Tourism-conservation enterprises for community livelihoods and biodiversity conservation in Kenya

Abstract

Tourism-conservation enterprises (TCEs), such as eco-lodges, are a relatively new strategy of the African Wildlife Foundation for enhancing community livelihoods and wildlife conservation in wildlife-rich areas outside state-protected areas in sub-Saharan Africa. This article investigates the extent to which TCEs succeed in meeting these objectives by focusing on two enterprises in Kenya: the Sanctuary at Ol Lentille and the Koija Starbeds. Empirical data were gathered between October 2010 and March 2013 through semi-structured interviews, focus group discussions, literature review and document analysis. Both TCEs demonstrated significant contributions to employment, education, healthcare and security. Compared with the Starbeds, the Sanctuary realised a much larger conservation area and more pronounced biodiversity recovery. The analysis showed that the contribution of TCEs to livelihoods and biodiversity conservation depends on the nature of the partnership arrangement, as well as the local, national and international contexts in which they operate.

Keywords:

• tourism
• biodiversity
• conservation enterprises
• livelihoods
• partnerships

1. Introduction

Until the 1970s, wildlife conservation efforts in eastern and southern Africa mainly focused on state-protected areas (Adams, 2004). In the 1970s and early 1980s, this ‘top-down’ approach began to be viewed as ineffective and unsustainable in realising conservation objectives (Gibson & Marks, 1995; Brown, 2002). Increased poaching, decline of wildlife populations and local people's hostility towards protected areas formed serious conservation challenges giving rise to the ‘community conservation paradigm’, which hinges on the premise that empowering local communities will enable them to become protectors of their natural resources (Western & Wright, 1994; Van der Duim, 2011; Kieti et al., 2013).

Interventions were consequently initiated to address the goals of conservation and development simultaneously (Western & Wright, 1994; Hackel, 1999; Brown, 2002; Adams, 2004; Igoe, 2006; Mahanty et al., 2006). In line with this discourse, community-based tourism enterprises, such as eco-lodges and cultural villages, were set-up in wildlife-rich areas outside protected areas to generate revenues as incentives for conservation (Saarinen, 2010; Butcher, 2011). From the 1990s onwards, communities also established joint ventures with private operators to contribute to development needs through revenue-sharing schemes. The conservation enterprises of the African Wildlife Foundation (AWF), defined as commercial activities that ‘generate economic benefits in a way that supports the attainment of a conservation objective’ (Elliott & Sumba, 2010:4), are part of this partnership approach.

In this article, we focus on tourism-conservation enterprises (TCEs) such as a lodge or tented camp. In these TCEs, a partnership is created between communities (land-owners) and a private investor. The AWF operates as a third and ‘neutral’ partner to promote a ‘fair deal’ between communities and private investors, secure donor funding and monitor the outcomes (Sumba et al., 2007; Nthiga et al., 2011; Ahebwa et al., 2012; Lamers et al., 2014; Van Wijk et al., 2014). Community land-owners (i.e. group ranches⁵) own all of the immovable assets (land for conservation and the lodge). A private investor owns the movable assets and operates the lodge on behalf of the community. In a trust, managed by a board in which all three partners are represented, the community benefits generated from the lodge are managed and distributed (see Figure 1).

Figure 1: Organisational chart of TCEs

Although a number of studies have investigated the outcomes of such partnership arrangements (Kiss, 2004; Spenceley, 2005; Manyara & Jones, 2007; Sumba et al., 2007; Mitchell & Muckosy, 2008; Snyman, 2012), there is still limited understanding of how the particularities of the institutional arrangements contribute to local livelihoods. This is important because recent research in East Africa has shown that failure to deliver sustainable livelihoods might lead to discord, unrealised conservation objectives and the collapse of the project (Southgate, 2006; Ahebwa & Van der Duim, 2013; Lamers et al., 2013). To address this gap, we analyse two TCEs located in Laikipia Country, Kenya: the Sanctuary at Ol Lentille and the Koija Starbeds (see Figure 2).

Figure 2: Map of the study area

2. Conceptual framework

To analyse the contribution of the two TCEs to livelihoods and conservation, this article focuses on the five capital assets of the Sustainable Livelihoods Framework (SLF): human, physical, social, financial and natural capitals (Scoones, 1998; DFID, 1999; Ashley & Hussein, 2000; Ellis, 2000; De Haan & Zoomers, 2005). The SLF (see Figure 3) consists of five main aspects (Scoones, 1998; Ashley & Hussein, 2000): livelihood capital assets, livelihood strategies, livelihood outcomes, institutional arrangements (e.g. policies and institutions) and the vulnerability context.

Figure 3: The Sustainable Livelihood Framework

The livelihoods concept has been widely used in the analysis of development interventions. It was conceptually inspired by Chambers & Conway (1992), who defined a livelihood as the means of gaining a living. Consequently, development organisations have applied the concept to unravel and evaluate the outcomes of poverty reduction and development interventions (Farrington et al., 1999; Allison & Ellis, 2001; Scoones, 2009). Although the livelihoods concept has been critiqued as being normative and analytical (Scoones, 2009), it has been widely applied for being people-centred, holistic, recognising multiple influences, outcomes, strategies and actors, and bridging micro-level and macro-level development (DFID, 1999; Solesbury, 2003; Scoones, 2009; Simpson, 2009).

Both TCEs were established with the aim to diversify community livelihood strategies by extending the ways in which communities make a living (pastoralism), improving community well-being and reducing poverty (outcomes), and decreasing community vulnerability to external factors (climate change, drought). Diversification of livelihoods and reducing poverty and vulnerability are supported by strengthening the livelihood building blocks; that is, livelihood capitals or assets.

Human capital consists of knowledge, skills and education that support livelihoods. Social capital relates to community pride and identity, and formal and informal relationships, including networks and partnerships. Buildings, health and education infrastructure, water supply machinery, or crops and livestock supporting livelihoods constitute the physical capital, and the employment opportunities, income and benefits that are available to individuals and communities comprise financial assets. Finally, the TCEs aimed at changing the relation between the way people make a living and the natural environment, and biodiversity conservation (see also Ahebwa & Van der Duim, 2013). Owing to methodological difficulties of establishing the impacts of project interventions on biodiversity (Kangwana, 2001), this article assesses this contribution of the TCEs on the natural capital assets, and biodiversity conservation, by looking at their impacts on biodiversity threats, including human–wildlife conflicts, land-use changes, land degradation and poaching (Salafsky & Margoluis, 1999; Kangwana, 2001).

This article examines the influence of the institutional arrangements on the capital assets in relation to the local, national and international contexts, and power-relations. At the local level, contextual factors such as the dynamics in the group ranch, culture and ecological conditions are examined. National-level policies that influence the partnerships are also included.

3. Methodology

The research employed a case-study design (Yin, 2003, 2009), with a clear subject (TCEs) and object of study (livelihood enhancement) resulting in exemplary knowledge (Flyvbjerg, 2006). Primary data were collected through semi-structured interviews and focus group discussions, complemented with secondary data, such as academic literature and project documentation. Between October 2010 and March 2013, the AWF offices in Nairobi and Nanyuki were visited. The group ranches directly involved in the TCEs (i.e. Kijabe, Koija and Nkiloriti) were each visited on multiple occasions, covering approximately 12 weeks in total. Access to project documentation (i.e. agreements, trust-deeds and reports) was granted and collected at the AWF headquarters in Nairobi as well as in the field.

The choice of interviewees was guided by the prominence of their role and involvement in the group ranch, the TCEs and spin-off projects. Regional AWF staff facilitated the initial contacts and access to the communities and private investors. Community interviewees mainly consisted of adult male respondents, selected through snowball sampling by the first author. In total, 46 individual semi-structured interviews, ranging between 30 minutes and two hours, were conducted with Kijabe, Koija and Nkiloriti group ranch leaders and members, the directors of the private investor companies and key AWF staff. Over time, some respondents were interviewed more than once. Further, five focus group discussions comprising of women and youths, were conducted on the respective group ranches, assisted by a Maasai translator. The number of participants ranged from 15 to 20. A representative sample was ensured to capture the different views of the three main stakeholder groups, as well as a fair representation of the interests and views within the community.

The interview and focus group discussion protocols consisted of general questions and discussion points relating to the establishment of the TCEs, decision-making processes, generation and distribution of income and governance challenges. Interviews and focus group discussions were audio-recorded, which made it possible for the interviewers to concentrate on the verbal and non-verbal aspects of the information exchange while obtaining a detailed account. Recordings were transcribed verbatim, coded and analysed in light of the SLF. The storyline and argumentation that emerged from this analysis was triangulated and enriched with the secondary data. Anonymity was guaranteed for the interviewees; therefore, a referencing system⁶ is used to relate information and quotes in the article to the respondents.

4. Introducing the two tourism-conservation enterprises

4.1 Sanctuary at Ol Lentille

The ‘Sanctuary’ is a 16-bed eco-lodge with four private, fully serviced houses, situated in the Kijabe group ranch, Laikipia County, Kenya. Each of the houses is architecturally designed to exhibit different themes, each having a private butler, valet, guide and chef. Other lodge facilities and activities include a swimming pool, spa, library, wildlife-viewing deck, game drives, horse riding, mountain biking, quad-bikes and community visits.

The Kijabe group ranch, with a total area of 6070 hectares, has set aside 2000 hectares as a conservation area (A-10; C-SOL-1). Lodge construction was started in 2003 by the AWF through funding from the United States Agency for International Development (USAID) and the Dutch Embassy, Nairobi, amounting to US$100 000 (P-SOL-1; P-SOL-2). The private investor, Regenesis Limited (RL), became involved in 2005 and together with the AWF and the community received US$400 000 from the European Union and Government of Kenya Tourism Trust Fund (A-10; P-SOL-2). RL spent a further US$1.1 million on furnishings, interior design and other movable assets (P-SOL-2), while the group ranch members contributed around US$14 000, raw materials and labour to the project (C-SOL-10).

In 2005, the Kijabe group ranch signed a 25-year management agreement with RL to manage the lodge and conservation area at a fixed fee per year, as well as a bed-night and conservation fee for every guest at the lodge. The fixed fee paid, referred to as the ‘rights fee’, is rent paid by the private investor to conduct business in their lodge and live on their land (C-SOL-1). The lodge opened for business in January 2007 (P-SOL-2). After RL had signed the contract with the Kijabe group ranch, the neighbouring Nkiloriti group ranch claimed that part of the conservation area belonged to them. A Memorandum of Understanding was signed between the two group ranches stating that Nkiloriti was to get 20% of the community share of bed-night and conservation fee if it would add a further 600 hectares to the conservation area, which it did (A-10; P-SOL-5). A ‘cultural manyatta’⁷ for women and a Financial Services Organisation (FSO), a micro-credit and savings organisation, have been set up as spin-off projects for boosting the community economic status.

The main actors in the partnership are the Kijabe group ranch, RL and the AWF. Decision-making occurs in three main platforms:

• the Kijabe Conservation Trust (KCT-SOL) or the ‘trust’;

• the group ranch; and

• the Ol Lentille Trust, Kenya (OLTK).

KCT-SOL is a charitable entity designed as a decision-making body for the operations of the conservation area, lodge maintenance and distribution of benefits. KCT-SOL is also mandated with the role of attracting donor and investment funding for the Kijabe community, a role it has not managed to execute. KCT-SOL is governed by an eight-member board of trustees, comprising three members from RL, three from Kijabe group ranch and one from the AWF and a chairperson from the community. The Nkiloriti group ranch is not represented. The group ranch is managed by a committee of 10 members, elected annually. Lastly, the OLTK is a Kenyan registered organisation affiliated to the private investor, which manages the Ol Lentille conservancy, and attracts donor, social and conservation funds for community and conservation benefits.

4.2 Koija Starbeds

The ‘Starbeds’ is an eight-bed eco-lodge on three platforms located at the Koija group ranch, Laikipia County, Kenya. The wooden platforms are partially covered by a thatched roof and contain a ‘Starbed’ that can be wheeled out onto the open deck for a night under the stars (Lamers et al., 2013). This design is a replication of the Starbeds^® concept from Loisaba Wilderness ranch (P-KS-2). Additional lodge facilities are a small lounge area for coffee, tea, lunches, dinner and relaxation.

The lodge idea was conceived in the late 1990s by the directors of the private investor company, which had leased the neighbouring Loisaba Wilderness ranch. They approached the AWF and proposed to replicate the Starbeds concept in partnership with the community as a conservation incentive (Sumba et al., 2007). The AWF was at the time implementing USAID's Conservation of Resources through Enterprises programme, which had a strong focus on enterprise development (A-3; A-4). The initial investment of US$48 000 for the lodge construction was a grant from the USAID's Conservation of Resources through Enterprises programme (Sumba et al., 2007; P-KS-2). The AWF's facilitation costs for community mobilisation, capacity-building and securing of property rights were estimated at US$30 000 (Sumba et al., 2007). The private investors, Oryx Limited (OL), provided the working capital (e.g. debtors, stock, creditors) in return for management rights (A-4). The community provided land for constructing the lodge and labour, dry wood and stones, as well as goats that were sold for US$2000 to obtain the land title-deed (C-KS-1; C-KS-2; C-KS-3; C-KS4). The cultural manyatta for women has been set up as spin-off project for livelihood diversification.

Similar to the Sanctuary, the main actors in the partnership include the Koija group ranch, OL and the AWF. At the core of the partnership is the Koija Conservation Trust (KCT-KS), governed by a five-member board of trustees; two from the community, two from OL and one from the AWF (Deed of Discharge and Appointment, 2000). On the board of trustees, the AWF holds the position of the chairperson, OL is the treasurer and the community is the secretary. The group ranch has a nine-member committee entrusted with decision-making and management. In addition, the private investors run the Loisaba Community Conservation Trust (LCCT), a charitable trust engaged in education, health and conservation issues. Table 1 compares the partnership arrangements of the two TCEs. The three main platforms are as follows:

• the Koija Conservation Trust (KCT-KS) or the ‘trust’;

• the group ranch; and

• the LCCT.

Table 1: Comparison of the arrangements

Aspect	Sanctuary at Ol Lentille	Koija Starbeds
Initial capital investment	>US$2 million (USAID/DGIS grant, TTF/EU, private investor)	US$48 000 (USAID grant)
Occupancy rate	<30%	<20%
Operations	Single business	Managed with the Loisaba Wilderness tourism enterprises, which allows economies of scale
Average annual community income from the partnership	US$20 000 to 30 000 (bed-night, conservation and lease fees)	US$20 000 to 30 000 (bed-night and conservation fees)
Bed-night and conservation fees	Paid as an extra amount from the non-resident rate (US$700)	Paid as an extra amount from the non-resident rate (US$395)
Division of bed-night and conservation fees	US$80 for all	US$85 for non-citizens
	• KCT-SOL (US$30)	• Entire amount to KCT-KS
	• RL (US$20)
	• Kijabe (US$24)
	• Nkiloriti (US$6)
Conservation area	2 000 hectares + 600 hectares from Nkiloriti group ranch	200 hectares
Average annual philanthropy benefits	>US$ 250 000 on education, health and conservation	US$20 000 to 30 000 on education, health and conservation
Main benefit-sharing platform	Group ranch	KCT-KS

DGIS, Directorate General for International Cooperation; TTF/EU, Government of Kenya Tourism Trust Fund/European Union.

5. Contributions to livelihoods and conservation

In this section we present and compare the contributions of the two TCEs to the livelihood assets and conservation. The contributions of the TCEs to livelihoods are further divided into the direct and expected outcomes of the TCEs and spin-off projects, and the indirect and unexpected outcomes.

5.1 Physical capital

5.1.1 Buildings

An essential characteristic of the AWF's TCEs is that lodge buildings and immovable assets belong to the communities (P-KS-2; P-SOL-3). Moreover, the AWF facilitated the construction of offices for the three group ranches and communication-related infrastructure, such as solar panels and radio communication equipment for health emergencies and security in the ecosystem. Women at Kijabe and Nkiloriti have bought iron sheets for some members to construct more modern houses, and for other members bought beehives for their beekeeping projects with income from the cultural manyatta (C-SOL-5). Challenges related to management and record-keeping, however, confront the women initiatives due to low literacy levels (C-SOL-5). The benefits are also limited since the cultural manyatta visits and bead-sales depend on the guests to the Sanctuary (C-SOL-6), which operates on an average of less than 30% occupancy (P-SOL-2).

Indirectly, the TCEs have led to an expansion of trading centres, such as Koija, Munishoi and Kimanjo (C-SOL-9; P-KS-2). One respondent who has witnessed the area transform argues: ‘if you looked at Koija ten years ago, there was no town, there was only one shop, and there were two buildings in that whole village’ (P-KS-2). Currently, the Koija group ranch hosts multiple shops, offices, a primary school and health centre, as well as two other lodges.

5.1.2 Water

At Kijabe and Nkiloriti, there has been drilling of boreholes, laying of water pipes, and maintenance and construction of dams, mainly funded through philanthropy of the OLTK and guest donations (P-SOL-3; C-SOL-2). The OLTK has also provided water tanks for schools, the group ranch office and a dispensary for rainwater harvesting (P-SOL-3). At Koija, the money received by the Koija group ranch management from KCT-KS is partly spent for maintenance of water pipes, as well as fuel and maintenance of the power generator that supplies water for the school and community (C-KS-4).

5.1.3 Health

The OLTK has constructed a 24-bed hospital, with an X-ray room, a maternity unit, a youth centre and doctor's quarters, at the Kimanjo trading centre, handed over to the Government of Kenya in September 2013. Prior to the construction, the trust supported the Kimanjo health centre with equipment, bicycles and a motorbike for community health workers (C-SOL-14). At Koija, the LCCT donated a refrigerator among other facilities for the health centre (C-KS-1).

5.1.4 Education

The OLTK has constructed buildings and provided desks, chairs, books, laptops and printers in over six primary schools and one secondary school in the area (C-SOL-1; P-SOL-3). The secondary school has been constructed by the OLTK and the infrastructure by January 2013 included four classrooms, a kitchen, a store and two latrine blocks, and plans were underway to construct an administration wing, science laboratory, teachers’ housing and a library. At Koija, the main beneficiary is the Ewaso Primary School, where LCCT has constructed classrooms, a dining hall and dormitories, and provided books and reading materials (P-KS-2; C-KS-1).

5.2 Social capital

The TCEs have contributed positively to the communities’ pride, belonging and identity. In the initial stages, the AWF facilitated the acquisition of land title deeds. Both areas had been designated as group ranches in the 1970s but did not have the legal documents to prove ownership (A-4; C-KS-1). As one leader at Koija pointed out:

we can credit the existence of Koija as a group ranch to the Starbeds … now we can say that we own the land, we have the title deed … before we just used to live here and maybe someone could have come to claim ownership … thanks to the Starbeds we are the real legal owners of the land. (C-KS-4)

The private investor at the Sanctuary supports the local football club and a sports exchange programme for primary school pupils (C-SOL-2). The lodge also provides transport to school for pupils to functions, such as music festivals and other extra-curricular activities (C-SOL-2; C-SOL-9). Koija, Kijabe and Nkiloriti communities have also been able to organise themselves in various committees, women and Moran⁸ groups (C-KS-3). At Koija, the LCCT aims at improving the community's social fabric through the purchase of a television set, video recorder and a power generator for youth education, guidance and counselling on issues such as HIV/AIDS (C-KS-1). Moreover, the roles of and respect for women in the community are also gradually improving as a result of linkages with other communities, as well as from their improved financial status (C-KS-4).

Negative social impacts include conflicts and power struggles, which have increased with the influx of money into the communities from the TCEs. For example, before the lodge was established in Kijabe there was harmonious co-existence with the neighbouring Nkiloriti (C-SOL-12). The construction of the lodge and consequent benefits has created some tension between the two group ranches, which have now been formally settled by a Memorandum of Understanding. Internal conflicts and power struggles have also been witnessed at Kijabe between those living within and those living outside the group ranch (C-SOL-1). These conflicts are associated with leadership, and associated benefits from the lodge. In addition, interactions with other communities as a result of increased commercial activities have contributed to cultural changes and reduced adherence to traditions, especially by the young generation (C-KS-5).

5.3 Human capital

5.3.1 Education

The Sanctuary provides meals, bursaries and payment of teachers’ salaries. At Kijabe, 40% of the group ranch's income from the partnership finances bursaries and teachers’ salaries, as stipulated in the Community Investment Plan. However, its distribution is hampered by a non-updated group ranch membership register with two-thirds of the members living elsewhere. The money is also shared among all members for secondary school, colleges and universities, and thus each receives a small amount. The decision-making power of the elected leaders also makes it difficult for the members to claim accountability and transparency. In addition, the RL bursary programme in 2011 paid school fees for 12 teacher-training college students, five university students and 12 secondary school students (P-SOL-2). By 2010, over 380 children had received bursaries through the partnership (P-SOL-1). Moreover, by January 2013 the secondary school student population had grown to 140 students.

At Koija, the largest percentage of income to KCT-KS is used for bursaries for secondary school, professional colleges and university. By 2007, a total of US$42 096 had been given to the community for bursaries (Sumba et al., 2007). Other available statistics indicate that in the second term of 2010, a total of US$5775 was used to pay bursaries for 58 secondary school students and two university students. In 2011, a total of US$10 462 was spent on bursaries for 43 secondary school students, one student in a professional college and two at the university (C-KS-1). LCCT also provides books, meals, bursaries and teachers’ salaries. As at November 2011, the LCCT had educated five students from Koija through secondary school up to university (P-KS-2).

While there are no available statistics to quantify the extent to which the benefits to education have influenced the illiteracy rates, one leader at Koija argued that the existence of the Starbeds has contributed to a reduction in illiteracy levels and changed the community attitude towards education: ‘The quality of education has improved, we now have around five graduates, many with diplomas and many in secondary school, people are now taking children to school since the lodge was opened’ (C-KS-4). However, the majority of the children at the group ranch are not in school for reasons such as livestock rearing, joining the Moran group at the cultural manyatta and early marriages (C-KS-1).

5.3.2 Health

At Kijabe, the OLTK is engaged in the provision of a food programme for HIV/AIDS patients on Anti-Retrovirals (C-SOL-14; Trustees Annual Report, 2011). The trust also paid salaries for 34 community health workers, whose duties include support for the community health programme at the grassroots level. The OLTK also provides mobile clinic services to the community, and an average of 10 mobile sessions reaching on average 400 patients were conducted in 2011 (Trustees Annual Report, 2011).

At Koija, the private investor supports the family planning project at the health centre, which has received a setback as a result of minimal acceptance due to culture and traditions (P-KS-2). The investor also provides transport for emergency treatment (C-KS-1). At Kijabe, a portion of the group ranch income from the partnership is used to pay hospital bills for those admitted to hospitals (C-SOL-1). However, there is an absence of proper guidelines, rules and accountability mechanisms for the use of the money, creating a loophole for misappropriation (C-SOL-1). Despite the development of healthcare facilities, health problems are still severe as community members continue to rely on traditional forms of medicine. Illiteracy further restricts the follow-up of medical staff instructions. Moreover, poor nutrition as a result of poverty and drought influence health (C-SOL-14).

5.3.3 Skills development

At the initial stages of both partnerships, the women's groups were enrolled in a capacity-building programme aimed at improving their skills in bead-making, savings and money management (C-SOL-5; A-4). AWF sponsored familiarisation tours for community leaders to other regions to learn about TCEs (A-4). At Kijabe, AWF also sponsored a tour for over 30 community members to learn about micro-finance and personal financial management (C-SOL-6). There has also been an awakening of the communities to an entrepreneurial culture; for example, a Koija community member is constructing a lodge.

5.4 Financial capital

The AWF claims that between 2006 and 2012 the total community income from the Sanctuary (e.g. lease fee, conservation and bed-night fees and wage benefits from direct employment) exceeded US$330 000. For Koija, the AWF estimated one-half of this amount (A-11).

5.4.1 Employment

The direct and indirect employment opportunities are summarised in Table 2. The Sanctuary employs a total of 21 permanent employees, eight of which are from Kijabe (C-SOL-13). In addition, through the OLTK, RL pays salaries for 34 community health-workers, one nurse, 22 schoolteachers and 29 scouts. The FSO also provides employment for two persons. The money received by the group ranch from the KCT-SOL also contributes to employment. Part of the group ranch income is used to pay three primary schoolteachers, two nursery schoolteachers and eight scouts (C-SOL-11). The lodge also purchases goat meat from the community, which amounts to over US$125, milk for US$94 and eggs for US$38 per month (P-SOL-2).

Table 2: Direct and indirect employment opportunities

Place/type of employment	Sanctuary at Ol Lentille	Koija Starbeds
Lodge	21: paid by RL	Five: paid by Oryx Limited
	• eight are from Kijabe	• All from Koija
Health centre	35: paid by the Ol Lentille Trust	Two: paid by LCCT
	• 34 community health workers and one nurse	• one nurse, one assistant
Schools	27: both primary and nursery school teachers	Three: two paid by LCCT
	• five paid by Kijabe group ranch
	• 22 by Ol Lentille Trust
Security + administration	37: security scouts for the conservation area	Five: paid by Koija group ranch, part income from KCT-KS
	• eight paid by Kijabe group ranch
	• 29 by KCT-SOL
FSO (village bank) To change to a SACCO	Two: salaries paid by the AWF	–
Others	–	>40 Koija community members employed at Loisaba Wilderness

SACCO, Savings and Credit Cooperative.

The Starbeds has five employees: one chef, two waiters and two room stewards (C-KS-2). Other employment opportunities created include one nurse and two teachers paid for by OL, one cultural manyatta supervisor and three scouts, one radio operator, one educational secretary, one cleaner for the health centre and one watchman, paid for by the group ranch with funds from the KCT-KS (C-KS-4). In addition, out of an estimated 170 employees of the Loisaba Wilderness, around 40 are from Koija (P-KS-2).

5.4.2 Other individual income

At Koija, Morans earn about US$250 per person per year from the dances. From bead sales, women can buy about 10 goats or one cow per year. Around 160 women are members of the cultural manyatta, and due to this large number each takes home around US$25 per year from entry fees. These low levels of income from the cultural manyatta can also be attributed to the limited number of guests visiting the Starbeds (C-KS-1). At the Sanctuary, the manyatta provides income for around 300 women from Kijabe and Nkiloriti, from entry fees (US$38) and beads sales. Entry fees are shared among the Morans and the women (C-SOL-5). Numerous challenges face the management of the cultural manyatta, including low literacy rates, limited visitor numbers, uncertain market conditions and poor record-keeping (C-SOL-5).

5.4.3 Other benefits

At Koija the private investor provides pasture and interbreeding for the community livestock during periods of drought, at a fee of US$4 per month for every cow (P-KS-2). However, this arrangement has been critiqued by the community for lacking proper organisation and information (C-KS-3).

The FSO became fully operational in April 2011 through the assistance of the AWF (C-SOL-6). It has a membership of over 1000 shareholders from different group ranches in the region (C-SOL-6) and operates as a bank and a micro-finance organisation. The FSO provides saving opportunities, with charges of US$0.12 per withdrawal, and loans and credit services, with interest rates from 1.5 to 3% per annum (C-SOL-5). From an initial capital of US$575 in 2008, this had grown to US$112 500 in August 2012. By that time, it had disbursed over US$93 023 in loans to shareholders who are eligible to get a loan of four times the amount of shares and earnings each year. In February 2012, the AWF gave a grant to the FSO of US$10 625 to support savings and loans (C-SOL-6). The FSO, however, faces several challenges. Only 10% of the shareholders are educated, which inhibits most shareholders from embracing financial opportunities, as well as understanding the financial details (C-SOL-6).

Generally, ‘illiteracy levels have gone down, people have added livestock, women can now make and sell beads, warriors used to engage in livestock theft but have now stopped, we have identity and peace and (medical) treatment is now available’ (C-KS-3).

5.5 Natural capital and biodiversity conservation

The existence of the TCEs has also had both positive and negative effects on the natural capital. Before the TCEs were established, land-use changes and degradation were identified as threats to biodiversity in the region. The AWF therefore assisted in zoning the group ranches into conservation, settlement and grazing zones, and initiated soil erosion control initiatives (A-2). For conservation, Koija has only set aside the initial 200 hectares (A-2). Kijabe set aside an initial 2000 hectares for conservation, with Nkiloriti adding 600 hectares to the conservation area to share in the benefits. Over time, the Tiamamut group ranch and Narupa Conservancy each added 2225 hectares, and by 2011 the area under conservation at the Sanctuary had grown to over 7050 hectares (P-SOL-2). These exclusive conservation areas have served as important wildlife habitats where vegetation recovery has been reported (A-10). Additionally, there has been an increase in numbers and diversity of both resident and migratory mammals, such as elephants, zebras and wild dogs (A-10).

However, there are incidences of non-commitment by the communities in terms of not sustaining soil erosion strategies or respecting the land-zoning rules (P-SOL-2; P-KS-2; A-10). Incidences of poaching by some community members have also been reported (P-KS-2). The dispute resolution mechanisms and sanctions tend to be weakly enforced by the Maasai communities, who often do not want to bring a community member to the police or to legal courts (see also Lamers et al., 2013). In addition, lack of property rights over wildlife on their land also serves as a disincentive to protect wildlife, particularly since communities receive minimal assistance and recognition as critical players in wildlife management in national policies and legislations (Van Wijk et al., 2014).

In terms of human–wildlife conflicts, the communities are reported to have become more tolerant of wildlife (C-KS-1; P-KS-2; A-2), as evidenced by the lack of retaliatory attacks on wildlife after loss of their livestock to wild dogs and other predators (AWF, 2008): Moreover,

there is a sense of security for the dispersal of wildlife to community land … they have a more positive attitude towards wildlife … and also the caring for that environment … they do not cut trees … they do not burn charcoal … the biggest disappointment is that they have not been able to expand the conservation area. (P-KS-2)

The tolerance of the communities can be partly attributed to the livelihood benefits from the eco-lodges and other related projects.

On the other hand, increased income from the TCEs has had negative impacts on biodiversity and natural capital. For example, income from bead sales and manyatta entry fees has been used for buying livestock, which has increased overgrazing levels and hence land degradation (A-10). Others have used their income to buy seeds for subsistence farming on the riverbank at Koija (C-KS-5), leading to increased land-use changes and human–wildlife conflicts.

6. Discussion and conclusions

This article discussed the livelihood and conservation impacts of two TCEs initiated by the AWF. The analysis was based on the SLF, with a focus on the livelihood capital assets and impacts on biodiversity threats.

The foundation of the AWF's TCEs clearly responds to the need to protect wildlife outside national parks, a task that has clearly been beyond the capacity and mandate of the Kenya Wildlife Service (see Van Wijk et al., 2014). As we have shown, the resulting experimentation with TCEs to address the challenges of conservation and development outside national parks and reserves in Kenya has had important, but also mixed, impacts on livelihoods and conservation.

In terms of livelihoods, our findings illustrate that the two eco-lodges have clearly stimulated local employment, income, education and health provisioning, which have steered community development. Moreover, the TCEs have leveraged a significant amount of indirect funding for additional development projects. Our findings also indicate that the majority of the benefits are not direct from the TCEs, but are indirect effects, resulting from philanthropy, grants and donations, among others, from the guests staying at the eco-lodges. Moreover, the direct benefits, although substantial, are highly influenced by the low occupancy rates, among other factors. This raises the question of the viability of tourism as a consistent and durable strategy for livelihood enhancement and conservation.

In terms of conservation, the main contribution of the partnerships has been the setting aside of exclusive conservation areas by the communities. However, the conservation of biodiversity is pressured by illegal grazing in the conservation areas, increased crop farming and growing livestock populations, which arise out of, among other factors, the ‘commons’ nature of the land (see Ostrom, 1990). Moreover, several fundamental ecological questions remain, including whether the conservation areas are large enough to support the different wildlife populations, and whether the areas are connected well enough to other conservation areas by corridors. Another major concern is the durability of such community interventions into the future, given the current increases in human population and the corresponding increases in livestock populations in the region.

These mixed contributions of the TCEs to livelihoods and conservation can be understood in relation to the local particularities of these institutional arrangements. The mixing of group ranch, trust and governmental rules and regulations, power-relations within and between group ranches and their members, and related governance issues have all influenced the varied impacts of the TCEs. The surge of partnerships, such as the two TCEs, reflect wider trends in which conservation organisations around the world increasingly adopt market-based approaches to the protection of wildlife by partnering with businesses in nature-based tourism activities (see for example Visseren-Hamakers, 2013). Most of the studies so far have focused on the effectiveness and desirability of these partnerships for nature protection, communities and businesses, while the longer-term consequences for governance remain understudied. Research in this area is timely and relevant, as the trend to work through multi-actor partnership arrangements continues, and recent evidence suggests that, in many of these partnerships, political struggles and governance challenges seem to increase rather than decline, particularly in Eastern Africa (e.g. Ahebwa et al., 2012; Lamers et al., 2013).

This article has shown that the SLF enables a detailed, thorough analysis and explanation of the relative contribution of a particular policy instrument, like the partnerships studied. By focusing the analysis on livelihood capital assets, the framework enables the unravelling of the different specific types of contributions. It also allows the analysis of the specific mechanisms through which the contributions are made, with changes in assets enabling the changing of livelihood strategies over time. Since the partnerships studied here are rather young, a separate analysis in terms of mechanisms has not yet been included. But for older initiatives it would certainly add value, as it could also shed light on the evolution of the contributions over time, and the durability of the solutions found. The inclusion of the contextual factors allows for explaining these various contributions.

In conclusion, based on our findings and in view of the limited available options to address conservation and development in the region, we argue that tourism-related arrangements such as the TCEs studied are promising. However, to be successful they need to be built upon, and better linked with, other conservation and development interventions since tourism alone will not be sufficient. Moreover, while the TCEs contribute to diversifying livelihood strategies, they can also create new dependencies of the local communities. In addition, the community governance structures should be enhanced if the intended benefits are to be meaningful to the community. The durability of the interventions should also be considered through viewing them as long-term interventions as opposed to projects. Thus, while important steps have been made through the partnerships until today, especially in terms of physical, financial and human assets, perhaps future efforts should be focused especially on social and natural assets, in order to assure the durability of the solutions created by the TCEs.

Notes

⁵A group ranch is communal land that has been demarcated and legally allocated to a group, such as a tribe, clan, section, family or other group of persons, under the Land Adjudication Act cap 284 Laws of Kenya.

⁶Interviews with community leaders for the Sanctuary (C-SOL-1 to C-SOL-14), with the Sanctuary's private investor (P-SOL-1 to P-SOL-5), with community leaders for the Starbeds (C-KS-1 to C-KS-13), with the Starbed's private investor (P-KS-1 and P-KS-2) and with AWF officials (A-1 to A-11); and focus group discussions (FGD-1 to FGD-5).

⁷A manyatta is a Maasai settlement (kraal); a cultural manyatta is used to denote a purposely built Maasai-style visitor centre located near tourist resorts or attractions.

⁸Morans are groups of young male Maasai warriors that benefit from the TCEs by dancing in the cultural manyattas.