Nairobi Athman A. Mohammed Project Officer, Council of Imams and preachers, Mombasa Lydia Mukama Chair, Constitution Observation Group Godfrey Musila African Center for International Legal Policy Samuel Musumba Administrative Officer, Provincial Commissioner’s Office, Nakuru Hassan Ole Naado CEO, Kenya Muslim Youth Alliance Stephen Ng’etich Project Coordinator, Yurith Water Resources Initiative, Kericho Thomas Ng’etich Executive Director, Good Samaratans International, Eldoret Ezikiel Njenga Project Coordinator, Kamukunji Based Community Network, Nairobi Patrick Njuguna Executive Director, Kenya Youth Parliament Charles Nyachae Chair, Commission for the Implementation of the Constitution (CIC) Ken Okello IT specialist, County Law Courts, Eldoret Charles Omanga Secretary General, Philip Ronoh Coordinator and actor, Hosana Empowerment Group, Kericho 66 Mwai Sirasi Paralegel, Kamukumji Community Based Community Network, Nairobi Mohamed Swazuri Professor of Geography, Mombasa Polytechnic College and Chair Land Commission of Kenya Samuel Tanui Secretary, Kericho Interfaith Network, Kericho Naima Wamaitha Treasurer, Kamukunji Community Based Community Network, Nairobi Fred Yego Director Wareng Youth Initiative for Peace and Development, Eldoret Individuals interviewed during Phase II of field investigations (June, 2013) General (26) Francis Ang’ila Aywa Chief of Party, State University of New York (SUNY) Parliamentary Strengthening Program Vincent Bartoo Bureau Chief, Standard Group, Eldoret Michael Chege Consultant and Former Policy Advisor, Ministry of Planning John Githongo CEO, Inuka Trust, Nairobi Rama Gullam Finance Manager, Muslim Human Rights Institute (MUHURI), Mombasa Karuti Kanyinga Associate Professor, Institute for Development Studies, University of Nairobi and Advisor to Kofi Annan Maina Kiai Director, InformAction and former Chair, Kenya Human Rights Commission Kithinji Kiragu CPA, Price Waterhouse Coopers and gubernatorial candidate, Embu Cornelius Korir Bishop, Catholic Diocese for Eldoret Salim Lone Journalist and advisor to Prime Minister, Raila Odinga S.E. Migot-Adholla Board Member, Equity Bank and former Permanent Secretary, Ministry of Agriculture Fred Matiangi Cabinet Secretary for Communications and Information Technology Philip Meli Chief of Staff, Office of the Governor of Usasin Gishu County Mwakera Mwajefa Reporter (stringer), Nation Media Group, Mombasa Ibrahim Mwathane Chairman, Land Development and Governance Institute David Ndii Independent consultant on governance and political economy Sharllen Njoroge Program Manager, Muslim Human Rights Institute (MUHURI), Mombasa Peter Anyang Nyongo Senator for Kisumu County Mary O’Hagan Country Representative, National Democratic Institute (NDI) Martin Rirei Community Development Officer, ACK church, Eldoret Heidy Rombouts Project Leader, Promotion of Social Justice, Reconciliation and National Cohesion, GIZ (German Aid) Stephen Sang Senator for Nandi County and former KTI Project Development Officer in Eldoret Peter Shambi Program Manager, Muslim Human Rights Institute (MUHURI), Mombasa James Smart Journalist and Presenter, Nation Television Network (NTN) 67 Michael Yard Chief of Party, International Foundation for Electoral Systems and advisor to the Independent Boundaries and Electoral Commission Tom Wolf Iysos Synovate Public Opinion Research US Department of State/USAID/KTI Staff (7) Karen Freeman Mission Director, USAID/Kenya Megan German USAID/OTI, Washington Robert Godec U.S. Ambassador to Kenya Galeeb Kachra Deputy Director, USAID/OTI/KTI John Langois Country Representative, USAID/OTI/KTI Patricia Ochieng Development Program Specialist, USAID/Kenya John Smith-Sreen Director, Office of Democracy and Governance, USAID/Kenya Implementing Partners (DAI and Chemonics) Staff (14) Masoud Muhammed Ali Grants Manager, Chemonics, Mombasa John Bett Project Development Officer, Chemonics, Eldoret Mary Kagunyi Project Development Officer, Chemonics, Nairobi Sam Kona Former Project Development Officer, DAI, Nairobi Elly Kurgut Project Development Officer, Chemonics, Eldoret Mark Lawler Regional Office Director, Chemonics, Eldoret Noel Martinez Deputy Chief of Party,Chemonics, Nairobi Husna A. Mbarak Project Development Officer, Chemonics, Mombasa Margaret Muyanga Project Development Officer, Chemonics, Mombasa Suki Nyadawa Grants Manager, Chemonics, Eldoret Kennedy Onyango Grants Assistant, Chemonics, Eldoret Daire O’Reilly Regional Office Director, Chemonics, Mombasa Ryan Smith Chief of Party, Chemonics, Nairobi Zainob Ronoh Grants Manager, Chemonics, Mombasa KTI Kenyan Grantees (Grant Recipients) (2) Nuh Nassir Abdi Speaker of County Assembly, Tana River County and Chair of County Speakers’ Forum Stephen Riechi Chief Magistrate, High Court of Mombasa 68 ANNEX II KTI GRANT PROJECTS REVIEWED, GRANT STATISTICS *Grants suggested for inclusion by CR or DCR **Grants included in random sample and also suggested for inclusion by CR or DCR Eldoret DAIELD005* Rural Womens Peace Link (Burnt Forest) DAIELD021** Wareng Youth Initiative for Peace and Development DAIELD036 Good Samaritans International (Engaging Local Politics) DAIELD037* Eldoret Magistrate Court (computerization of case records) DAIELD043 Emgwon Youth for Change (Agenda 4 café) DAIELD050* Good Samaritans International (programs to counter hate speech on KASS FM) CHXELD005 Good Samaritans International CHXELD012 Rift Valley Development Trust CHXEKD019 Regional Committee, Eastern Rift Valley (International Peace Day) CH0XEKD22 Hatua Moja Youth Group, Eldoret Kericho DAIKIS019 Kipsigis Consultative Group DAIKIS036 Community Communitarian, Sotik DAIKIS055 Kericho Interfaith Network CHXKER003 Kericho Interfaith Network CHXKER007 House of Law and Justice CH0XKER10 Borabu Sotik Womens Group CHXKER013 Youth Water Resources Users Kisumu DAIKIS002 Diakonia Compassionate Ministry DAIKIS008* Civil Societies Organization Network DAIKIS013 Betta Theatrix DAIKIS018 Kotina Bunge Group (youth) DAIKIS023 Women in Fishing Industry Program (business/development) DAIKIS029 Kisumu Cultural Forum (peace and reconciliation) DAIKIS035 Timely Limited Consulting Firm (prevent land grabbing) DAIKIS047* Nyanza Youth Coalition 69 Mombasa CHXMSA001 Pwami Leadership Council CHXMSA003 Coast Education Center CHXMSA013 Kituo Mombasa CHXMSA018 STTA Research Land in Kwale CHXMSA025 Council of Imams and Preachers in Kenya Nairobi DAIHRG002 Human Rights grantee (anonymous) DAINBO012 District Peace Building Committee, Nakuru DAINBO053 Provincial Peace Forum, Nakuru DAINBO068 Kamukunji Based Community Network DAINBO108 Kenya Muslim Youth Alliance DAINBO121 Governance Forum CHXNBO005 Youth Platform for Change CHXNBO023 Constitution Observation Group CHXNBO042 African Center for International Legal Policy Summary Statistics for All Grants by Field Office (through October 16, 2013)* DAI CHEMONICS CHEMONICS (5/2008-5/2011) (5/2011-8/2012) (9/2012-10/1613) Eldoret 73 46 46 Kericho 38 14 Kisumu 59 Nairobi 155 69 64 Nairobi Eastleigh 46 38 Mombasa 31 66 Total 287 230 228 Total $14,191,934 $7,314,830 $10,328,577 Average amount of grants (in US dollars): Eldoret $41,684 $31,232 $38,207 Kericho 24,697 24,710 Kisumu 36,851 Nairobi 57,902 34,200 55,860 Nairobi Eastleigh 31,232 46,715 Mombasa 36,878 43,559 All $49,449 $31,804 $45,301 70 *Data is for all grants that were closed, and does not include an additional 23 grants made by DAI and 23 made by Chemonics which were canceled. Total number of grants (May, 2008 through October 16, 2013), including grants that were canceled: 791 Number of grants canceled under DAI: 23 Number of grants canceled under Chemonics: 23 Cancelation rate** under DAI: 7.4% Cancelation rate under Chemonics. 4.8% Cancelation rate for all grants under both implementing partners: 6.2% **The cancelation rate is the number of grants canceled as a percentage of all grants attempted Type of Recipient Organization (May, 2008 through October 16, 2013) NUMBER OF GRANTS VALUE IN US DOLLARS Community Organizations: 50 8.8% 1,766,647 7.1% Local Non-profit organizations: 311 54.7 11,605,668 46.8 Local for Profit Organizations: 39 6.9 1,755,114 7.1 Local Faith Based Organizations: 39 6.9 1,763,882 7.1 Government of Kenya 96 16.9 6,447,289 26.0 International Organizations 3 .5 199,328 .8 Direct Procurement 31 5.4 1,235,126 5.0 N = 569 100.0 $24,773,054 99.9 Sector of Grant (May, 2008 through October 16, 2013) Conflict mitigation 132 16.6% Youth engagement 157 19.7 Media 33 4.1 Public institution strengthening 149 18.7 Evaluation and assessment 56 7.0 Civic action (e.g. support to CSOs) 90 11.3 Alternative leadership 30 3.7 Advocacy (e.g. support to CSOs 81 10.2 Accountability 44 5.5 Access to Justice 25 3.1 N = (797) 100.0 71 ANNEX III FINDINGS FROM SURVEY OF GRANTEES The survey of KTI grantees was conducted during the last two weeks of February, 2013 by enumerators hired and trained for the exercise. The enumerators attempted to contact a universe of 309 potential respondents, the total number of individuals heading organizations that had received grants from KTI through January, 2013. The questionnaire for the survey was developed by Joel Barkan with assistance from Njuguna Ng’ethe. Ng’ethe also trained the enumerators who conducted the survey and oversaw the initial preparation of the data for analysis via the Statistical Package for the Social Sciences (SPSS). Barkan, with the assistance from Jamy Felton at the University of Cape Town prepared the final SPSS file from which the tables presented in this annex were generated. Although KTI had made over 500 grants by the time of the survey, the number of potential respondents was much lower because more than one third of all grantees reported that their organizations had received two or more grants. The proportions of grantees receiving one or more grants is given in Table 2 below. One hundred seventy-eight (178) individuals were ultimately interviewed for the survey. The enumerators were required to make three attempts to interview all potential respondents. Some refused to be interviewed but most of the 131 (of the 309) individuals not interviewed were dropped because their mobile phone numbers had changed since their organization received its grant from KTI, and could not be contacted in time for the survey. The tables below present summary “one way” frequencies for all questions included in the survey, and are presented here without comment. Selected findings from the tables are reported in the main text, mostly in Section IV of the main report. TABLE 1: GRANTEE’S ORGANIZATION (I.E. RECIPIENT ORGANIZATION) A central government ministry or department 10.6% The National Assembly, Judiciary or government commission 29.8 A county level government 1.1 A large and/or established civil society organization (CSO) 10.6 A new and/or small CSO 39.4 A private company or business 7.4 Other: 1.1 N (number responding) = (94) 72 TABLE 2: NUMBER OF GRANTS THE GRANTEE’S ORGANIZATION RECEIVED FROM KTI 1 65.2% 2 20.8 3 8.4 4 or more 5.6 N = (178) TABLE 3: YEAR GRANTEE’S ORGANIZATION RECEIVED ITS FIRST GRANT Survey 2008 12.4 2009 16.3 2010 21.4 2011 9.0 2012 32.7 2013 7.9 N = (177) TABLE 4: IMPLEMENTING PARTNER FOR GRANTEE’S GRANT Development Alternatives Inc (DAI) only (before May 2011) 46.9% Chemonics only (after May 2011) 48.0 Both 5.1 N = (177) 73 TABLE 5: ESTIMATED TOTAL VALUE OF ALL GRANTS RECEIVED FROM KTI (IN US DOLLARS) Less than $3,000 3.4% $3,001 to $10,000 4.6 $10,001 to $20,000 7.8 $20,001 to $30,000 15.0 $30,001 to $40,000 17.6 $40,001 to $50,000 20.9 $50,001 to $60,000 8.5 $60,001 to $70,000 4.6 $70,001 to $80,000 5.2 $80,001 to $100,000 3.9 $100,001 to $200,000 3.9 $200,001 to $600,000 2.0 $600,001 to $850,000 2.0 N = (177) 69.9 percent of all grants were between 10 and 60 thousand dollars 53.6 percent of all grants were between 20 and 50 thousand dollars TABLE 6: GRANTEE’S GENDER Male 75.6% Female 24.4 N = (172) TABLE 7: GRANTEE’S AGE 21-30 12.0% 31-40 34.7 41-50 25.2 51-60 21.5 61-65 6.6 N = (167) 74 TABLE 8: GRANTEE’S LEVEL OF EDUCATION Secondary school graduate 4.3% Some post-secondary or some university education 23.8 University graduate 36.6 Some post graduate work 6.7 Post-graduate degree 27.4 Other 1.2 N = (164) TABLE 9a: WAS GRANTEE EMPLOYED BEFORE HIS/HER ORGANIZATION RECEIVED A GRANT FROM KTI? Yes 81.5% No 18.5 N = (173) TABLE 9b: GRANTEE’S AREA OF EMPLOYMENT BEFORE KTI GRANT Government (excluding teaching, military or police) 13.2% Education, teaching 15.9 Military or police .7 Health services, medicine 2.6 Other professional, e.g. legal services, accountant, etc. 11.9 Small business—a firm with less than 20 employees 1.3 Moderate to large business—a firm with over 20 employees 5.3 Creative arts, e.g. musician, theater, artist .7 NGO or community based organization 35.8 Self employed 4.6 Other 7.9 N = (151) 75 TABLE 9c: NUMBER OF YEARS GRANTEE EMPLOYED BEFORE RECEIVING GRANT FROM KTI 1 to 5 years 33.1% 6 to 10 years 20.1 11 to 15 years 15.9 16 to 20 years 8.9 21 to 25 years 2.8 26 years or more 13.7 N = (139) TABLE 10: NUMBER OF HOURS PER WEEK GRANTEE DEVOTES TO HIS/HER ORGANIZATION 10 hours or less 10.0% 11 to 20 hours 14.4 21 to 30 hours 10.9 31 to 40 hours 31.2 41 to 50 hours 23.8 51 hours or more 10.6 N = (160) TABLE 11: HAVE YOU BEEN INVOLVED WITH OTHER NGOS SIMILAR TO (NAME OF ORGANIZATION THAT RECEIVED GRANT FROM KTI) TO IMPROVE THE LIVES OF THE KENYAN PEOPLE? No, this is the first organization with which I have been involved 12.7% Yes, but only one other organization 12.7 Yes, I have been involved with several (two or more) other NGOs 74.6 N = (166) TABLE 12: HAVE YOU EVER LIVED IN A MAJOR URBAN AREA (NAIROBI, NYAKURU, MOMBASA, ELDORET OR KISUMU)? Yes 91.6% No 8.4 N = (167) 76 TABLE 13a: HOW FREQUENTLY DO YOU USE THE INTERNET EACH WEEK? I do not use the Internet .6% Once a week 2.3 Two or three times a week 9.2 Almost every day 39.7 More than once a day 48.3 N = (173) TABLE 13b: THINKING BACK TO THE TIME BEFORE YOUR ORGANIZATION RECEIVED ITS GRANT FROM KTI, DID YOU USE THE INTERNET LESS FREQUENTLY, MORE FREQUENTLY, OR ABOUT THE SAME AMOUNT OF TIME AS YOU DO NOW? I did not use the Internet 2.3% I used the Internet less frequently 38.2 I used the Internet more frequently 30.1 I used the Internet about the same amount of time 29.5 as I do now N = (173) TABLE 14: PRIMARY PURPOSE OF GRANT RECEIVED BY GRANTEE’S ORGANIZATION To prevent violence and conflict 39.5% To resolve disputes 1.3 To reduce hate speech and xenophobia 3.3 To Inform and educate public about the new constitution 19.7 To strengthen key democratic institutions 3.3 To strengthen CSOs to monitor governmental operations 4.6 To implement reforms specified in the National Accord 7.2 Other 21.1 N = (152) TABLE 15: HAS YOUR ORGANIZATION RECEIVED FUNDING OR SUPPORT FROM OTHER INTERNATIONAL DONORS OR NGOS? Yes 48.8 No 51.2 N = (168) 77 TABLE 16: ESTIMATED TOTAL VALUE (IN US DOLLARS) OF GRANTS RECEIVED FROM OTHER DONORS OR NGOS Less than $10,000 9.4% $10,001 to $20,000 5.7 $20,001 to $40,000 7.5 $40,001 to $60,000 9.4 $60,001 to $80,000 7.5 $80,001 to $100,000 5.7 $100,001 to $120,000 11.3 $120,001 to $200,000 11.3 $200,001 to $800,000 17.0 Cannot be determined: 15.1 N = (53) Median value: $90,588. TABLE 17: DID YOU OR YOUR ORGANIZATION APPROACH KTI FOR FUNDING OR DID KTI APPROACH YOU (PLEASE ANSWER IN RESPECT TO FIRST GRANT RECEIVED)? We approached KTI 73.0% KTI approached our organization 21.5 Neither side initiated the proposal. Members of our organization and KTI met each other and the conver- sations about our activities and what activities KTI wanted to support evolved over time 5.5 N = (163) HERE ARE SOME QUESTIONS ABOUT YOUR EXPERIENCE WITH KTI’S PROJECT DEVELOPMENT OFFICER (PDO) WHO HELPED YOU DEVELOP YOUR GRANT. TABLE 18a: HOW FLEXIBLE OR “EASY” WAS IT TO WORK WITH THE PROJECT DEVELOPMENT OFFICER(PDO)? Very flexible and easy 82.5% Somewhat flexible and 14.0 Not flexible or easy 3.5 N = (171) 78 TABLE 18b: HOW EFFECTIVE WAS THE KTI PROJECT DEVELOPMENT OFFICER IN HELPING TO MAKE THE ACTIVITIES OF YOUR ORGANIZATION BETTER THAN THEY WOULD HAVE BEEN HAD THERE BEEN NO PDO? PUT DIFFERENTLY, DID THE PDO IMPROVE THE PROJECT AND RAISE THE PROBABILITY THAT IT WOULD ACTIVE ITS OBJECTIVES? Very effective 86.5% Somewhat effective 12.3 Not effective 1.2 N = (171) TABLE 18c: ALL IN ALL, HOW WOULD YOU RATE THE ASSISTANCE YOU RECEIVED FROM THE PDO? Excellent 56.5% Very good 34.7 Fair 7.6 Poor 1.2 N = (170) TABLE 19a: HOW FLEXIBLE OR “EASY” WAS IT TO WORK WITH THE GRANTS MANAGER (GM) ASSIGNED TO YOUR GRANT? Very flexible and/or easy 75.7% Somewhat flexible 17.3 Not flexible or easy 6.9 N = (173) TABLE 19b: HOW EFFECTIVE WAS THE GM IN HELPING YOU DEVELOP AN APPROPRIATE BUDGET FOR THE GRANT TO YOUR ORGANIZATION SO THAT IT COULD CARRY OUT ITS WORK? Very effective 80.9% Somewhat effective 16.7 Not effective 2.5 N = (162) 79 TABLE 19c: HOW EFFECTIVE WAS THE GM IN ARRANGING FOR THE “IN-KIND” SUPPORT RECEIVED BY YOUR ORGANIZATION? Very effective 78.5% Somewhat effective 17.8 Not effective 3.7 N = (163) TABLE 19d: ALL IN ALL, HOW WOULD YOU RATE THE ASSISTANCE YOU RECEIVED FROM THE GRANTS MANAGER? Excellent 46.8% Very good 40.9 Fair 9.4 Poor 2.9 N = (171) TABLE 20a: WHAT TYPE OF GRANT DID YOUR ORGANIZATION RECEIVE? An “in-kind” grant 78.7% A cash (i.e. money) grant 14.8 A “fixed obligation” grant 3.0 Other 3.6 N = (169) TABLE 20b: ANSWER ONLY IF YOU RECEIVED A GRANT “IN-KIND” The in-kind support created more difficulties than benefits. We would not do it again 9.4% The in-kind approach has both benefits and problems. 15.4 The in-kind support made it easier to implement the project, because we were able to focus on the activities and not on finance or procurement 75.2 N = (149) 80 TABLE 20c: WERE THE FUNDS AND IN-KIND GRANTS RECEIVED FROM KTI SUFFICIENT TO CARRY OUT THE ACTIVITIES AGREED TO IN THE GRANT AGREEMENT? Yes, the funds were sufficient 78.3% No 13.4 Partially—we ran out of resources. 8.3 N = (157) TABLE 21: WHEN YOU LOOK BACK ON THE ACTIVITIES CARRIED OUT BY YOUR ORGANIZATION WITH THE GRANT IT RECEIVED FROM KTI, WOULD YOU SAY THAT THE ACTIVITIES REFLECTED WHAT YOUR ORGANIZATION ORIGINALLY WANTED TO DO WHEN YOU SOUGHT A GRANT FROM KTI? OR, WOULD YOU SAY THAT THE ACTIVITIES REFLECTED WHAT KTI WANTED TO DO? OR, WAS IT A MIXTURE OF BOTH? The activities funded by the grant largely reflected what our organization wanted to achieve. 70.3% The activities funded by the grant largely reflected what KTI wanted to do and less of what we wanted to do. 5.2 It was a mixture of both—the grant helped us carry out activities that we wanted to do, but they also furthered the objectives of KTI 24.4 N = (172) TABLE 22: HOW SUCCESSFUL WERE THE ACTIVITIES OR PROGRAMS FUNDED BY THE KTI GRANT TO YOUR ORGANIZATION IN TERMS OF ACHIEVING THEIR OBJECTIVES? Very successful 86.5% Somewhat successful 12.4 Not successful .6 Other .6 N = (170) 81 TABLE 23a: LOOKING BACK ON THE SUPPORT YOUR ORGANIZATION RECEIVED FROM KTI, WOULD YOU SAY THAT THIS SUPPORT STRENGTHENED YOUR ORGANIZATION IN TERMS OF BUILDING ITS CAPACITY TO CARRY OUT ITS ACTIVITIES? Yes 91.8% No 8.2 N = (171) TABLE 23b: LOOKING BACK ON THE SUPPORT YOUR ORGANIZATION RECEIVED FROM KTI, WOULD YOU SAY THAT THIS SUPPORT STRENGTHENED YOUR ORGANIZATION IN TERMS OF BUILDING ITS CAPACITY TO CARRY OUT ITS ACTIVITIES, AND IF SO BY HOW MUCH? The support strengthened the capacity of my organization a lot 88.6% The support strengthened the capacity of my organization somewhat 10.1 The support neither strengthened nor reduced the capacity of my organization. .6 The support led to a lot of problems at my organization .6 N = (158) TABLE 24: LOOKING BACK ON YOUR EXPERIENCE WITH KTI, WOULD YOU WORK WITH KTI AGAIN? Yes 98.8% No 1.2 N = (171) TABLE 25a: THINKING ABOUT THE COUNTY IN WHICH YOUR ORGANIZATION CARRIED OUT THE ACTIVITIES OF THE FIRST GRANT IT RECEIVED FROM KTI, WAS THIS AN AREA OF POST-ELECTION VIOLENCE (PEV) IN JANUARY AND FEBRUARY 2008? Yes 86.6% No 13.4 N = (172) 82 TABLE 25b: COMPARED TO THE SITUATION THAT EXISTED IN THE COUNTY AFTER THE LAST ELECTION, IS THE COUNTY PRESENTLY MORE PEACEFUL, LESS PEACEFUL OR ABOUT THE SAME AS IT WAS THEN? More peaceful 83.0 Less peaceful 7.0 Same 9.9 N = (171) TABLE 26a: LOOKING AHEAD TO THE COMING ELECTION, DO YOU EXPECT THERE TO BE OUTBREAKS OF VIOLENCE? Yes 29.9% No 70.1 N = (164) TABLE 26b: IF “YES” WHY? Animosity between members of different ethnic Groups has continued to thrive 34.8% Security has not been enhanced 17.4 PEV education has not been effective 21.7 Other 26.0 N = (23) TABLE 26c” IF “NO” WHY? Kenyans have matured; considerations of ethnicity are less important than before 13.6% Security has been enhanced 11.1 PEV education was effective 58.0 Other 17.3 N = (81) 83 ANNEX IV THE “BRIDGE” TO USAID’S DEVOLUTION SUPPORT PROGRAM (AHADI) IV.1 Overview This evaluation was not required to address the current involvement of KTI in the just-beginning devolution exercise in Kenya. However, in the course of part 2 of the fieldwork, it became obvious that devolution has now become one of the main theatres of political contestation in the country, and therefore the issue could not be avoided. On its part, KTI has chosen devolution as one of its exit strategies and is currently engaged in a number of devolution activities. These include: working with the Centre for Parliamentary Studies to help county assemblies and executives understand their roles; training for capacity development in such areas as county￾based planning, procurement and financial management; developing new training modules for counties at the Kenya School of Government. The focus on devolution has led to some changes at the management level. For example grants are now going to counties not to CSOs as before, and PDOs have requested and received some training on devolution at the Kenya School of Government. The PDOs were keen to discuss devolution, especially in the context of their own careers. Grantees had an interest in the subject because they want to be party to the devolution exercise and also because most of them sense not just the potential gains but also the potential pitfalls of devolution. Others, including non- grantee key informant saw devolution as a bold political and development endeavor but one fraught with many challenges. The consensus in the country is that there is no consensus on the what, the why and the how of devolution even though it is constitutionally mandated. For this reason, devolution has become the “hot ugali” of Kenyan politics. To get engaged in it, therefore, is to engage Kenyan politics. In this brief appendix we review the politics of devolution, the space available for donor assistance, the key issues that the proposed AHADI will have to confront and very briefly where KTI might make a contribution. We argue that devolution is an intensely political issue, that though room exists for external donor interventions, it is not all technical in nature and neither can it be expected to be. Governance and democracy issues including peace-building, conflict￾mitigation and devolution within counties are a key part of this space. Furthermore, the proposed AHADI program will do well to understand the nature of devolution and prepare in advance before entering the arena. Finally, we argue that KTI can contribute to devolution and indicate a few areas where it can do so. IV.2 The Politics of Devolution Devolution has been mandated by the 2010 constitution which laid down the structures and also prescribed the transitional arrangements between the passing of the constitution and the 2013 elections. In theory, devolution should not be contentious at least not from constitutional and legal perspectives. But it is, and there are several reasons for this. 84 i. Different actors, including political actors and local communities have differing visions of devolution. To some, it is a form of federalism which means the country is no longer a unitary state. To others, it is the traditional form of political decentralization within a unitary state. Still others see devolution in social terms as the platform for local social/tribal autonomy that is antithetical to the national state. This group of closet secessionists views the state as the enemy even though they will not go as far as openly advocating secession as the MRC is doing in the former Coast province. Others refuse to admit that devolution by definition entails power redistribution and want to view it in purely technical/ developmental terms. ii. Devolution has become part and parcel of “political tagging.” The CORD Alliance has all along sought to portray itself as the champions of devolution while portraying the Jubilee Alliance as reluctant supporters of, if not actually opposed to it. The Jubilee Alliance on its part seems to view CORD as reckless and populist on matters devolution, even though it has not said this loudly. Perhaps because of this, the Jubilee Alliance has not managed to shed the image of being anti-devolution. In this tagging war, discussion of devolution is invariably politicized. iii. As the implementation of devolution gets under way there are genuine disagreements on the pace at which devolution should be implemented and the sequencing of the responsibilities to be devolved. The constitution and the Transitional Authority anticipated a phased transition during which responsibilities would be transferred to the counties gradually and in sequenced manner depending on the state of capacity to handle the devolved activities. The newly elected county governors seem to have a different view. They want the responsibilities transferred at a much faster pace and are now seeking to amend the constitution to have more central government revenue transferred to the counties. Because of the politics of it, no leader would dare tell the governors that the quickest way to kill devolution would be to transfer responsibilities too quickly without the requisite capacity not even when this is accompanied by more revenue. Clearly, the governors have their own vision of devolution. iv. Granted that devolution is about power sharing through redistribution in the first instance, still the country did not fully anticipate the on-going power struggles between institutions. The relationship between the National Assembly and the Senate is now the subject of a power struggle, as well as the relationship between the governors and the county assemblies in some counties. There is also a supremacy war between the senators and the MPs and between the senators and the governors though the latter is low key at the moment. One suspects that it is a matter of time before the judiciary is sucked into the inter-institutions supremacy wars when it will required to interpret the constitution. v. The politics of resource sharing between the central government and the counties is complicated by the fact that some counties are more resourced than others, which introduces inter-county politics on the formula for sharing equalization funds and other special funds. The formula worked out by the Commission on Revenue Allocation is by no means uniformly accepted, and as the country discovers more natural resources, the issue will become acute. How much of the oil revenue from Turkana—as the deposits have now been 85 declared commercially viable—will the county accept and how much will the central government accept in order to boost its redistribution capacity? vi. In order to appreciate the politics of devolution, one also has to appreciate that so far, the implementation of devolution has been poorly planned. Following the promulgation of the constitution in 2010, only the constitutionally mandated legislation was passed between 2010 and the 2013 elections, and quite a number of these crucial pieces of legislation were passed in a hurry in order to beat the constitutional deadlines. No wonder that the politics of devolution seems to be a free for all given that the pieces of legislation were not sufficiently discussed and subsequently they have not been internalized by the public and the political elite. This explains why the governors would more or less casually seek a referendum on an issue that has already been legislated and that can presumably be subjected to judicial interpretation without resorting to a referendum so soon after a very divisive election. Finally it should be said that the civil society does not have the technical capacity to offer a guiding hand in the politics of devolution. It too is caught in the tagging game. Under these circumstances the biggest challenge that will face any external interventions are not likely to be at the national level because there are well established modalities for this. Rather the challenges are likely to be in the counties where the rules of engagement and modalities for doing so are still evolving and doing so under less than optimal circumstances. Still, there are opportunities for external interventions at the county level but the path is treacherous. IV.3 External Interventions for the Counties There are two types of spaces available for useful external interventions—the technical and the political. In the past, donors have been engaged in both. We have some feel of how this has worked out even though the impact of the work is often difficult to measure. With devolution the job of capacity building is now more complicated, even when the issues appear to be purely technical in nature. This is because devolution is new to everyone and also because of the politics of it. The technical problems can be reduced to lack of county capacities in areas critical for them to start functioning. Some of the most important of these are: budgeting, planning, financial management, procurement, instilling public ethics and avoiding corruption, inventorying local resources and assessing their revenue potential; understanding the roles under the constitution, and understanding legislation affecting counties. i. Budgeting. Following the installation of the county governments in March 2013, the country has been treated to the spectacle of counties producing spectacularly large budgets that are way beyond even the proverbial wish lists. These were clearly produced by individuals without any technical know- how or who did not care to consult those who did. Alternatively, the budgets were produced with the mindset that the central government will, if forced, provide the counties with more resources than currently mandated under the constitution on the assumption that the central government has the resources. Either way, a useful intervention would seek to train county officials on the meaning of a budget, both at the county and national levels and the linkages between the two. The training should also seek to 86 emphasize the obvious fact that a budget is highly intertwined with fiscal policies and capacities. ii. Planning. Kenya has plenty of national capacity on planning although it is only recently that local authorities have been hiring planners versus engineers. There is also a long history of top down planning that fails to capture basic priorities on the ground and include citizen feedback. Furthermore, if one mines the history of decentralized planning in the country, one is likely to find plenty of lessons on the “does” and “don’ts” of local planning which is, of course, a political process as well. One of the lessons is the need for integrated local planning based on data-based inventory of local resources. The District Development Officers in the now defunct District Focus For Rural Development acquired plenty of experience on local planning and on how to generate the requisite data. Unfortunately, any mention of past practices is now being dismissed as “decentralization” as opposed to “devolution”. In a rational environment, planning and budgeting should be undertaken simultaneously and both should learn from the past. Thoughtful Neutral external intervention can help underscore this. iii. Financial management. Poor county financial management is chronic because it has been inherited intact from the defunct local authorities, all of which had little capacity in public finance management. The problem could be even more serious for the counties in that unlike the now defunct system, the central government cannot exercise day to day supervision, something which is good for devolution, but which also means the counties have to get it right. The counties themselves agree with the central government that they have little, if any capacity on public finance management. One would, therefore expect, as the local media is now pointing out, that the counties should first demonstrate prudent public finance management before demanding more funds from the central government. Regardless of this politics, the need to train county officers is extremely urgent, otherwise some of the governors and the some county officers will end up in jail. This is such a huge problem that donor intervention should be well planned and if possible medium to long term. iv. Procurement, public ethics and corruption. The one single issue with the highest capacity to devolve corruption is procurement. This is because the counties have little positive to learn from the central government or local authorities. The problem here, is that the issue is not merely one of skills and competencies but largely one of public ethics. Kenya has plenty of procurement capacity that can easily be transferred to the counties. What is in serious deficit is public ethics in procurement. The dilemma for capacity building interventions is, therefore, how to facilitate the transfer of skills, minus the pervasive culture of corruption. Still, some external intervention could be useful in facilitating basic training on the legal procurement frameworks and specific legal instruments on procurement. v. Local resources, fiscal policies and revenue generation. Devolution pessimism has it that, under the current central government financial transfers, most counties are not likely to take off. The easy and optimistic solution, therefore, is to drastically increase the county entitlements from the central government as the governors are seeking to do through a referendum. The not so easy but equally optimistic solution lies in local revenue generation, a perennial problem of devolution everywhere. There are two, but related capacity issues here. The first is the capacity to take an inventory of local resources. The second is the capacity to 87 establish the resource’s revenue generation potential. Counties currently lack both capacities but especially the second. This matter is highly political especially as the country gears up towards exploiting the newly discovered natural resources. Empowering counties to demand more will invariably be seen as political by the central government. On their part, the counties will seek to underestimate their own potential to generate local revenue, especially if they can force more from the central government without having to increase unpopular local taxes. vi. Understanding roles and boundaries. The country is at the moment full of confusion on the exact roles and mandates of key actors in implementing devolution. The problem seems to be that, even though the constitution is in place and the key pieces of legislation are also in place, none seems to have stamped its authority, perhaps because the provisions are not well known, and in some cases, the constitutional and the statutory instruments are ambiguous. One example is the relationship between the Inspector General of Police and the Police service Commission. Another is the relationship between the National Assembly and the Senate, which has led the Senate to seek an advisory opinion from the Supreme Court. At the county level, the governors, the assemblies, the speakers, and the county executive committees, are all trying to understand the nature and extent of their mandates. The capacity building job is relatively simple; to take the county members, including the governors, through the constitution and the legislations so that they understand what they can/cannot do. While the central government, through the Kenya School of Government can do most of this, external financial injection would hasten the pace of delivery. The technical problems confronting the counties are almost exactly the same as those experienced by the now defunct local authorities except, this time the political environment is different as are the administrative structures. Because the problems are historical and well known in the devolution literature, they ought not be contentious but because of the new political and social environment, they too are likely to be viewed politically and socially rather than technically the way they should be. Another difference is that, whereas the previous local authorities experienced the problems existentially when they already existed, the counties have yet to take off, which explains the current impatience, especially by the governors and the senators. But, like the old local authorities, the degree of poor capacity differs from county to county, except this time, the voices from the counties are louder and better coordinated through the Governor’s Forum and the senate. This means the issue of relative capacity is likely to find powerful political voice. This in turn, has implications on how donor assistance for capacity development is targeted. It is also important to observe that unlike previously when the minister for local government enjoyed immense powers, including powers to coordinate external assistance to local authorities, it is not clear yet how much power the minister for planning and devolution can exercise on external assistance to counties. What is clear at the moment, is that the central government is uneasy about counties going out on their own to seek external assistance. 88 IV.4 How to Develop Institutions of Accountability at the Local Level The political capacity issues are largely democracy and governance related. The main issue here is how to develop local accountability institutions. The issue has several threads. i. Most of the devotion politics outlined above is likely to be replicated within the counties in the form of intra-county politics of devolution as social and political actors push for devolution within counties and some form of accountability. As one would expect, the county centers are likely to resist the push for intra- county devolution as well as that for more accountability. ii. The demand for accountability is closely related to the local demand for democracy. Unfortunately, local peasant populations are, on account of parochial loyalties, highly susceptible to manipulation by those in power. This is compounded by the fact that rural￾based civil society is not as articulate as its urban counterpart with the exception of the producer-based civil society, such as agricultural and pastoralist cooperatives and professional organizations such as the teaching profession. Even then, it is not clear how well this segment of rural civil society is positioned to articulate issues of its concern within the devolved system of governance and the anticipated local democracy. It almost certainly is not positioned to articulate human rights issues based on abstract principles of democracy. iii. As with other local resources, we have a poor idea of who these local actors are or are likely to be. This in turn means we have a poor idea of the likely local demand for accountability and democracy. In all, then, county civil society is desperately needed, given that county assemblies are weak and therefore likely to be compromised by the executive, and in particular the governors. The capacity building implications of the three threads are fairly clear. These are: map the local nodes of potential demand for accountability and democracy; select those with the highest potential for multiplying the demand especially membership organizations such as farmers, unions, cooperative societies, and teachers. This is the second generation civil society that is capable of sustaining itself compared to the first generation of human rights organizations that has been wholly dependent on donors; provide some training on the nature of intra-county devolution and democracy. In essence, this means helping create county-specific civil societies. This kind of local capacity building is likely to prove more politically sensitive than national capacity building. It is also likely to prove more controversial than the first generation civil society because it will resonate more with the local people and will, therefore be closely “protected” by those in power. Any external assistance has to be very well designed and extremely well marketed. Among other things, donor interventions would be required have in￾built mechanisms to manage central government- county relations as well as intra-county relations. 89 IV.5 The Road to AHADI The proposed AHADI program of assistance to counties is still being planned but is expected to cover 10- 15 counties. Bidding for the implementation of the program is currently under way, with the winner of the bid expected to be announced in the next few months. It is important for AHADI to start by reflecting on several strategic and operational difficulties that could very well derail the program if they are not thought through at the very beginning. Some of major ones are: i. Choice of counties. The prevailing politics of devolution is such that power and resources￾competition is not only between the central government and the counties, but also among the counties themselves. This means that any assistance to the counties that does not cover all them will have to justify the choice of the counties to be assisted, to both the central government and to the counties. Failure to do so will result in accusations of favoritism. Given that AHADI is planning to cover 10-15 counties out of the 47 counties, it is important to develop transparent criteria for choice of counties and to subject these to discussion by both the central government and the counties. The discussion will likely not lead to a consensus but the aim should be to obtain support from the central government and hopefully a majority of counties. The choice of counties further requires that USAID coordinate its efforts with the other donor agencies planning to enter this field so that there is not an overlap and duplication of effort in some counties while others receive no assistance. ii. Involvement of the central government. As indicated earlier, the central government is somewhat uneasy about direct county contacts with donors, no doubt on account of the divisive aftermath of the 2013 elections but also because devolution is yet to settle in a predictable and manageable pattern. While direct contacts cannot be entirely avoided, a large program like AHADI cannot afford to model itself after the traditional civil society support programs. Doing so would be seen as giving credence to some strange political theories floating around and almost suggesting that the counties are sovereign and not simply distinct as stated by the constitution. An easy way out of this problem would be to first agree with the counties on county priorities, then provide assistance, especially training, through national institutions such as the Kenya School of Government. This is already happening, is what happened under the District Focus For Rural Development and can provide aspects of the training model. The major advantage of training centrally is that counties can be clustered on the basis of identical priorities, design a curriculum based on themes for each cluster and offer the training at one go to all the counties falling under that theme/cluster. iii. Sequencing for capacity building. County capacity needs are immense and will most certainly take a decade or more to meet. Great care is, therefore needed when promising what to provide and when. The temptation is to announce a “big bang figure” that creates huge expectations that might not reflect the larger context. It is imperative therefore, to manage this politics of donor assistance and one way of doing so would be to inventory capacity development needs along the lines outlined earlier and sequence them in order of accepted priorities. Agreeing on the menu of priorities and sequencing of the menu items are all likely to prove highly contentious and this time, for very genuine developmental reasons as opposed to politics. The 47 counties are highly heterogeneous between them and within them. They have genuine reasons to opt for different sequencing. The danger for external 90 assistance is that it risks getting hopelessly caught up in a complicated development debate on these matters. One way out of this is for donors to quickly develop/acquire county￾specific expertise alongside country expertise. iv. Timing and pacing. Related to sequencing is the timing and pacing of delivering assistance. The problem here, is that if the assistance is not delivered at the right time and on time, it will get caught up in the current quicksand that is devolution today. The traditional mode of delivering assistance has often entailed delays in delivery and quite often, poor pacing as the assistance is delivered either too slowly or too quickly. The danger now seems to be the opposite. Because of the novelty of devolution and the sheer size of the reforms it entails, both in central government institutions and in the counties, the temptation is to rush the assistance, especially in the initial stages of devolution. While this might satisfy some players, it could lead to mistakes that might poison the atmosphere for years to come. In this context, it is imperative to keep remembering that devolution is not an event and neither is it about the next election cycle. It is a long term process that, properly implemented, is meant to reconfigure the state. There is no need to rush in before the major issues have been thought through and these include timing and pacing. v. It’s about resources stupid! The controversies about devolution are not unique to Kenya in that all over Africa and elsewhere, they are primarily about resources distribution. It is unlikely that donor assistance will escape this central contestation. The problem here is about balancing the assistance portfolio. In the Kenyan context, this problem has two dimensions, the regional and the sector dimensions. The regional dimension is especially sensitive to the tribal nature of Kenyan politics now compounded by the regional nature of both the governing coalition and the opposition. The problem is an extension of that of choice of counties discussed above but this time, manifesting itself in budgetary allocations. The sector dimension is related to the regional in that some sectors are more important in some areas than others. If the allocations are viewed as skewed towards some sectors in such a manner as to favor some areas, then the program will be politicized. While politics of this nature cannot be avoided altogether, it would help to keep it to a minimum. This is especially important if donor assistance goes beyond institutional capacity development to development projects. IV.6 KTI Contribution to AHADI KTI’s exit strategy through devolution, could provide useful tips on how to implement AHADI. Even if the exit strategy was not expressly designed to yield lessons for AHADI, those involved must have encountered and hopefully reflected on the issues outlined above and could, therefore still make a useful contribution to the proposed program. KTI could contribute useful lessons in at least three areas: i. How to manage county politics. Perhaps the biggest bequest of the KTI program is the wealth of local politics that has been accumulated by the program. This has been documented in KTI records but a great deal of it is in the heads of KTI personnel. Fortunately the PDAs in particular are already part of the devolution exercise and are already passing on the political 91 expertise. Hopefully, many of them will continue to be players in the exercise when AHADI comes on board. ii. Choice of civil society. Creating demand for accountability at the county level will very much depend on which civil society actors are chosen for assistance. At the county level, the choosing is more difficult than at the national level because the donor have little experience here. KTI again, has enormous amounts of knowledge on local civil society, having essentially created and nurtured some of this rural civil society iii. Linking devolution and peace at county level. The discussion on devolution has not yet embraced the linkage between the success of devolution and existence of peaceful means to resolve problems of resource distribution and minority representation in the counties. Yet, devolution is threatened in several counties precisely by the absence of the rule of law and continuing use of hate speech and ethnic mobilization at the local level to ensure an exclusive block controls county government. These dynamics have been a problem in Tana River, Marasabit, Taita Taveta, Kilifi, Lamu and Kwale. In Kwale for example, devolution is viewed suspiciously, by some, as a trick to dissuade those who might want to join the MRC from doing so. Unfortunately the potential for emergence of similar secessionist movements is quite real especially in light of the numerous natural resources now being discovered in the country. The need to sustain peace-building in the counties is, therefore, of paramount importance to the success of devolution. KTI knows this better than anyone else. It is well positioned to push this agenda using its extensive local networks.