Kenya

Conrad M. Bosire

July 2019

pdf download

Recommended Citation:

Bosire, Conrad M. 2019. “Kenya”. In Autonomy Arrangements in the World. www.world-autonomies.info.

This content is licensed under a Creative Commons Attribution 4.0 International license.

1. Essential Facts and Figures

Kenya’s sub-national autonomy arrangements are secured through a devolved system of government comprising the national government and 47 county governments. The devolved system of government was established through the current Constitution, which was adopted in August 2010 and devolves resources and powers, through a symmetric arrangement, to the 47 county governments. The 2010 Constitution replaced the old Constitution that had provided a strongly centralized and unitary system of government that vested the President with immense political powers. Before delving into the history and current structures of sub-national autonomy, it is important to provide essential facts and figures on the Kenyan state.

Kenya’s last national census (2009) placed the total country’s population at 38.4 million people, although the population projections show that, at an estimated annual growth rate of 1 million, this number has been surpassed to approximately 50 million in 2018 (Kenya National Bureau of Statistics 2009; United Nation 2018). The majority of Kenya’s population is composed of those aged below 35. Kenya is a linguistically and ethnically diverse country, comprising as many as 43 recognized ethnic communities. As per the 2009 census, the largest ethnic community, the Kikuyu, was over 6.5 million. The second largest community, the Luhya, was 5.3 million while the third largest community, the Kalenjin, was 4.9 million. The other larger communities, as per the 2009 census, were the Luo at 4 million, the Kamba at 3.9 million, the Somali at 2.4 million, and the Kisii at 2.2 million (Kenya National Bureau of Statistics 2009). There are 38 other ethnic communities, bringing the total number of ethnic communities in Kenya to 43 in total. The criteria for classification of ethnic groups is, however, general and haphazard and has, in some cases, classified smaller groups, who identify themselves as distinct, within larger ones. Smaller communities include non-indigenous groups (classified generally as “non-Kenyan” or “other Kenyan”), such as Kenyan Asians (mostly from the Indian sub-continent), Kenyan Europeans, and Kenyan Americans (Ibid.) most of whom are based in major urban areas across the country.

Kenya borders Tanzania and Uganda to the southeast and east respectively. To the north and northwest lie the Republics of South Sudan, Ethiopia and Somalia. The current Kenyan borders, which contain a geographical area of approximately 569 300 km2, are a product of the arbitrary colonial boundary demarcation between imperial colonial powers. As a result, along the Kenyan borders, communities share the same culture and heritage (even families) can be found in both Kenyan territory and neighboring countries. For example, Somali communities live on both sides of the 600-kilometre Kenyan–Somali border. 

Despite the fact that almost every community has its own local language, Kenya has only two official languages, English and Kiswahili. The two languages can be described as ethnically or politically neutral (not associated with any local community) and this may explain the universal embrace of the two languages; there is a small community of ethnic Swahili people along the Kenyan coast. The law prohibits the use of local languages in official forums[1] and the language of instruction in schools, all the way to university, is English. According to the 2009 census, Kenya’s large majority, around 31 million were Christian while the 4.3 million were Muslim. The Constitution expressly states that there is no state religion.[2]

Kenya is a developing economy with a GDP per capita of 1,377 USD (World Bank 2017, 45). The largest contributors to GDP growth in Kenya are agriculture, trade and industry, and services (Ibid.). While subsistence agriculture is the mainstay of the majority of Kenya’s rural-based population, urban areas make the largest contribution to GDP. There are great disparities in terms of development and access to basic services between rural and urban areas. For instance, while the national average access for improved water sources is 63.2 %, this translates to 81.6 % for urban areas and 56.8 % for rural areas (Kenya Institute of Public Policy Research and Analysis 2016, 152). Furthermore, while rural–urban migration and rural–rural settlement across the country has altered the ethnic composition across the counties, most of the counties have majority ethnic communities mostly spread out in vast rural areas. The 2009 census revealed that 67.7 % of Kenyans were living in rural areas. However, a rapid urbanization rate of 4.3 %, shows that half of Kenyans will be living in cities or major urban areas by 2050 (World Bank 2016, 3.).

2. Autonomy and State Structure

The 2010 Constitution establishes 47 county governments to which symmetrical powers are devolved.[3] While the Constitution recognizes, in principle, the need for accommodation of diversity, there are no asymmetrical autonomy arrangements within the constitutional framework. The 47 county governments replaced the former eight provinces[4] that had been carved out in a manner that created ethnic homelands for each of the five largest ethnic communities (see Map 1).[5]

Map 1. Boundaries of Kenya’s regions (pre-2010) and counties (post-2010)

The boundaries of the 47 counties are also inherited from the colonial administrative boundaries that were ethnically defined. The larger ethnic communities (the Kikuyu, Kalenjin, Kamba, Luhya, and the Luo) have been split into several counties (around 22 of the 47 counties) while the smaller ethnic communities each have a “home county” or two. However, there are several other communities in the smallest category, with no home county and are minorities within the current 47 counties.

The Senate, which is the upper house of the bicameral national legislature (see section 5.2. below) represents the counties at the national level.[6] The senators are directly elected by voters from the 47 county governments and are mandated with special powers to safeguard and protect the autonomy of the county governments.[7] It is worth noting that the 2010 Constitution of Kenya explicitly bars the formation of regional parties. The Constitution provides that all parties must have a “national character” and they shall not “be founded on a religious, linguistic, racial, ethnic, gender or regional basis or seek to engage in advocacy of hatred on any such ideas.”[8]

The national executive structure is that of a presidential system of government. The president, who is elected by a margin of votes and geographical proportion of votes,[9] is the head of the executive. The Constitution provides that, where no candidate attains the prescribed margin of votes, a run-off is held between the two candidates with most votes.[10] The president appoints ministers, who have to be approved by the National Assembly.[11] The national executive is required to represent regional diversity of the country in its composition.[12]

With regard to sub-national autonomy arrangements, the Constitution contains broad principles that reflect the need to accommodate autonomy. Among the objectives of the devolved system of government are to recognize diversity for purposes of national unity, facilitate communities to exercise self-rule, and protect ethnic minorities and marginalized communities.[13] However, there is a historical and political context to the symmetric devolution of power to the 47 county governments. At independence, there were strong claims for regional autonomy that would have necessitated asymmetrical accommodation. Indeed, two of the regions (the one bordering the Somali Republic and the Coast region) had secessionist claims. These claims found a compromise in a semi-federal structure of government that was negotiated by the British government. However, the post-independence government dismantled the regions and imposed on the country a strongly centralized system of government.

In order to achieve the above, the political elite from regions that clamored (and continue to clamor) for regional autonomy was co-opted into the centralist philosophy of the successive governments. These events gradually weakened the popular demand for regional autonomy, leading to the current situation where symmetric devolution of powers is an acceptable arrangement across the country.

3. Establishment and Implementation of Autonomy

The history of Kenya’s approach to ethnicity can be understood in three main phases. First is the colonial era where policies pursued heavily influenced Kenya’s post-colonial state structures and practices with regard to autonomy. The second phase is the independence and post-independence period where the state structure underwent substantial structuring and restructuring as a result of negotiations and differing approaches to autonomy. The third period is of constitutional review which began in the 1990s that led to the adoption of the 2010 Constitution. The issues that have shaped these three phases are discussed below.

3.1. Colonial beginnings of the post-colonial Kenyan state

Being a post-colonial African country, specifically a former British colony, the country’s governance and political system, including approaches to autonomy, were, and still are, heavily influenced by colonial policies. In order to understand this, a brief history of how Kenya’s territory was consolidated is essential. The indigenous communities within present-day Kenya had informal traditional leadership structures, with limited inter-ethnic trading and interaction; there was no overarching socio-political arrangement that would have brought these communities together. At the coastal region, however, the Sultan of Zanzibar had authority over the 10-mile strip along the Kenyan coast, and an undefined territory in the interior (Ghai and McAuslan 1970, 6–7). In the 1880s, European powers (the German and the English) divided East Africa into “spheres of influence” leaving most of the current Kenyan territory under British rule (Ibid., 5). Initially, the British Government granted a royal charter to the British East African Association (which later changed to Imperial British East Africa Company) to administer the territory and gave the Imperial British East Africa Company power to establish administrative structures.

On 15 June 1895, the British government declared a protectorate status over much of what is Kenya’s current territory (Ibid., 12). In the decades that followed, the protectorate government made adjustments to the territory of the Kenyan state that created the current boundaries. First, an agreement of protection of the Ogaden Somali in 1896 led to the expansion of the Kenyan territory northwards (a part of this was transferred to Italy in 1926 to create Italian Somaliland). In 1902, the British government made an administrative transfer of the Eastern Province of Uganda to Kenya, along with a portion of Lake Victoria. In 1925, the Rudolph Province of Uganda (i.e., the current Turkanaland in Northern Kenya) was transferred to the Kenyan colony (Ibid., 13–14). It is important to note that the consolidation of the territory was done through administrative fiat of the British government and sometimes with the coercion of the local communities.

Unlike areas such as Northern Nigeria and Uganda, where traditional kingdoms or leadership structures were allowed limited powers of self-rule, administration of the Kenyan territory was direct, centralized, and hierarchical (Oyugi 1978, 7). The British government set aside areas for exclusive white settlement across the colony, an act that led to displacement of communities from areas that were isolated for white settlement. For the African communities, ethnically exclusive native reserves were created where movement and inter-ethnic interaction was restricted (Kanyinga 2009, 327–28). While the government avoided indirect rule, it created administrative units defined along ethnic lines throughout the colony and restricted African political activity to the district/ethnic level (Okoth-Ogendo 1972).[14] Local Native Councils  were established in 1924 and African political activity was restricted to the councils, which were placed under the direct supervision of the District Commissioners (Oyugi 1978, 7). Indeed, many of the independence leaders started their political careers from the Local Native Councils (Odinga 1976, 15). The policy of “ethnic divide and rule” and pursuit of ethnic enclaves in the colony had an impact on the post-colonial political dispensation in Kenya. 

3.2. Independence talks: unsuccessful claims for regional autonomy and the entrenchment of the centralized state

The colonial policy of pursuit of ethnic divide and rule, and especially the policy of limiting political activity to the district/ethnic level ensured that the African leaders arrived at independence talks divided along ethnic lines. Two main political fronts emerged: leaders from the largest ethnic communities (then the Kikuyu and Luo) formed the Kenya African National Union (KANU). Fearing domination from the larger communities, smaller communities formed the Kenya African Democratic Union (KADU). KANU, composed of the largest and dominant ethnic communities, called for a strong unitary and centralized government, which the party argued was necessary for national unity and development (Ogot and Ochieng’ 1995, 70). The smaller communities, under the umbrella of KADU, with the backing of settler (European) politicians and the Asian community, called for a semi-federal system of government (Ibid.). KADU argued that the there was a danger of domination by a political party, person, group or tribe, and that this risk would be eliminated through decentralization of power where sub-national units would not entirely depend on the center (Ibid.).

While the smaller ethnic groups sought to assert their autonomy through a regional system of government within the Kenyan state, two regions in the country had a bigger agenda, secession. The Northern Frontier District, composed mainly of the Somali community and other smaller communities, wanted to join the Somali Republic that had become independent in 1961 (Castagno 1964) to form a so-called “Greater Somalia” (Soomaaliweyn).[15] A political party, the Northern Province Peoples Progressive Party, was formed to assist in the secessionist cause (Ibid). However, the British government was opposed to the idea of the break-up of the Kenyan territory. Despite setting up two commissions of inquiry to look into the “Somali question” that received strong secessionist sentiment from the region (to the surprise of the colonial government), it declined to grant regional autonomy arrangements (Ibid). In the coastal region, the Arab leaders of the party Mwambao United Front (MUF) championed the secessionist cause for a 10-mile coastal strip (Brennan 2008: 831–61). They relied on the 1895 agreement between the British government and the sultan of Zanzibar that handed the 10-mile strip over to be incorporated as Kenyan protectorate, at the time when the British were consolidating the territory (Ghai and McAuslan 1970, 12–13). The Mwambao United Front used this agreement to seek independence from the Kenyan state. None of the regions were given a right to secede. The British government believed that autonomy questions could be addressed through a territorial arrangement that gave space for regional autonomy. The coastal question was thus resolved through an agreement, which was signed by the Kenyan prime minister and the sultan of Zanzibar in 1963.[16] In the agreement, the sultan agreed to the incorporation of the coastal strip to Kenyan territory on condition that the people were allowed to practice their religion, apply Islamic law to personal law matters such as succession and to ensure, as far as is practicable, that the leaders in the region were locals. The secession threats in the Northern Frontier District, on the other hand, were quelled through military action in the region. Initially the Somali government gave covert support to the secessionist groups, but this support gradually ceased, as internal conflict started crippling the Somali Republic.

While the British government saw that autonomy demands would be accommodated through territorial arrangements, KANU, which had majority support, was opposed to any form of decentralization of power or autonomy. In the end, KANU agreed with a semi-federal structure of government, referred to as majimbo in Kiswahili, but its leaders later confessed that they accepted it only as a means to gain independence (Mboya 1963, 75; Odinga 1976, 146). It is worth noting that the British government pushed for a semi-federal system of government, although the British themselves had established and maintained a strongly centralized and hierarchical colonial system of government (Ghai and McAuslan 1970, 178). There was no experience or learning upon which the new semi-federal system of government would have operated.

The Independence Constitution, adopted in 1963, established a regional system of government that was composed of eight regions. The regions were fashioned along the colonial administrative provinces. The two regions that wanted to secede (North Eastern and Coast) were granted regional governments but only Nairobi, the capital city becoming the eighth region, was granted special status.[17] There was no special status for the two regions with secessionist tendencies. There was no mention of the international agreement between the sultan of Zanzibar and the then prime minister Jomo Kenyatta over the 10-mile coastal strip. The regional system of government provided for symmetrical devolution of powers to the regions. However, while the 1963 Constitution established powers of the regional governments, it contained several claw-back clauses that seemed to re-centralize powers granted to the regions (Ghai and McAuslan 1970, 197–98). 

KANU won the May 1963 elections and interpreted its victory as a referendum against the regional system of government (Okoth-Ogendo 1972, 18). As KANU prepared to take over the government, the British government (to the disappointment of KADU and British settlers) relinquished all responsibility of implementation of the Constitution and other post-independence arrangements to the KANU government (Anderson 2005: 562). Soon after, the KANU government frustrated all efforts to implement the regional system of government. It is through the Senate that the secessionist regions and smaller communities found a voice at the national level (Proctor 1965, 393). The Senate was composed of 41 members elected from 40 senatorial districts countrywide and Nairobi.[18] One of the major decisions that the Senate successfully blocked, albeit temporarily,[19] was the decision to deploy the military to the North Eastern region, much to the chagrin of the KANU government (Okoth-Ogendo 1972, 19).

KANU continued its plans to dismantle the regional system of government. Carefully crafted constitutional amendments gradually weakened the regional system of government, finally abolishing it in 1968. Senators were offered positions in the House of Representatives (the lower house), without undergoing an election, in exchange for abolishing the Senate. The parliamentary constituency boundaries were redrawn to accommodate the senators leading to the abolition of the Senate (Muigai 2001, 131–32). In a span of four years, Kenya resorted to the centralized system of government reminiscent of the colonial governance.

One may wonder why the country did not go into political chaos after abolishing structures that guaranteed regional autonomy. Several factors have been put forward to explain this. Firstly, KADU, which had the support of the smaller ethnic communities and secessionist regions, was gradually weakened by defection of its members to KANU, leading to its folding up in 1964 (Anderson 2005, 563). KADU was the only opposition party at independence thus, KANU was left with little political opposition to its centralization agenda. Secondly, the regional system of government was “embarrassingly complicated” and even KADU admitted that its implementation would face various hurdles (Ibid., 561). Finally, leaders from minority communities and the secessionist regions were enticed with positions in government and co-opted into government largesse for personal gain in exchange for dropping their demands for regional or local autonomy (Kanyinga 2009, 337). 

The push by the KANU government for a centralized and unitary system of government has a broader context. Between the 1960s and 1970s, many African countries gained independence. The emerging leaders called for centralized systems, which they said were necessary for the much -needed unity and development at the time. The semi-federal systems of government that were sponsored by colonial powers were dismissed as “vestiges of colonialism” (Anderson 2005, 562).  In many countries across the continent, independence Constitutions faced similar changes as those in Kenya. Leaders reverted to one-party systems and in worse cases, no political party at all (Wunsch and Olowu 1990, 47; Neuberger 1979, 187; Mueller 1984, 22)   

Claims for autonomy and accommodation of intra-state diversity were shoved aside and replaced with centralized planning and development. Centralized decision-making replaced regional or local decision-making. In many states, sub-national autonomy arrangements were viewed with suspicion by the central government leadership (Neuberger 1979, 174–76). The Cold War between the East and the West contributed to the flourishing of dictatorial rulers who centralized state powers and resources and suppressed political opponents. Kenya was no exception: constitutional amendments handed president Kenyatta immense powers and regional autonomy demands fizzled out. Kenya, unlike many other states, escaped the curse of coup d’états that gripped the region at this time.[20] In 1982, the government sponsored a constitutional amendment to make the then ruling party KANU the only political party;[21] the provision was repealed in 1991, paving the way for the first multi-party elections in 1992. The collapse of the Berlin Wall inspired democratic reforms throughout the African continent. In Kenya, this was accompanied by comprehensive reforms that included decentralization reforms.

3.3. Constitutional reforms: the decline of autonomy demands and the rise of decentralization for developmental effectiveness

The end of the Cold War led to collapse or end of dictatorial regimes across the continent. In Kenya, external pressure from donor governments and local pressure from civil society, church groups, and opposition politicians led to a repeal of the constitutional provision that established Kenya as a one-party state and to other reforms that facilitated the first multi-party elections in 1992. KANU won the 1992 and 1997 elections, courtesy of an ethnically split political opposition and the use of state resources and largesse for campaigns, vote rigging, and other electoral malpractices.

Furthermore, the re-introduction of multi-party politics coincided with politically instigated ethnic violence during elections. Not surprisingly, electoral violence was at its most severe in the Rift Valley and Coast provinces, which were champions of regional autonomy (Akiwumi Commission 1999, 59–190 and 233–78). In both provinces, the violence was accompanied by calls for “outsider communities” to leave and return to their home regions (Ibid.). It soon became clear that multi-party elections alone were not adequate to ensure a proper transition to democracy and rule of law. Comprehensive constitutional reforms became the vehicle through which democracy, rule of law, and constitutionalism was advocated (Bannon 2007, 1832).

The demand for constitutionalism was accompanied by a popular clamor for decentralization of powers and resources. While local governments survived the onslaught by the KANU governments on regional autonomy arrangements, their operations were severely weakened by the KANU’s centralization policy (Southall and Wood 1996). Local service delivery was taken over by ineffective central government entities that were unaccountable to local communities (Ibid.). Administrative and security agencies became increasingly repressive and only accountable to the central government. This led to a popular call for devolution of powers and resources and replacement of central government administrative structures with locally elected governments (Oyugi 2005, 76).

The call for decentralization of powers was different from the regional autonomy demands at independence. The demands for decentralization had a strong element of developmental and service delivery effectiveness as opposed to autonomy demands on the basis of ethnic or regional identity. Calls for devolution of powers were accompanied by objectives of local accountability, participation in development, and effectiveness of local and regional governments. While there were political differences among national politicians and political parties on the manner of devolving powers, the situation was not comparable to the events that occurred at independence.

There were no more secessionist demands in the North Eastern and Coast regions. Furthermore, the Luo community, who had teamed up with the Kikuyu to call for a centralized system of government, were in the opposition. Thus, the calls for the decentralization of powers were fueled by a united disenchantment with the inefficiency of a centralized bureaucracy and a common dislike of a dictatorial regime. While there were hints of regional autonomy demands in the constitutional review process, there were no clear or coherent demands to establish regional autonomy in a manner similar to independence-era negotiations.

The constitutional review process can be divided into two phases: the first began in 1998, under the Constitution of Kenya Review Commission and ended with an unsuccessful constitutional referendum in 2005 when the government's proposed draft fundamental law was defeated. The second phase began in 2008, when constitutional reform was identified as a means of consolidating political stability, after the violence that followed the disputed presidential elections in late 2007 and early 2008. During both phases, there was universal support for the devolution of powers and resources. Minority communities who participated in the constitutional conferences were all in support for the devolution of powers and resources.  However, despite the uniform public support of devolution during both phases of constitutional review, no coherent or detailed structures (number of units, the kinds of powers to be devolved, arrangements at the national level, etc.) were put forth (Constitution of Kenya Review Commission 2005, 39). Furthermore, there were no specific calls for asymmetrical autonomy arrangements for the independence-era secessionist regions or any other community for that matter.

In the first phase, the national delegates conference proposed in the so-called “Bomas draft” four levels of government: national, regional, district, and locational.[22] The Bomas draft proposed lists of powers for the national government, 14 regional governments and 74 district governments.[23] It also proposed a bicameral parliament with the Senate representing regional interests.[24] However, the state’s attorney general Amos Wako’s final draft (commonly referred to as the “Wako Bill”), which was subjected to the 2005 national referendum, altered substantially the Bomas draft of the delegates. The Wako Bill abolished the Senate and removed regional and locational levels of government. It also gave the central government more sweeping powers over the devolved units than the Bomas draft had done and weakened the strong parliamentary system features and replaced them with a more powerful presidency.[25]

The two main political parties, at the time, were sharply divided on the system of government.  While the opposition favored a parliamentary system, the ruling coalition wanted a hybrid system with a stronger presidency. None of the parties wanted a pure presidential system since the overly powerful presidency had been stigmatized as a result of years of abuse of state power. In reality, the bid to frustrate the devolution proposal was the work of the Kikuyu elite who never wanted to share powers, through a devolved system of government, with other communities (Kanyinga 2009, 339). Suggestions to return to the independence-era system of regional governments were dismissed as promoting “ethnic Balkanization”. However, the ruling elite at the time of the constitutional review was unwilling to share powers, either through devolution, or through a strong parliamentary system of government where executive powers were to be shared between the president and a prime minister (possibly from an opposition party).

The second phase of the constitutional review process, which led to the current Constitution, was led by a Committee of Experts on Constitutional Review (CoE). Unlike its predecessor (i.e., the Constitution of Kenya Review Commission), the CoE had a limited mandate: isolating contentious issues only and facilitating consensus building on those issues in order to have a new Constitution. Among the issues that it identified were devolution and the system of government. The CoE prepared three drafts at key stages of the review process: the “harmonized draft Constitution” (17 November 2009), the “revised harmonized draft Constitution” (8 January 2010) and the “proposed Constitution” (23 February 2010). In the harmonized draft Constitution, the CoE provided for three levels of government: national government, eight regional governments (similar to the independence-era regions) and 74 county governments. However, in the proposed Constitution, the CoE removed the regional level and retained the county governments but reduced the latter to the current 47 county governments (Committee of Experts on Constitutional Review 2010, 90–91).

The reasons advanced for the changes in the number of government levels and number of administrative-territorial units are important for this discussion. The CoE decided to do away with the regional governments because the regions, as provided for in the harmonized draft Constitution, were not allocated any substantial powers and had no clear sources of revenue. The feedback the CoE received from public views and comments were that the regions would be an unnecessary economic burden (Ibid.). The CoE decided to reduce the number of counties in order to ensure that they are more effective and are better able to check the power of the Center (Ibid.). The negative or hostile attitude towards regions may be attributed to the fact that regional governments were associated with the secessionist and autonomy demands of the past. In Kenyan public discourse, the mention of federalism or a regional system of government is usually equated with the independence-era majimbo system. In turn, the majimbo system is usually associated with ethnic Balkanization and break-up of the state. This may be the very reason why any talk about establishing or maintaining regional administrative-territorial units was not seriously entertained. The then government, a grand-coalition executive that was formed between the ruling and opposition parties as a peaceful solution to the 2007–2008 violence, supported the proposed Constitution which was approved in the 2010 referendum with a 68.5% vote in favor.

One can conclude that the manner in which leaders of minority ethnic communities and secessionist regions were co-opted into the centralist philosophy of the KANU government served to erode the legitimacy and authenticity of regional autonomy demands. Even the design and implementation of the devolved system is focused more on development effectiveness than accommodation of regional autonomy claims. 

4. Legal Basis of Autonomy

Kenya’s sub-national autonomy arrangements are protected under the country’s 2010 Constitution. The principle of devolution of powers is recognized in the founding provisions and is one of the core constitutional principles underpinning the fundamental law. First, the Constitution provides that sovereignty, which emanates from the people, is exercised at the national and county level.[26] Devolution and sharing of powers is recognized as one of the national values and principles of governance.[27] Furthermore, the county boundaries are recognized as the state’s internal boundary divisions in the Constitution.[28]

The vertical division of state power between the national and county level is, however, only limited to legislative and executive powers; county governments do not have powers to establish and run courts or exercise judicial power. The devolved system is reified in the Constitution by allocation of powers to the county level.[29] The Fourth Schedule contains two separate lists of governmental functions, one for the national government and one for the county governments. While counties have legislative and executive powers over the areas allocated to the county level, there is no provision for county governments to enact their own constitutions, leaving the national Constitution as the sole primary means of protecting their autonomy.

The 2010 Constitution provides for two amendment procedures. The first is amendment by parliamentary initiative, which involves introduction of an amendment bill in any of the two houses of parliament followed by voting. A provision of the Constitution will be deemed amended if both houses support the amendments by a two-thirds majority.[30] The second is amendment through popular initiative signed by at least one million registered voters. The draft bill is then forwarded to each county assembly for a vote. If a majority of county assemblies approve the draft bill, it is passed on to parliament. The amendment via popular initiative only requires the support of a simple majority in both houses of parliament. If the legislative body fails to pass the draft bill, the proposed amendment is subjected to a national referendum.[31]

However, there is a special entrenchment of county government autonomy. The amendment of constitutional provisions that touch on “the objects, principles and structure of devolved government”, “national values and principles of governance” such as devolution of power, and “sovereignty of the people”, must be approved by a national referendum.[32]

It is thus clear that county autonomy enjoys a special protection under the Constitution. While county governments have a limited role in constitutional amendment (i.e., voting to approve amendments via popular initiative), the core provisions relating to county autonomy and devolution of powers are safeguarded by popular sovereignty via national referendum. Support for devolution remains high in the public and it is highly unlikely that any bill that seeks to weaken county autonomy could secure a positive result through a referendum.

It is worth noting that neither the 1963 international agreement signed by Kenya’s government and the sultan of Zanzibar regarding the 10-mile coastal strip, nor its content are explicitly featured in the county autonomy arrangements of the 2010 Constitution. The exclusion of the special autonomy arrangements in the Coast region from the constitutional provisions on autonomy was not a major issue during the constitutional review.  The centralist approach of the post-colonial Kenyan state has suppressed legitimate autonomy demands and this is reflected in the current county autonomy arrangements.

The objectives of devolved government include, inter alia, fostering “national unity by recognizing diversity”, recognizing “the right of communities to manage their own affairs and to further their development” and protecting and promoting “the interests and rights of minorities and marginalized communities”.[33] Yet, historical claims to self-rule have not been accepted. The Constitution contains symmetrical arrangements for developmental effectiveness as opposed to accommodating claims to self-determination and self-rule. 

Issues related to group autonomy have been regulated in the Bill of Rights included in the 2010 Constitution. For instance, the Islamic traditions of the coastal region find protection in the guarantee of the freedom of religion.[34] The Bill of Rights also protects  the rights of persons belonging to cultural and linguistic communities to enjoy their culture, use their language, as well as establish, join and maintain cultural and linguistic associations[35] Moreover, the Bill of Rights compels the state to put in place affirmative action measures to ensure that minorities and marginalized groups develop their cultural values, languages and practices, participate and are represented in governance and other spheres of life, and enjoy special opportunities in educational and economic fields.[36] 

The Constitution does not make reference to or single out any particular groups within the Kenyan society. There is no reference to groups or communities that have historical claims to territorial autonomy. Furthermore, generic terms such as “minorities” and “marginalized communities” have proved elusive and hard to pin down in practical contexts.

5. Autonomous Institutions

In Kenya, autonomy is based on a symmetric devolution of powers to the 47 county governments that exercise executive and legislative power. Accordingly, each county has its own executive body and assembly. The parliament’s upper house (Senate) is the formal link between the county and the national levels of government. The primary mandate of the Senate and county institutions is to safeguard and exercise autonomy.

5.1. County level institutions

The county assembly is composed of representatives elected through a first-past-the-post system from single member constituencies called wards, for a mandate of five years.  The number of wards per county varies from one county to the next.[37]  The 2010 Constitution provides that, in addition to the ward representatives, a county assembly comprises several special representatives, that is, a number of “special seat members” necessary to ensure that no more than two-thirds of the membership of the assembly are of the same gender, and to ensure representation of members of minorities and marginalized groups, including persons with disabilities and the youth. [38] These special seats are allocated on the basis of proportional representation by use of party lists (for details, see below).[39] The speaker of the county assembly is an ex officio member. [40]

A county governor heads the county executive committee. The county governor is directly elected by a simple majority of county voters. Upon assuming office, the county governor nominates the members of the county executive committee with the approval of the assembly, from among persons who are not members of the assembly.[41] Among the criteria that the nominees have to collectively meet is to reflect a county’s ethno-cultural diversity and gender balance.

It would appear that the county institutions are designed to ensure effective exercise of autonomy at the county level. However, the design of the county institutions raises several challenges which affect the implementation of these legal provisions. One could point to issues that range from the design of the electoral system and manner of choosing elected representatives to the criteria to ensure diversity in the composition of the legislative and executive bodies.

First, while the Constitution provides for the election of special representatives, the entire electoral system is based on a majoritarian system. County governors are elected through a first-past-the-post system where the candidate with the most votes, regardless of margin, is elected. Given that national and sub-national elections are widely determined by ethnic voting patterns, the fate of ethnic minorities within counties is almost permanently consigned to the ethnic majority. This is worse in cases where inter-county migration has created ethnic majorities that are not originally from the county. Indeed, in some counties (e.g., Nakuru, Laikipia, Trans-Nzoia) the governors come from the “migrant ethnic majorities” as the ethnic communities who are indigenous to the county are outnumbered.

Secondly, while it is possible to ensure territorial representation of minorities through creation of special wards or areas of special representation, the delimitation of ward boundaries overly relies on population threshold, as opposed to geographical concentration of ethnic and cultural minorities. Indeed, the general criteria for ward boundary delimitation give disproportionate weight to population-based factors, rather than to identity-related factors or consultation with communities.[42] The report of the commission that delimited ward boundaries ahead of the March 2013 general election stated, incorrectly in this author’s view, that “boundaries delimitation does not resolve the issues regarding representation of marginalized groups” (Independent Electoral and Boundaries Commission 2012, 27). The Commission called for other means (other than boundary delimitation) to accommodate minorities and marginalized groups. While it is difficult to enhance representation of minorities and marginalized groups, especially those who are dispersed within larger or dominant voter groups, it is possible to accommodate geographically concentrated groups by establishing electoral consistencies within their areas. 

The 2013 general election results confirmed the ineffectiveness of electoral mechanisms. There is paucity of ethnically disaggregated data per county that can assist to understand the performance of ethnic communities in each county. However, the aggregate national data can provide a general picture. The largest political parties in Kenya, which presumably have the support of the five largest ethnic communities in the country won a total of 1074 of the 1450 ward seats that were up for election (Nyabira and Ayele 2016,146). This represents 74 % of the total ward seats. While this is an aggregated result, it still shows the general trend that dominant parties and communities within counties secured a vast majority of county seats. Non-dominant political parties (and independent candidates) were left to share the remaining 26 % of county assembly ward seats.

While the Constitution seeks to cure this problem through the inclusion of “special seats” in the county assembly, the manner of determining who are the special representatives negates the purpose of the mechanism. Political parties are required to submit party lists before elections to fill the various categories of seats in the county assemblies. However, the seats allocated to special representatives (that take into consideration gender, minority background, disability, age, etc.) are filled on the basis of the party’s performance in the first-past-the-post elections. For example, if party A won 70 seats in a county with 100 wards, it will be awarded 70 % of all the seats available in all categories of nomination. While the Constitution refers to this system as proportional representation,[43] the mechanism produces a further disproportion in the county assemblies as the majority parties get to choose seats for minority groups.

In the March 2013 general election, 95 % of the people that were elected as ward representatives were men. This necessitated the topping up of “special seats” in the county assemblies with 770 women representatives, in order to ensure compliance with the two-thirds gender rule. However, the filling of the other categories, especially minorities or marginalized communities was complicated and confusing. The lack of precise criteria for identifying groups such as minorities within the counties proved a major challenge. For example, do persons from the Kikuyu community (i.e. the largest ethnic group nationally) qualify as minority in areas where they have migrated to? What about sub-clans or smaller groups within the known ethnic communities, do they also qualify as a marginalized/minority sub-group within the county’s majority ethnic group? These questions have no clear answers in the Constitution or the laws dealing with ethnic inclusion.

Thus, while the electoral rules provide that each county should nominate four representatives to cater for the categories of minorities/marginalized communities, persons with disabilities, and youth, the criteria that were applied across the counties differed and produced wildly different results from what was intended in the Constitution. Indeed, even where the nominations were challenged in courts of law, the courts lacked definitive criteria against which to evaluate the nominations. In the end, the dominant parties at the different levels ended up appointing their own supporters to the special seats in the county assembly without any objective criteria of ensuring that minorities or marginalized groups are actually represented. Due to lack of data on the ethnic background of elected representatives at the county level, it is impossible to ascertain the current ethnic composition of special representatives in the 47 county assemblies across the country.

With regard to the county governors, all the 47 governors, with the exception of the cosmopolitan Nairobi, are from the ethnic majority group of the respective county or one of the ethnic groups that voted together as a bloc in the county election. The process of ensuring that each county’s executive body reflects its ethnic and cultural diversity has also met similar challenges. There are no criteria through which persons can be identified as belonging to a particular community in a county. In some cases, the names of individuals have been used but this is not always accurate as there are shared names across ethnic communities. Furthermore, religious names (Christian and Islamic) are ethnically neutral and cannot be used to determine one’s ethnicity. There is no requirement in the law for one to declare one’s ethnicity for any official purposes, hence the lack of ethnic data at the county level and generally in the country.

The requirements for inclusiveness in the 2010 Constitution have prompted official initiatives to collect ethnic data to demand ethnic identity information. For instance, public service employees have now the option to state their ethnic identity in a headcount. However, no one is compelled to state his or her ethnic identity in the process. In 2016, the National Cohesion and Integration Commission published an “ethnic and diversity audit” of county governments. Based on research conducted in all counties of Kenya between 2014 and 2015, the report concluded that “employment in the county public service is not only inequitable but skewed towards the dominant groups in the county. Additionally, there are communities that are highly underrepresented yet others, such as the Dasenach Shangila, Leysan, and Galjeel are completely excluded from accessing these opportunities” (National Cohesion and Integration Commission 2016, 211).

However, not all is lost in terms of representation within autonomous county institutions. In each of the 47 county governments, the local majorities are communities that historically or currently desire some level of autonomy. By default, the elected representatives are from these communities. This is the case, for instance, with the governors and ward representatives elected in the counties established in the former North Eastern and Coast regions. While the current administrative-territorial structures do not reflect the claims for regional autonomy of these communities, there is some form of representation in the county institutions. The very fact that these communities have been split into smaller counties as opposed to the original regions at independence, weakens their collective autonomy demands. But the communities and their representatives can still achieve, in a limited way, some of their autonomy goals through the county institutions. 

5.2. National level institutions

Kenya has a bicameral parliament consisting of the National Assembly (lower house) and the Senate (upper house). The National Assembly represents “the people” and is composed of 290 members directly elected by the registered voters of single member constituencies, 47 women elected from the county constituencies, and 12 members nominated through party lists to cater for vaguely defined special interests (e.g. youth, persons with disabilities, workers).[44] As the National Assembly is a chamber of popular representation, the largest ethnic communities or areas with high population density have the most members. For instance, while Nairobi (which is the county with the highest population) has 18 members in the National Assembly, Lamu county (which is the least populated and the smallest geographically) has only 2 members. However, in the Senate, the two counties are represented by one senator each.

The Senate represents and safeguards county autonomy at the center and is composed of 47 senators elected by the registered voters of the counties, each county constituting a single member constituency and 20 senators nominated through party lists as follows: 16 women, two representing persons with disabilities (one woman and one man) and two representing the youth (also one woman and one man).[45] The manner of determining the 20 special seats in the Senate is similar to that of county assemblies. They are allocated to parties on the basis of mandates won through the first-past-the-post system of electing senators from the counties. The party with the largest number of elected senators is allocated the most seats across the categories of nomination, in the proportion of the seats won. 

There are special voting procedures in the Senate that reflect its county representation role. All senators who were registered as voters in a particular county constitute a “county delegation”. The head of the county delegation is the senator elected in the respective county. On matters that affect counties (whether laws or other issues), each county delegation cast one vote on behalf of the county. Accordingly, only 47 votes will determine the decision on a matter affecting counties. On all other matters, each senator has one individual vote.[46] This mechanism is put in place to ensure that the territorial representation is enhanced in decision-making in the Senate.

The Senate, thus, provides the appropriate forum for advocating and protecting regional autonomy. However, the Senate has structural design features that do not support its role as an institution to protect county autonomy. First, there is no special category provided for representation of territorial minorities within a county. Indeed, senators are elected at the county level in the same manner as governors. Accordingly, majority ethnic communities will, in most cases, win the governor and senator seats. In some cases, pre-election arrangements may be made to share senator and county governor seats between the two largest communities, but there is no guarantee that the voters will respect such local power-sharing arrangements at the ballot. One would expect that special representation of territorial minorities and marginalized communities within counties would be at the center of the design of the Senate. However, special representation is given to non-territorial categories such as youth, women, and persons with disabilities.

6. Autonomous Powers

County powers are guaranteed and protected by the Constitution, which provides that county governments can exercise legislative and executive powers over matters allocated them under the Fourth Schedule to the Constitution.[47] Kenya’s fundamental law recognizes, in principle, the three kinds of powers: exclusive, concurrent, and residual. The Fourth Schedule contains a list of national government functions and a list of county government functions. The Constitution provides, without providing a separate list of concurrent functions that any matter that appears in the two lists is a concurrent function that belongs to the two levels of government.[48] However, it is not easy to decipher which matters are concurrent as the language and wording used in the two lists to describe the respective functions differs. This means that matters that are concurrent between the two levels of government have to be determined through interpretation (Bosire 2015).

The Constitution further provides that any function that does not appear in the two lists in the Fourth Schedule is a residual function allocated to the national government.[49] However, the above scheme of division of functions between the two levels of government has proved to be extremely problematic in its implementation and has provided the most serious threat to county autonomy.

Firstly, while the Constitution stipulates that “[f]or greater certainty, parliament may legislate for the Republic on any matter”,[50] the nature and extent of this provision is not clear. There is a potential for such a provision to be abused by the national legislature.

Secondly, the ambiguity and total lack of clarity on the powers that are concurrent, exclusive or residual is affecting the effective exercise of county autonomy. On several occasions, the parliament has passed laws that seem to infringe on areas that seem to belong to county governments. Courts have, on several occasions, struck down such laws for being unconstitutional (Bosire and Wanjiru 2015). The Senate should be playing a role in ensuring that no national laws that infringe on county autonomy are adopted. However, political rivalry between the senators and governors has undermined the protective role that the former are supposed to perform in national legislative business. Indeed, on some occasions, the Senate itself has unleashed legislation meant to undermine county government powers.[51] The drafters of the Constitution foresaw the complexities and ambiguities in defining national and county functions and thus put in place institutions and mechanisms to ensure that functions are objectively divided.[52] However, for various reasons, they did not manage to reach a comprehensive framework for the division of functions between the national and county governments.

Thirdly, given the historical context of political and institutional centralization of resources and power, the ambiguity of the constitutional and legal framework regarding national and county governments seems to maintain the status quo, which favors the central government. Courts have demonstrated eagerness to protect the functional autonomy of county governments, however, judicial intervention is only limited to disputes that come before the courts for adjudication.

Fourthly, the powers that are most relevant to exerting autonomy, given Kenya’s historical context, are not allocated to counties. For instance, land is a politically, socially and economically sensitive issue. During the independence-era talks, the claims for regional autonomy were heavily laced with fears that communities from other regions would take up land belonging to indigenous communities. In some cases, regional autonomy was equated to expulsion of “settler communities” to their home regions. Accordingly, sub-national units wanted powers to determine and vary land policies including ownership and address historical injustices (Constitution of Kenya Review Commission. 2012, 27). However, the powers over land management are reserved for the national government and county governments’ role in land administration and management is peripheral. County government powers are limited to holding unregistered community lands in trust for communities. Administration of land is vested in the national government and the National Land Commission. Thus, county governments are not in a position to address claims with regard to land ownership. 

Lastly, the national government has powers to set and enforce national policies and standards. As a result, the national government can, under certain circumstances intervene in county governments (where a county is not performing its functions properly, or where a county is not maintaining a financial management system prescribed by national legislation[53]) or suspend county governments. As mentioned above, one of Senate’s functions is to check the exercise of such powers over counties.

7. Financial Arrangements

One of the principles of devolved government provides that counties should have adequate resources to perform their functions.[54] The Constitution puts in place various measures intended to guarantee the financial autonomy of county governments. These include setting aside some taxes and revenue raising powers for the counties and a guaranteed share for county governments from revenue collected nationally. The Constitution also provides that county governments can, subject to certain conditions, borrow loans to finance operations. However, the central government transfers are, for various reasons, the main and primary source of revenue for the county governments. Accordingly, the Constitution has attempted to put in place a transparent process of sharing resources between the national and county governments and among the counties. However, the national government retains the bulk of resources and the county governments generally do not have adequate resources to perform their functions and thus realize the potential of their constitutionally guaranteed autonomy.

While the county governments have taxing powers, the Constitution places major tax bases with the national government. Indeed, all but two taxes (property and entertainment taxes) belong to the national government. The Constitution provides that parliament may authorize county governments to charge any other tax. County governments, as well as the national government, are empowered to charge fees for the services that they provide.[55] The Constitution requires that no government should charge any tax or licensing fees except as provided by legislation. The Constitution does not specify whether this is national or county legislation, but the High Court ruled that county legislation is sufficient in the case of county taxes.[56]

While county governments have limited powers to tax, locally generated revenue is an insignificant source of revenue for most of the counties. Except for a handful of county governments with relatively large urban centers (such as Nairobi and Mombasa) or those rural counties that host national game reserves and tourist resorts, the local revenue generation is dismal. In 2013–2014, county governments targeted to collect 1.2 % of GDP as own-resource revenue but they end up collecting an average of only 0.5 % of GDP (World Bank 2014, 27). Reasons for such dismal performance include: low tax base for most counties that have vast rural areas, inadequate capacity to generate/collect revenue, revenue leaks due to mismanagement and corruption within the counties, and a lack of willingness to enhance revenue generation capacity among counties, which is facilitated by increasing dependence on central government transfers.

It is for the above reason that the county share from the national level becomes the main and primary source of county finance. The Constitution provides that at least 15 % of revenue collected nationally should be set aside for county governments.[57] However, early analyses showed that the 15 % was grossly inadequate to meet the functions that the county governments were expected to perform (World Bank 2012, 53–57). The Constitution sets up an independent Commission on Revenue Allocation to advise on sharing of revenue collected nationally and to build the fiscal capacity of county governments. Furthermore, the Senate participates in the vertical and horizontal division of revenue collected nationally.

Since March 2013, counties have received four annual allocations. In all allocations, the percentage of the county share has been well above the minimum 15 % that is guaranteed in the Constitution. In 2013–2014 when counties came into operation, they received 19.94 % of the revenue and this was increased to 24.18 % in 2014–2015 the next financial year (Kenya Institute of Public Policy Research and Analysis 2016, 181). This shows that there is an incremental trend in the county share from the revenue collected nationally. The recommendation of the Commission on Revenue Allocation for the county share has always been slightly higher than what is finally provided in the Division of Revenue Bill.

The county share is usually divided among the 47 county governments using criteria that take into account various factors including: population, geographical size, poverty index, fiscal responsibility, and a uniform share. Counties have full autonomy to plan, budget and utilize the county share. There is no national government restriction, e.g., in terms of block grants earmarked for specific sector spending to county governments or conditions on the spending of the county share. The only restriction is that withdrawals from the county revenue fund, where county disbursements are held, can only be done with the authority of the Controller of Budget, an office that is independent from the two levels of government and is mandated to monitor budget implementation of the national and county governments. Counties are also required to publish quarterly expenditure reports and have their accounts audited annually by the Office of the Auditor General.

While limited revenue autonomy seems to be well protected, there are a number of challenges in implementation. Firstly, the overall county share is generally inadequate to enable county governments to fulfil all their functions (Council of Governors 2016). As a result, the national government provides additional funding through conditional grants to the county governments. There has been a consistent growth in the number and total amount of conditional grants that are given to county governments (Office of the Auditor General 2016, 27–28). The risk with this trend is that the national level may end up undermining county autonomy through these grants.

Secondly, the confusion over the division of functions between the two levels of government seems to hurt county finances. There are instances where the national government seems to unjustifiably retain resources while substantial functions have been transferred to the counties. In the health sector, for instance, the Ministry of Health retains around 47% of the total national and county expenditure but over 90 % of health institutions have been transferred to the counties (Ministry of Health 2016). This problem is replicated across most of the sectors where there are shared functions between the two levels of government.

Thirdly, there is usually a delay in the disbursement of funds to the county governments. This disrupts service delivery by county governments and sometimes leads to crises such as strikes due to delayed payment of salaries. The overall effect is that the autonomy of county governments is undermined as a result of the management of national revenue by the national government.

Fourthly, the Constitution requires county governments to seek administrative authorization before borrowing and this has a great potential to undermine the autonomy of county governments from the center.[58]

Lastly, the county governments do not have resources to address the core issues that underlie regional autonomy claims. For instance, in the coastal region, the county governments do not have resources to support the restoration of land to communities in the region. The national government fully funded one such project, where households were settled on a farm after the national government purchased the land (Ngirachu 2016). Important coastal resources such as the Mombasa port, which is currently the gateway to the East African region, is controlled by the national government and revenue from the port is collected at the national level and redistributed nationally.

Economic self-determination, as a concept, has not been strongly advocated over the years. The lack of substantial mineral deposits, oil or other natural resources in the country, as is the case for instance in Botswana or Nigeria, may have contributed to this. The discovery of oil deposits in the northern part of the country, however, has led to commencement of debates around benefit sharing from locally extracted resources. County governments, as well as communities within mining areas, are demanding a fraction of the overall benefits from oil that will be extracted from the area.

The collective impact of the factors discussed above is to undermine and weaken the financial autonomy of the county governments.

8. Intergovernmental Relations

The Constitution provides broad principles for cooperation and consultation between the national government and the county governments and encourages them to carry out their respective functions on the basis of mutual respect to each other. In case of disputes, the Constitution requires the two levels of government to ensure an amicable settlement and encourages them to seek alternative dispute resolution mechanisms.[59] 

Enabling legislation provides for intergovernmental institutions to enhance cooperation. The Intergovernmental Relations Act of 2012 established the National and County Government Coordination Summit, the Council of County Governors (CoG) and the Intergovernmental Relations Technical Committee (IGRTC). The Summit is “the apex of intergovernmental relations” and is composed of the president and the 47 county governors.[60] It meets at least twice a year and its functions include, inter alia: evaluating the performance of national or county governments; monitoring the implementation of national and county development plans; coordinating and harmonizing the development of county and national governments policies; facilitating and coordinating the transfer of functions, power or competencies from and to either level of government.[61] 

The CoG brings together the 47 county governors and operates as a body for horizontal cooperation and consultation among the county governments. Whenever there are issues of common concern to the county governments, the CoG is the forum through which deliberations and consultation are carried out with the national government.

The IGRTC is a body composed of experts on intergovernmental relations, which acts as the secretariat of both the Summit and the CoG, facilitating their activities and implementing their decisions. While the IGRTC is expected to serve the two levels of governments, its operation is supposed to be supported by the national government. Furthermore, the appointment of the IGRTC’s members was spearheaded by the national ministry in charge of devolved governance instead of a joint process involving the two levels of government.

Few other bodies, some formed by statute and others by administrative fiat, act as intergovernmental consultation bodies for sectoral issues such as. health, education, security. The Public Finance Management Act of 2012 established the Intergovernmental Budget and Economic Council which brings together the finance minister, members of the county executive in charge of finance, and representatives of the Commission on Revenue Allocation. This body is chaired by the deputy president and its role is to provide a forum for intergovernmental fiscal relations and consultation.[62]

While the legal framework provides for principles, mechanisms and institutional structures to enhance intergovernmental relations, effectiveness has not been achieved for a number of reasons. First, the transition to devolved governance has slowed down due to the lack of commitment to settling matters such as the division of functions, restructuring of institutions, and transfer of resources. Intergovernmental relations provide avenues through which such issues can be ironed out. However, it appears that the forums are not used to address the actual problems facing implementation. On many occasions, the governors have complained about the lack of follow-up on decisions that are agreed at the Summit and other sectoral forums (Council of Governors 2016).

9. Inter-group Relations within the Autonomous Entities

In Kenya there are at least 43 major ethnic communities which are living dispersed around the 47 counties. Within a county there are indigenous communities and migrant communities concentrated in rural or urban areas. The legislative and executive bodies of a county must reflect the different communities and the cultural diversity of the respective county. The legal framework must prescribe mechanisms to protect minorities within counties.[63]

However, there are a number of institutional design factors that impact on inter-group relations. As discussed at section 5 above, the electoral system’s design favors local majorities and does not facilitate effective participation of smaller groups. Moreover, there are no precise criteria, in policy or law, for the identification and accommodation of distinct local groups. Group autonomy is largely suppressed by policy and is seen (often irrationally) as divisive. As a result, there is a kind of enforced homogeneity that tends to reinforce the dominance of the majority or politically dominant groups.

Due to an implicit suppression of local identities, there is a dearth of statistics and information for purposes of recognizing, preserving, and promoting cultural diversity at local level through official channels. For instance, it is almost impossible to get disaggregated data for different communities in multi-ethnic counties. Furthermore, the general institutional design stopped with the county governments. There are no specific constitutional arrangements, beyond the level of the counties; no consideration was given to sub-county identities and counties have a near unilateral power to design institutions below the county level.

In these circumstances, it is difficult to talk, with certainty, about formal intergroup relations at the county level. An exception, perhaps, are the pre-election power-sharing arrangements that happen in some counties in the North Eastern region. For instance, Mandera county is dominated by the Somali community, which is divided into clans and sub-clans. They have arrangements, facilitated by elders where seats are shared or rotated between clans. However, politicians who run for elective office do not always heed these arrangements; the current governor of Mandera county vied for the seat against the wishes of elders. In other counties, pre-election arrangements for the sharing of seats are usually fluid, uncertain, and heavily influenced by the interests of individual politicians from those areas (as opposed to legitimate group interests).

10. Membership, “Quasi-citizenship” and Special Rights

Section 10 is not applicable to the case of Kenya.

11. General Assessment and Outlook

The rise of popular resentment against centralized rule in the 1990s was not a direct result of suppression of regional autonomy claims. There were no strong claims for regional autonomy but a countrywide outcry over the inefficiency and ills of centralized authority and the personalization of state powers and resources.

Throughout the 1990s, the World Bank initiated several reform programmes to improve local service delivery in Kenya, but they did not bear much fruit (World Bank 2002). These decentralization reforms focused on developmental goals and service delivery effectiveness and did not respond to sub-national claims for autonomy. This point is validated by the fact that there was no strong opposition to the removal of the regional level of government during the constitutional review process. Yet, the provinces came closest to structures that would have been used to accommodate the independence-era claims to regional autonomy. Furthermore, the process through which the drafters settled on the current 47 counties reveals that they mainly considered the developmental outcomes as opposed to regional or local autonomy and identity demands. The establishment of county governments has had the effect of attracting a number of national politicians to the county level. Many of the county governors were national politicians who opted to vie for the position of governor. This very fact may have the effect of elevating autonomy issues to the national political scene. However, this will very much depend on how county governors collectively push for autonomy arrangements. However, it is also fair to comment that most county governors may end up being absorbed into the business of local development and local service delivery and neglect questions of autonomy.

Public and political consciousness of regional autonomy has significantly reduced over the years. Most Kenyans (including those from territories with strong historical autonomy claims) supported the removal of provinces when they overwhelmingly voted for the 2010 Constitution. The only exception is the Rift Valley where the majority voted against the Constitution, mainly because the leading political figures in the region opposed it. However, regional autonomy and secessionist sentiment re-emerged after the 2017 presidential election. In Kenya, presidential elections are usually hotly contested, pitting the larger ethnic communities against each other. In the August 2017 election, Uhuru Kenyatta, who led a coalition that drew support mainly from the Kikuyu and Kalenjin communities in Central and Rift Valley regions, was declared winner. The opposition, which drew support mainly from the Western, Coast, and Nyanza regions, challenged the election result in the Supreme Court. The Supreme Court cancelled the election citing irregularities and illegalities and ordered a fresh election. However, the main political opposition boycotted the second election on grounds that its demands regarding electoral reforms had not been met by the time the second election as convened. The second election was, thus, a one-sided affair where the incumbent was announced the winner.

The presidential election has ignited a debate on secession of regions and communities in the political opposition. A member of parliament from the opposition drafted the Constitution of Kenya (Amendment) Bill 2017 seeking the secession of 40 out of the 47 counties to form “Peoples Republic of Kenya” (Vidija 2017). The draft Bill is grounded on claims that successive governments have entrenched ethnic discrimination and marginalized certain communities and regions The Bill also alleges that instead of devolving powers and resources, the current government continues to centralize them.

The topic of secession is yet to gain traction in the mainstream Kenyan public discourse. The current context is not comparable to the independence-era secessionist movements. The latest secessionist claims may actually be seen as a knee-jerk reaction, springing from dissatisfaction with the 2017 presidential election result. However, it is important to note that secessionist talks were earlier confined to proscribed political groups with little involvement of mainstream politicians.

After the 2017 elections, the political opposition resolved to form “peoples’ assemblies” to act as forums for public consultation on the future of the country. More than 15 county assemblies in opposition friendly regions have adopted motions to establish peoples’ assemblies. Kenya’s attorney general has challenged at the High Court these motions arguing that the purported formation and exercise of powers by the peoples’ assemblies have no constitutional basis since there are established government institutions to exercise power on behalf of the people (Wambulwa 2018). In March 2018, the opposition suspended the peoples’ assemblies process following an agreement between the president Uhuru Kenyatta and the opposition leader Raila Odinga.

Bibliography

–      Akiwumi Commission. 1999. "Report of the Commission of Inquiry into Tribal Clashes in Kenya". Nairobi.

–      Anderson, David M. 2005. “Yours in the struggle for Majimbo. Nationalism and the party politics of decolonization in Kenya”. Journal of Contemporary History 40(3): 547–64.

–      Bannon, Alicia L. 2007. ”Designing a constitution-drafting process: Lessons from Kenya”. Yale Law Journal vol. 116, issue 8, 1824–72. Accessed June 20, 2019. https://digitalcommons.law.yale.edu/ylj/vol116/iss8/3.

–      Bosire, Conrad. 2015. “Powers and functions of county governments in Kenya”. In Kenyan–South African Dialogue on Devolution, edited by Nico Steytler and Yash Pal Ghai, 181–200. Cape Town: Juta.

–      Bosire, Conrad and Wanjiru Gikonyo, eds. 2015 Animating Devolution in Kenya: The Role of the Judiciary. Rome, Nairobi: International Development Law Organisation, Judiciary Training Institute, and Katiba Institute.

–      Brennan, James R. 2008. “Lowering the Sultan’s flag: Sovereignty and decolonization of coastal Kenya”. Comparative Studies in Society and History 50(4): 831–61.

–      Castagno, A. A. 1964. “The Somali–Kenya controversy: Implications for the future” Journal of Modern African Studies 2(2): 165–88.

–      Committee of Experts on Constitutional Review. 2010. “The Final Report of the Committee of Experts on Constitutional Review”. Nairobi. Accessed June 20, 2019. https://katibaculturalrights.files.wordpress.com/2016/04/coe_final_report-2.pdf.

–      Constitution of Kenya Review Commission. 2005. “Final Report of the Constitution of Kenya Review Commission”. Nairobi.

–      Constitution of Kenya Review Commission. 2012. “Verbatim report of constituency public hearings, Saboti Constituency”, held at Kitale Town Museum Hall.

–      Council of Governors. 2016. “Statutory Annual Report: 2015/2016”. Nairobi.

–      Ghai, Yash P. and J. P. W. B. McAuslan. 1970. Public Law and Political Change in Kenya. Oxford: Oxford University Press.

–      Independent Electoral and Boundaries Commission. 2012. “Revised Preliminary Report of the Proposed Boundaries of Constituencies and Wards” Vol. 1. Nairobi. Accessed June 20, 2019. https://www.iebc.or.ke/uploads/resources/WHXao7x83D.pdf.

–      Kanyinga, Karuti. 2009. “The legacy of the White Highlands: Land rights, ethnicity and the post-2007 elections violence in Kenya”. Journal of Contemporary African Studies 27(3): 325–44.

–      Kenya National Bureau of Statistics. 2009. “Kenya Population and Housing Census”. Nairobi.

–      Kenya Institute of Public Policy Research and Analysis. 2016. “Kenya Economic Report 2016. Fiscal Decentralization in Support of Devolution”, Nairobi: KIPPRA. Accessed June 20, 2019.  http://kippra.or.ke/kippra-publications-2/#7-kenya-economic-report.

–      Mboya, Tom. 1963. Freedom and After. London: MW Books.

–      Ministry of Health. 2016. “National and County Health Budget Analysis 2016/17”. Nairobi.

–      Mueller, Susanne D. 1984. ”Government and opposition in Kenya 1966–9”. The Journal of Modern African Studies 22(3): 399–427.

–      Muigai, Githu. 2001. "Constitutional Amendments and the Constitutional Amendment Process in Kenya (1964 – 1997)". Unpublished PhD Thesis. Nairobi: University of Nairobi.

–      National Cohesion and Integration Commission. 2016. “Ethnic and Diversity Audit of the County Public Service”, Vol. 1. Nairobi.  Accessed June 20, 2019. https://www.cohesion.or.ke/images/docs/Ethnic-and-Diversity-Audit-of-the-County-Public-Service-2016.pdf.

–      Neuberger, Benyamin. 1979. “Federalism in Africa: experience and prospects”. In Federalism and Political Integration, edited by J.D. Elazar, 171–88. Tel Aviv: Turtledove Publishing.

–      Ngirachu, John. 2016. ”Mini Budget seeks Sh1 billion for Waitiki land”. Daily Nation.

–      Nyabira, Ben Christopher and Zemelak Ayitenew Ayele. 2016. “The state of political inclusion of ethnic communities under Kenya’s devolved system”. Law, Democracy and Development. Vol. 20, 131–53.

–      Odinga, Oginga. 1976. Not yet Uhuru, Nairobi: East African Educational Publishers.

–      Office of the Auditor General. 2016. ”Report of the Working Group on the Socio-Economic Audit of the Constitution of Kenya 2010”. Nairobi.

–      Ogot, Bethwell A. and William R. Ochieng’. 1995. “The decisive years 1956–63”. In Decolonization and Independence in Kenya: 1940–93, edited by Bethwell A. Ogot and William R. Ochieng’, 24–79. Athens: Ohio University Press.

–      Okoth-Ogendo, H. W. O. 1972. “The Politics of Constitutional Change in Kenya since Independence, 1963–69”. African Affairs 71(282): 9–34.

–      Oyugi, Walter O. 1978, Local Government and Development in Kenya, Institute of Development Studies at the University of Sussex, 7–11.

–      Oyugi, Walter O. 2005. “The Search for an appropriate decentralisation design in Kenya: Historical and comparative perspectives”. In The Anatomy of Bomas: Selected Analyses of the 2004 Draft Constitution of Kenya, edited by Kithure Kindiki and Osogo Ambani, 57–106. Nairobi: Claripress.

–      Southall, Roger and Geoffrey Wood. 1996. “Local government and the return to multi-partyism in Kenya”. African Affairs 95(381): 501–27.

–      Vidija, Patrick. 2017. “MP Kaluma drafts secession Bill to form 'People's Republic of Kenya’”. The Star, 13 November 2017. Accessed June 20, 2019.  https://www.the-star.co.ke/news/2017-11-13-mp-kaluma-drafts-secession-bill-to-form-peoples-republic-of-kenya/.

–      Wambulwa, Annette. 2018. “Oath, Peoples’ Assembly suit on the same day”. The Star, 24 January 2018. Accessed June 20, 2019. https://www.the-star.co.ke/news/2018/01/24/oath-peoples-assembly-suit-on-same-day_c1702729.

–      World Bank. 2002. "Kenya: An assessment of local service delivery and local governments in Kenya". Washington, DC. Accessed June 20, 2019. http://hdl.handle.net/10986/15415.

–      World Bank. 2012. "Devolution without Disruption: Pathways to a Successful New Kenya". Nairobi. Accessed June 20, 2019. http://hdl.handle.net/10986/27157.

–      World Bank. 2014. “Decision time: Spend more or spend smart? Kenya Public Expenditure Review”, Vol. I. Nairobi. Accessed June 20, 2019. http://hdl.handle.net/10986/21507.

–      World Bank. 2016. “Republic of Kenya: Kenya Urbanization Review”. Washington, DC. Accessed June 20, 2019. http://hdl.handle.net/10986/23753.

–      World Bank. 2017. “Kenya Economic Update, Housing – Unavailable and Unaffordable”, No. 15. Washington, DC. Accessed June 20, 2019. http://hdl.handle.net/10986/26392.

–      Wunsch, James and Dele Olowu. 1990. The Failure of the Centralized State: Institutions and Self-Governance in Africa. Boulder: Westview Press.

–      United Nations. 2018. World Statistics Pocketbook – Kenya. New York. Accessed June 20, 2019. https://unstats.un.org/unsd/publications/pocketbook/files/world-stats-pocketbook-2018.pdf.

Case law

High Court of Kenya

Council of Governors & 3 others v Senate & 53 others [2015] eKLR

Cereal Growers Association & another v County Government of Narok & 10 others [2014] eKLR

List of abbreviations

CoG – Council of County Governors

IGRTC – Intergovernmental Relations Technical Committee

KANU – Kenya African National Union

KADU – Kenya African Democratic Union

NFD – Northern Frontier District

About the author

Conrad M. Bosire is an expert in the areas of law, governance and development, and human rights. He currently serves as the Chief of Staff in the Office of the Chief Justice in Kenya, and is the acting Executive Director of the National Council on the Administration of Justice (NCAJ), an umbrella body that brings together (state and non-state) justice sector institutions in Kenya. Dr. Bosire holds an LLM in Human Rights and Democratisation in Africa from the University of Pretoria (South Africa) and a PhD in Law from the University of the Western Cape (South Africa). His doctoral research analyzed Kenya's system of devolution and examined how it pursues development, unity, and limited (democratic) central government.

[1] Section 18 of the County Governments Act 2012, for instance, provides that county assemblies shall only use English and Kiswahili in the conduct of the formal assembly business.

[2] Article 8 of the 2010 Constitution of Kenya.

[3] Article 6 (1) of the 2010 Constitution of Kenya.

[4] Before the entry into force of the 2010 Constitution Kenya was divided into eight provinces: Central, Coast, Eastern, Nairobi, North Eastern, Nyanza, Rift Valley, Western.

[5] The Kikuyu (Central Province), Luo (Nyanza Province), Kamba (Eastern Province), Luhya (Western Province), and Kalenjin (Rift Valley Province).

[6] Article 96 (1) of the 2010 Constitution of Kenya.

[7] Article 96 (2) and Article 98 of the 2010 Constitution of Kenya.

[8] Article 91 (1) (a) and (2) (a) of the 2010 Constitution of Kenya.

[9] Under Article 138 (4) of the 2010 Constitution, the president must win at least 50 % of the votes cast and win 25 % of votes in at least half of the 47 county governments.

[10] Article 138 (5) of the 2010 Constitution of Kenya.

[11] Article 152 (2) of the 2010 Constitution of Kenya.

[12] Article 130 (2) of the 2010 Constitution of Kenya.

[13] Article 174 of the 2010 Constitution of Kenya.

[14] The boundaries of the districts were delineated in such a way as not to split up (so far as possible) any ethnic group between different districts.

[15] The newly-born Somali Republic aimed to create a “Greater Somalia” by uniting all the territories inhabited by a majority Somali population in the Horn of Africa. “Greater Somalia” would comprise Somalia and territories belonging to Kenya, Ethiopia and Djibouti.

[16] “Kenya Coastal Strip Agreement between the Government of the United Kingdom, His Highness the Sultan of Zanzibar, the Government of Kenya and the Government of Zanzibar”, October 1963.

[17] Section 91 of the 1963 Constitution of Kenya.

[18] Section 36 of the 1963 Constitution of Kenya.

[19] After abolition of the Senate, the government was able to push through its agenda of military action against the secessionist movement in the North Eastern region.

[20] There were two unsuccessful coup attempts in Kenya:  The first (lesser known coup) was in April 1971 and the second one was in August 1982.

[21] Section 2A of the 1963 Constitution of Kenya introduced by the Constitution of Kenya (Amendment) Act 7 of 1982.

[22] Section 6 of the Bomas Draft (2003).

[23] First Schedule and Fourth Schedule of the Bomas Draft (2003).

[24] Sections 120 and 122 of the Bomas Draft (2003).

[25] While the Bomas draft provided, for instance, that the prime minister would be the leader of the majority party in parliament, the Wako Bill changed this to provide that the president was to appoint a prime minister from any of the members of parliament.

[26] Article 1 (4) of the 2010 Constitution of Kenya.

[27] Article 10 (2) (a) of the 2010 Constitution of Kenya.

[28] Article 6 (1) and First Schedule of the 2010 Constitution of Kenya.

[29] Fourth Schedule of the 2010 Constitution of Kenya.

[30] Article 256 (1) of the 2010 Constitution of Kenya.

[31] Article 257 of the 2010 Constitution of Kenya.

[32] Article 255 (1) of the 2010 Constitution of Kenya.

[33] Article 174 of the 2010 Constitution of Kenya.

[34] Article 32 of the 2010 Constitution of Kenya.

[35] Article 44 of the 2010 Constitution of Kenya.

[36] Article 56 of the 2010 Constitution of Kenya.

[37] Nairobi County (the largest) has 85 wards while Lamu County (the smallest) has 10 wards. The 47 counties of Kenya are divided into 1450 wards nation-wide.

[38] Article 177 (1) (b) and (c) of the 2010 Constitution of Kenya.

[39] Article 90 (1) of the 2010 Constitution of Kenya.

[40] Article 178 of the 2010 Constitution of Kenya.

[41] Article 179 (2) of the 2010 Constitution of Kenya.

[42] Article 89 (6) and (7) Constitution of Kenya 2010.

[43] Article 90 (1) of the 2010 Constitution of Kenya.

[44] Articles 95 (1) and 97 (1) of the 2010 Constitution of Kenya.

[45] Articles 96 (1) and 98 (1) of the 2010 Constitution of Kenya.

[46] Article 123 of the 2010 Constitution of Kenya.

[47] Articles 183 and 185 of the 2010 Constitution of Kenya.

[48] Article 186 (2) of the 2010 Constitution of Kenya.

[49] Article 186 (3) of the 2010 Constitution of Kenya.

[50] Article 186 (4) of the 2010 Constitution of Kenya.

[51] See, for instance, the judgement of Kenya’s High Court in Council of Governors & 3 others v Senate & 53 others [2015] eKLR. The Court found the County Governments (Amendment) Act 2014 that introduced the County Development Board as unconstitutional because this body, although chaired by the county’s senator and composed of members of the national legislature and national executive, were intended to perform county functions and to be funded from county funds.

[52] The Transition to Devolved Government Act 2012 establishes the Transition Authority to manage the transfer of functions and transition to county governments.

[53] Article 190 (3) (a) and (b) of the 2010 Constitution of Kenya.

[54] Article 175 (b) of the 2010 Constitution of Kenya.

[55] Article 209 (3) and (4) of the 2010 Constitution of Kenya.

[56] In Cereal Growers Association & another v County Government of Narok & 10 others [2014] eKLR, the High Court of Kenya held that a county government must have enacted a Finance Act in order to charge agricultural produce cess.

[57] Article 203 (2) of the 2010 Constitution of Kenya.

[58] Article 212 of the 2010 Constitution of Kenya.

[59] Articles 6 (2) and 189 of the 2010 Constitution of Kenya.

[60] Sections 7, 11 and 19 of the Intergovernmental Relations Act of 2012.

[61] Sections 8 and 9 of the Intergovernmental Relations Act of 2012

[62] Section 187 of the Public Finance Management Act of 2012.

[63] Article 197 (2) of the 2010 Constitution of Kenya and Section 3 (g) and (h) of the County Governments Act of 2012.