How the unbanked bank themselves in rural western Kenya : understanding small-scale farmers' and producers' decision-making in using microloans for income-generating activities

(1)

(2)

Cover photo: Members of Bujibo Community-Based Organization carrying out table banking in Bukiri, Samia District, Kenya. (By N.Lehmann, July 2014).

(3)

Master’s Thesis

How the Unbanked Bank Themselves In Rural Western Kenya:

Understanding small-scale farmers’ and producers’ decision-making

in using microloans for income-generating activities

In collaboration with:

Name: Nadia Lehmann Student number: 10635432

E-Mail: nadia.lehmann@student.uva.nl

Course: MSc International Development Studies 2013-2014 Supervisor: Dr. Nicky R.M. Pouw

Second Reader: Dr. Courtney Lake Vegelin Date: 15 December 2014

(4)

Supervisor

Dr. Nicky R.M. Pouw

Assistant Professor, International Development Studies Graduate School of Social Sciences

University of Amsterdam Nieuwe Achtergracht 166 1018 WV Amsterdam The Netherlands Phone: +31(0)20-5254105 E-mail: n.r.m.pouw@uva.nl http://home.medewerker.uva.nl/n.r.m.pouw/

Second Reader

Dr. Courtney Lake Vegelin

Assistant Professor, International Development Studies Graduate School of Social Sciences

University of Amsterdam Nieuwe Achtergracht 166 1018 WV Amsterdam The Netherlands Phone: +31(0) 20-5255033 E-mail: c.l.vegelin@uva.nl http://home.medewerker.uva.nl/c.l.vegelin/

Local Supervisor

Dr. Hannington Odame Executive Director

Centre for African Bio-Entrepreneurship Family Health Plaza Off-Langata Road P.O. Box 1285 Nairobi

Kenya

Phone: +254 (0) 20600040 Mobile: + 254 (0)724226893 E-mail: hsodame@gmail.com

(5)

Abstract

This thesis examines the internal and external factors influencing smallholder farmers’ and producers’ decision-making in using microloans for income-generating activities in Samia District, Western Kenya. Drawing on a gendered livelihoods analysis it particularly focuses on how women use microcredit and how gender relations influence women’s loan use choices. The study analyzes the particular case of table banking groups, an upcoming informal microfinance scheme in Kenya. Qualitative and quantitative data was collected through a combination of focus group discussions, semi-structured interviews, a survey and participatory exercises such as priority ranking.

The study reveals that a combination of inter-related factors such as personal priorities, perceptions of risk, socio-economic background, table banking groups’ capacities and structural factors can either constrain or enable microloan use for income generation. Overall, it was found that women face various gender constraints in using microloans to improve their economic situation and that informal microfinance can have an ambiguous effect on intra-household gender relations.

Key words: rural livelihoods, table banking, informal microfinance groups, gender, female farmers,

(6)

Chapter 1. Introduction

Problem statement and background

Since its introduction as a poverty reduction and economic empowerment tool in the 1970s, modern microfinance has become a popular and widely used approach by development actors around the globe. Targeting in particular women and the rural poor as they are often excluded from the formal banking sector due to low and volatile incomes, high banking fees, low education level and lack of collateral (World Bank 2013), it aims at providing microloans, savings and insurances to small-scale entrepreneurs. It is based on the idea that microfinance is used to invest in income generating activities (IGAs) or expands an already existing business, which ideally improves people’s livelihood outcomes (Rosenberg 2010). In a rural context, on which this thesis focuses, this can mean intensification of already existing livelihoods such as agriculture or engagement in non-farm livelihood diversifications strategies (Carletto et al., 2007).

However, empirical studies assessing the impact of microloans suggest mixed results. Authors such as Zinman and Karlan (2009) and Banerjee et al. (2009) find in case studies that microloans trigger business investments. Other studies such as by Zaman (2004) and Dupas and Robinson (2009) assert that microloans mainly contribute to consumption smoothing and people’s ability to cope with shocks and stresses. Hence, investment in income sources that would lead to increasing people’s livelihood outcomes on a longer-term basis is not as often pursued (Rosenberg, 2010). A gap in these academic debates is that such impact studies rarely examine how loanees actually use microloans and what influences their decision-making (Morduch, 2000). As argued in this thesis, this is a strong deficit given that understanding use of microloans is a necessary intermediate step before assessing impact. In addition, informal microfinance groups, which have been essential in shaping formal microfinance schemes, are not as well researched as formal microfinance (Schreiner, 2001). Understanding of (indigenous) informal microfinance schemes is crucial given that they offer a valuable source of credit to people who cannot access formal microfinance and are practiced around the globe.

One form of informal microfinance is table banking, which is an upcoming development tool in Kenya that aims at targeting the (rural female) poor, who have often been excluded from accessing formal sources of credit, which constrains their economic development (Mwangi & Ouma, 2012). Table banking attempts to overcome this through a group-based funding strategy in which group members contribute monthly savings and take loans from the group’s allocation. Through interest rate payments on the loans group capital accumulates over time, whereby the capital is owned by its members through their shares. The idea behind table banking is that members use these microloans to invest in their livelihood activities. The Centre for African Bio-Entrepreneurship (CABE), a Kenyan NGO working on rural development related projects, introduced a table banking project in Western Kenya’s Samia District in 2011 (CABE, 2012), which this thesis deals with. The majority of members in CABE’s table banking groups are female small-scale farmers and producers, who face, like many

(12)

4

Kenyan women, various gender inequalities at intra-household as well as political and socio-economic level (Mwobobia, 2012). In accordance with the literature on microfinance, investments into IGAs have not taken off as expected in CABE’s project, which this thesis tries to understand. Hence, by attempting to contribute to filling the above identified gaps in microfinance literature and paying particular attention to gender, this thesis aims to answer the following main research question:

What factors influence rural small-scale farmers’ and producers’ decision-making in using table banking loans for rural income generating activities and how does gender influence this process? Outline of study

The theoretical framework of this thesis, presented in the next chapter, builds on the sustainable livelihoods approach (SLA) in tandem with a relational perspective as well as literature on microfinance, IGAs and gender. After an introduction to the research context is provided in the third chapter, the research methodology and methods are described in the fourth chapter. This thesis employed a mixed-methods research design by drawing on qualitative and quantitative data collected through in-depth semi-structured interviews, focus group discussions (FGDs), a survey, participatory exercises, observations and a community workshop. The data analysis is presented in Chapter 5, 6 and 7: firstly, an overview over CABE’s table banking group is provided, which is followed by an analysis of how members use table banking loans – whether they invest them in IGAs, in social purposes such as food, health or school fees, or a mix of economic and social activities. Chapter 7 then turns to the heart of this thesis by describing how factors such as people’s priorities, risk, socio-economic background, group performance and structural constraints impact members’ loan use decision-making process. Finally, Chapter 8 answers the main research question by presenting the main findings. A reflection on theory and methodology as well as recommendations for further research and policy conclude this thesis.

Purpose

The nature of this thesis is exploratory and explanatory at the same time as it focuses on gaining a more nuanced and comprehensive understanding of people’s decision-making regarding taking up IGAs. By integrating subjective elements such as people’s perceptions and priorities regarding loan use this thesis attempts to stay close to the perspective of the people this thesis is about, which can make policy design more effective (McGregor, 2004). The purpose of studying this particular case and this topic is two-fold: As this research aims at delivering concrete input in the form of analysis and policy recommendation to CABE’s current development project, it has an instrumental and practical relevance. The theoretical contribution of the thesis lies in its strength to contribute knowledge on how informal microfinance activities work for women and how people use microfinance and what influences their decision-making. As table banking has never been conceptualized in academic literature, this thesis also attempts to provide an academic introduction into the topic.

(13)

5

Chapter 2. Theoretical Framework

This chapter presents the theoretical framework of this thesis. Grounded in the SLA and enhanced with a relational approach, the chapter entails the main concepts this study builds on: microfinance, rural income-generating activities (RIGAs) and gender. The relational approach is used in order to better understand the complex structural context in which table banking members’ decision-making takes place. A post-structuralist perspective is introduced to briefly outline some of the critiques of microfinance. Drawing on the main concepts and the SLA, factors which can enable or constrain table banking loan use for IGAs are presented. Given that table banking has not been researched and is not widely known outside of Kenya, this chapter also briefly introduces table banking in the Kenyan informal microfinance context. The conceptual scheme of Chapter 4.3 indicates the relationships between the main concepts.

2.1 The Sustainable Livelihoods Approach

2.1.2 Origins, purpose and framework of the Sustainable Livelihoods Approach

In the context of three decades of evolving thinking about poverty and development the SLA rose to prominence in development theory and practice in the 1990s (Ashley & Carney, 1999). Firstly introduced by the Brundtland Commission on Environment and Development, the 1992 United Nations Conference on Environment and Development expanded the concept and advocated for sustainable livelihoods as a poverty eradication goal (Scoones, 2009). After Chambers and Conway (1991) had started off the debate around the livelihoods approach various authors such as Scoones (1998, 2009), Moser (1998), Ashley and Carney (1999), Bebbington (1999) and Ellis (2003) contributed to the theoretical evolution of the concept. International development NGOs, research institutes and donors quickly took up and adapted the SLA when it was introduced into policy-making (Krantz, 2001).

Inspired by Amartya Sen’s (2000) classic capability approach the SLA aspires to understand the complexity of poverty and its reduction from a people-centered viewpoint that goes beyond defining poverty as a lack of material assets (Chambers, 2006; Scoones, 2009). It departs from a local perspective on poverty and focuses on human capabilities while it recognizes that the micro-level perspective of how people construct their livelihoods is linked to the institutional and policy level they are part of (Farrington et al., 1999). The SLA seeks to comprehend the strategies and aspirations of the poor, the reasons behind their decisions and how development interventions can “reinforce the positive aspects of these strategies and mitigate against constraints” (IDS, 2011, n.p.). As such the framework has an analytical as well as practical purpose guiding development efforts and is employable at various levels such as the household, community, region or country (Farrington et al., 1999). Scoones (1998), who proposed a modified version of Chamber’s and Conway’s original sustainable livelihoods description, defines sustainable rural livelihoods the following (cited in Krantz, 2001, p.1):

(14)

6

“A livelihood comprises the capabilities, assets (including both material and social resources) and activities required for a means of living. A livelihood is sustainable when it can cope with and recover from stresses and shocks, maintain or enhance its capabilities and assets, while not undermining the natural resource base.”

As expressed in Figure 1.1 below, the sustainable rural livelihoods framework consists of the livelihood context, resources, strategies, outcomes and institutional processes (Scoones, 1998). The theoretical relations indicated in the framework articulates that people use a combination of resources or assets in order to pursue multiple activities so as to achieve certain livelihood outcomes (Farrington et al., 1999). In order to understand how people construct their livelihoods one must not only analyze their livelihood resources and strategies but also look at how institutional processes and organizational structures are inter-related to these elements (Krantz, 2001).

Figure 1.1: Sustainable and rural livelihood, a framework for analysis (Scoones, 1998)

According to DFID (1999) and similar to Figure 1.1 livelihood resources consist of: 1) human capital (skills, knowledge, labor, health); 2) natural capital (natural resources and services); 3) financial capital (savings, credit, remittances); 4) physical capital (infrastructure, producer goods) and 5) social capital (social support networks, group membership, relations of exchange, trust and reciprocity). Depending on the livelihoods context and institutional processes people can employ various

(15)

7

livelihoods strategies such as agricultural intensification/extensification (such as more output per unit area through labor inputs/more land under cultivation), livelihood diversification (such as off-farm income earning activities) or migration (temporary or permanent) which lead to livelihood outcomes and possible trade-offs at the well-being and sustainability level. Having a large portfolio of these assets increases one’s chance to follow specific strategies to overcome poverty (Thorbecke, 2005). However, not all people have equal access to these capitals and can transform them into strategies to pursue their own priorities, goals and aspirations. As DfID’s (1999) operationalization of the SLA2 demonstrates poor people’s lives are also influenced by a vulnerability context encompassing shocks, trends, risks and seasonality. People can employ coping strategies as a response to risks and vulnerabilities if they combine various assets, which determines their resilience to external stresses (Moser & Satterthwaite, 2010). As Krishna (2004) illustrates selling assets such as crops and cattle, changing consumption patterns and pulling children out of school, helps poor people to cope in times of crisis. Adaptive strategies such as diversifying income sources and building up assets are households’ responses to long term trends, cycles and adverse events (Ellis, 2000).

2.1.2 Critique of the Sustainable Livelihoods Approach: gender and relational approach

Gender

The early SLA and its manifold elaborations can be critiqued on a number of grounds. Important input to elaborate on power relations in livelihood analysis came from various disciplines such as political ecology, gender studies and studies on political arenas (Haan, 2012). Given the context of this thesis the failure to incorporate gender in the earlier SLA is of particular importance. This is largely attributed to the fact that Chamber’s and Conway’s (1992) contribution to the SLA never explicitly had a gender dimension (Beall, 2002). Since the framework regards households rather than individual members of households as unit of analysis, it disregards the role gender and its underlying different roles, responsibilities and power relations between women and men (Danida, 2008) plays in accessing and distributing assets within households (Odebode, 2004). Even though the more recent livelihood frameworks of major aid agencies include gender analysis nowadays, they still lack a comprehensive gender-sensitivity as Krantz (2001) states, “It is one thing to ensure that gender is being addressed in principle, however, and another to make it possible for women to express their genuine perceptions, interests, and needs in relation to specific livelihood issues in practice” (p.23). This has (often) been recognized in the more recent theoretical elaborations of the SLA due to influences from gender and development debates yet at the practical level gender appears to be “lost in translation” (Beall, 2002; Krantz, 2001). Beall (2002) stresses that it is misleading to refer to the absence of gender in the SLA

2

In this research the environmental context of the SLA, which is a crucial part of the sustainability notion of the approach, is not considered as I focus on the institutional context regarding income generating opportunities.

(16)

8

as a ‘missing idea’, which could be overcome by bringing in ‘additional tools’ as Ashley and Carney (1999) state. Rather, according to Beall (2002) gender needs to be intrinsic to the livelihoods perspective itself.

Neglect of structural issues in livelihoods analysis – need for a relational approach

Localized and agency-centered approaches such as the SLA are difficult to generalize and tend to exclude inequalities (Moser & Norton, 2001). Scoones (2009) asserts that the structure orientation that the approach offers is undermined when applied in development practice through which it becomes mainstreamed engendering a “black-boxing of structural issue” (p.186). Its strong agency perspective may offer a complex understanding of what people are capable of doing themselves to escape poverty. However, given that access to resources, their distribution and unequal power relations are highly political problems this leads to a depoliticization of development as the underlying structures that create poverty remain unaddressed (De Haan, 2012).

An evolution of the SLA that incorporates structure-agency relationships is the Strategic-Relational Livelihoods Approach as outlined by van Dijk (2009). It builds on a relational approach which, according to Mosse (2010) “views persistent poverty as the consequence of historically developed economic and political relations (…) and that rejects the individualism of neo-liberal rational choice models by emphasizing the effect of social categorization (…) in reproducing inequality” (p. 1157). Proponents of the relational approach such as Harris (2001, 2007), Harriss-White (2006) and Mosse (2010) argue that poverty research focuses too much on describing characterizations of poverty instead of explaining the structural political-economic and social causes that create and perpetuate poverty. In that regard Harriss-White (2006) asserts that “poverty is continually being created and recreated under the institution of capitalism” (p. 1241).

Thus, while remaining within the SLA approach, the theoretical perspective of this thesis is enhanced with a relational approach. It hence pays attention to the larger structures and socio-political processes in which table banking members navigate their choices about loan use. Given that gender is regarded as a social construction and that gender relations are largely shaped by power and inequalities and are hence tightly linked to structural issues a relational approach proves to be particularly valuable.

2.2 Microfinance, gender and table banking

2.2.1 Microfinance and rural income generating activities

Microfinance

The modern microfinance (MF) concept is strongly associated with the Bangladeshi Grameen Bank and its founder Muhammad Yunus who pioneered microfinance (MF) in 1976. In 2011 there were 2.5bn unbanked people in the world (World Bank, 2013). In rural areas only 10 percent of communities have access to formal financial services, which is a constraint for poor people to acquire

(17)

9

financial capital (IFAD 2013a). According to Christen et al. (2004) MF is a way of bridging the financial access deficit of the poor as it is supposed to contribute to “a world in which as many poor and near-poor households as possible have permanent access to an appropriate range of high quality financial services, including not just credit but also savings, insurance, and fund transfers” (p. 2f).3 Connected to the idea of increasing the poor’s financial capital through MF services is the expectation of governments, microfinance institutions (MFIs) and NGOs involved in MF that the poor use microcredit for creating or expanding small and medium-sized enterprises (SMEs) (Rosenberg, 2010; CGAP, 2006). Hence, MF is viewed as “a road to economic growth” and social and economic development leading to poverty reduction (Dichter & Harper, 2007; Hulme, 2007). Since women worldwide suffer from societal gender inequalities and are often poorer than men, MF programs have particularly targeted women and other vulnerable populations (Brau & Woller, 2004).

MF itself has grown from a poverty reduction-oriented tool operating on donor-funded NGOs projects to becoming a commercialized for-profit sector that spread in many developing countries (Copestake, 2007; Drake & Rhyne, 2002). Many debates in MF literature center on interest rates, collateral issues, MF innovations and techniques and group-lending versus individual lending (Mersland & Strøm, 2012). In contrast, as Morduch (2000) and Rosenberg (2010) argue little attention has been paid to assessing the impact of what some believe to be a ‘magic bullet’ for poverty reduction (Hulme, 2007). Similarly, “almost no studies provide comparable and reliable evidence on attributes as basic as the incomes, occupations, or loan uses of clients”, as Morduch (2000, p. 618) remarks, which is a strong deficit given that understanding loan use it is a first step in examining the effect of MF and establishing whom MF serves (Ibid.). Empirical studies that do examine the impact of MF are largely shaped by methodological debates on how to assess impact and are rarely generalizable as they focus on small case studies (Hermes & Lensink, 2011).

Studies on MF impact can be divided into two somewhat opposing strands: One the one hand, Zinman and Karlan (2009) find in their studies of microloans in the Philippines that business profit of loanees rose due to access to MF. Banerjee’s et al. (2009) India study comes to a similar conclusion as they observed increased business activities but no effect on health, consumption or education. Hence, these studies assert a more economic development than social impact of microloans. On the other hand, Robinson and Dupas’ (2009) study of microsavings in Western Kenya reported improvements in welfare such as health, education and consumption. Studies by Halder and Husain (1999) and Zaman (2004) also suggest that poorer loanees in particular use their loans rather for consumption smoothing than for the claimed aim of business expansion or creation, which is also argued by Banerjee and Duflo (2011). This is in line with McKee’s (1989) argument that microentrepreneurs mainly focus on raising their consumption levels at the expense of business development. This group of studies hence

3

Microcredit, as a part of microfinance, is a narrower approach as it is only centered on proving small loans to the poor and does not include the wider financial services that microfinance offers

(18)

10

emphasizes the social impact of microcredit. Overall, it can be argued that the impact of MF on poverty reduction and socio-economic development is mixed (Verrest, 2013) and still unclear (Rosenberg, 2010). A more realistic picture of the contribution of MF to business and social development is given by Morduch (2000) and in Rosenberg’s (2010) review of MF impact studies. Rosenberg (2010) argues that “The poor use credit and savings not only to smooth consumption, but also to deal with emergencies like health problems and to accumulate the larger sums they need to seize opportunities (occasionally including business opportunities) and pay for big-ticket expenses like education, weddings, or funerals” (p.2).

Rural income generating activities

The use of microfinance to promote IGAs as a livelihood diversification strategy has been a widely applied approach by various development actors. Although agriculture, both on commercial and subsistence level, continues to be of importance for rural development, rural non-farm (RNF) activities as “complementary engines of rural growth” are increasingly significant for securing rural livelihoods (Carletto et al. 2007: 147). Both types of activities are wide-spread among rural households and are in a complementary relationship as RNF activities are often countercyclical with agriculture as farming underlies seasonality. Thus, according to Carletto et al. (2007) rural income-generating (RIGAs) can be defined as 1) agricultural activities generating livestock, crop or wage-employment income and as 2) non-agricultural activities generating wage, self-employment, transfer or other income whereby the latter are often small-scale, home-based and informal.

2.2.2 Gender and use of microloans

Gender

Gender can be defined as a social construction or institution that is learned and expressed through differing roles, responsibilities and power relations between men and women (Danida, 2008). Gender differences not only manifest at the household level but also at extra-household structures and institutions (Beard & Cartmill, 2007), whereby they differ within and between cultures and underlie a dynamic nature as they can change over time (Danida, 2008). As noted in Chapter 2.1.2 the livelihoods approach and the household as unit of analysis fail to recognize women’s productive function and hide intra-household gender inequality. O’Laughlin (1995) emphasizes that this disregards women’s very active participation in socio-economic processes. Similarly, Whitehead (1994, p.36) emphasizes women’s important role in farming systems in Africa which “(…) are based on a complex inter-relation of men’s and women’s work” that is shaped by political and socio-economic factors as well as gender norms. Yet gender bias in education, access to credit, legal system, political representation and agricultural services create institutional barriers for many women around the world to improve their wellbeing (O’Laughlin, 1995, GGGR, 2014).

(19)

11

Gender and microcredit loan use

In that context, microfinance has gained popularity as a (claimed) resource of women’s empowerment by granting women access to financial capital and enhancing their economic independence (IFAD, 2009). However, Kabeer’s Bangladesh study (2001) reported that women, in particularly poorer ones, mainly used their loans and business profits on meeting consumption needs while men used them for business investments. This view is also shared by Khan (1999), Mayoux (2000) and Pitt and Khandker (1998) who argue that women are more likely to spend their income or loans on household expenditures, food, clothing and children’s education, which constrains their business development (Ibid). Mehra’s et al. study (1995, p. 57) asserts, “They [women] seem to prefer to invest more in household consumption and human capital development (i.e. child education and health care) than in expanding their enterprises”.

A number of reasons explain why women are more likely to use their microloans for social needs of the family than men, which can constraints women’s economic development: One important aspect is that in many societies women are traditionally responsible for household subsistence expenditures, which limits women’s choices to invest in IGAs (IFAD, 2009). This in turn can have negative impacts on women’s ability to repay loans. Furthermore, men might reduce their financial contributions to the household once women have access to loans (Kabeer, 2001). Women then have to substitute and invest more of their loans in family matters (IFAD, 2009). Hence, Kabeer (2001, 1998), Rai and Ravi (2011) and Mallick (2002) argue that women’s access to loans can have ambiguous effects on intra-household gender relations and women’s empowerment in general. At the same time psychological stress can increase for women when they take out microloans as they are faced with the additional responsibility of having to repay a loan (debt) and carrying out economic activities on top of strenuous work at the household-level (Kabeer, 2005). Goetz and Sen Gupta (1994) argue that women are not free in their loan use decision-making as men exert influence and control over loans. This aspect is connected to the general view that women (might) have access but little control over resources such as land, incomes (Berry, 1989; IFAD, 2009) and in this case loans.

Given that women are often more affected by poverty, have lower education, financial literacy and income levels and less access to financial services than men women face strong structural deterrents to starting and running successful small and medium-sized enterprises (SMEs) (Ibid.; Mwobobia, 2012). They are also time-constrained in running businesses and gaining necessary skills due to the unpaid domestic work they carry out which is linked to division of gender roles in which the woman is supposed to be the prime caretaker of the home and children (Ibid.; Agarwal, 1997). Women may also face male resistance when they want to take up economic activities as men might see their traditional provider role, which is connected to their status as household head, threatened as has been pointed out by Kabeer (2001) and IFAD (2009). When women do take up economic activities they often remain within IGAs typically carried out by women (Ibid.)

(20)

12

This thesis focuses mainly on women in terms of gender, which is a theoretical limitation as a holistic gender analysis would also involve men. While recognizing this shortcoming, this choice is justified by the fact that the majority of table banking members are women and that (Kenyan) women are generally poorer and more exposed to vulnerabilities than men (CABE, 2012). However, data was also collected on men but mainly for comparative purposes.

2.2.3 Critique of microfinance from a post-structuralist perspective

The rise of MF since the 1980s/1990s can be contextualized within development thinking and practice that has increasingly emphasized the pivotal role of the private sector in generating economic growth and development (Verrest, 2013). This market-led growth regime, grounded in neoliberal economic ideology, went along with the retreat of the state and responsibilization of the individual (Rankin, 2002). In that belief it was worthy to support private individual (micro) entrepreneurs through MF program as that would foster people’s self-reliance, economic growth and be reconcilable with the decreased role of the state (Bateman, 2010) . However, Bateman and Chang (2009), argue that MF strong link to neoliberal ideology, fosters the replacement of state welfare programs with private sector provisions in which the individual is responsible for his/her own fortune. In their view the underlying logic of MF perpetuates “distorted structures of wealth and poverty that historically are mainly responsible for the creation and perpetuation of poverty” (Ibid., p.25). These arguments are also reconcilable with a post-structuralist perspective such as by Ferguson (1994) or Escobar (1995, 2000) who emphasize that neoliberal development promotes the responsibilization of the individual and depoliticization of development. Applying the critique of the underlying logic of MF to Escobar (2000) one could argue that MF offers no alternative to neoliberal development as MF by itself is founded upon and contextualized within the discourse of neoliberalism.

Chapter 8, which abstracts the table banking concept, takes up this critical perspective of MF again.

2.2.4 Informal microfinance institutions and table banking in Kenya

The Consultative Group to Assist the Poor (CGAP) (2006) classifies four categories of microfinance providers: Formal microfinance providers are NGOs like Opportunity International, Kiva or Prodem in Bolivia and formal financial institutions such as banks. On the informal end there are financial service providers such as moneylenders and member-based organizations including self-help groups and credit unions who operate through self-regulated and/or group-based rules (Hospes et al., 2002). Informal MF institutions are often tightly linked to social relations and networks within the respective community as friends, neighbors or family members might together engage in informal MF (Ben-Porath, 1980). They operate on a local and small-scale level, which allows them to be highly flexible, operate on low costs and understand the needs of their clients or members (Hospes et al., 2002). However, due to their informality regulation, supervision and possibly insufficient financial skills can be an issue (CGAP, 2006). Although informal MF institutions have largely shaped and influenced the formal MF sector, relatively little research exists on informal MF (Schreiner, 2001). As Putnam

(21)

13

(1993) states that social capital is employed in order to achieve certain ends, one could argue that the formation of informal microfinance institutions underlies the same logic – in that case social capital, along with human and financial capital is used to facilitate loans, savings and other financial services.

Informal (Microfinance) Institutions in the Kenyan Context

Local grass-roots associations and collective actions efforts are deeply rooted within Kenyan society pre- and post-colonialism (Thomas, 1987) as its national motto harambee (Kiswahili for “Let’s pull together”) symbolizes. For Kenyan women, women councils had an important role in defining women’s status and role in society during pre-colonial times (Robertson 1996). Due to socio-economic and political changes during colonialism and after independence Kenyan women associations shifted to primarily focusing on IGAs or related activities in order to mitigate the economic hardships that many households faced (MacKenzie, 1986, Ngau, 1987). Harambee activities, which can according to Ngau (1987, p. 524) be defined as a form of “social exchange” between community members, are still carried out in Kenyan society, in particularly in rural areas when it comes to fundraising for school fees, funerals or churches.

Merry-go-rounds

Nowadays, Rotating and Services Credit Associations (ROSCAs) or Accumulating Savings and Credit Associations (ASCAs), referred to as merry-go-rounds (MGRs) in the Kenyan context are still popular activities rooted in traditional mutual guarantee systems. Their appeal to Kenya’s rural poor is that most self-employed individuals do not have access to formal banks as they are often located in urban areas (Mwobobia 2012). Thus, participation in MGRs is high, in particular among women (Gugerty, 2007). Dupas & Robinson (2009) describe MGRs the following:

“Most ROSCAs have periodic meetings, at which members make contributions to the shared saving pool, called the “pot”. The pot money is given to one member every period, in rotation until everyone has received the pot.”

However, there are many variations of MGRs as they can also entail rotating goods, labor or livestock. One of the primary motivations for joining MGRs, which are often organized at village level, is that it is difficult for people to save at home due to self-control problems and claim for cash by other family members as Ambec and Treich’s (2007) and Dupas and Robinson (2009) report. For women in particular MGRs offer a space to protect savings from claims by husbands as Anderson and Baland (2002) remark. Although use of the credit that MGRs offers differs among members, MGRs are mainly perceived as a consumption smoothing activity (Brannen, 2010).

Table Banking Concept

Compared to the quite extensive literature on Kenyan MGRs, to my knowledge, no academic literature exists on the table banking concept. According to JOYWO (2013), Kenya’s largest table banking NGO, table banking “was initially developed by the Poverty Eradication Commission under the

(22)

14

former Ministry of Planning and Vision 2030, targeting MDG 1 on eradicating abject poverty”. It has hence received much attention among Kenyan NGOs and governmental ministries and found a prominent public supporter in Rachel Ruto, the wife of Kenya’s Deputy President.4

Table banking is used by a number of Kenyan NGOs such as in a prisoner rehabilitation program of Resource Oriented Development Initiatives (RODI) Kenya, among women groups of the Joyful Women Organization (JOYWO), the ADS (Anglican Development Service) in Western Kenya and among mainly female smallholder farmers and small-scale producers through CABE. The term table banking draws from the fact that all financial transactions (the banking aspect of saving and lending) are carried out in front of all members of a group sitting around a table on which (usually) a cash box (‘the pot’) is placed. Although the specifics of how table banking functions depends on the NGO or respective group, table banking generally works the following5: Members of a table banking group come together once a month and bring a share, which are their savings, to the group’s pot. From this pool loans are given to members, usually at an interest rate of 10% through which the group accumulates capital and can provide more and higher loans over time. At the end of the year dividends are paid out to the members. The official idea behind table banking is that its loans are used to create small-scale business development which can help to reduce poverty (Ruto cited in CitizenNews, 2013).6

CABE’s table banking scheme7

, which this thesis deals with, has grown between 2011, when the concept was firstly introduced, from four groups who owned a revolving fund of KES 4.700 to 13 groups who circulated a pool of KES 800.000 in 2013 (CABE, 2012). Hence, in comparison to other Kenyan NGOs, CABE’s table banking project operates on a small scale. Declared objective of the project is to “facilitating the communities’ involvement in income generating micro-enterprises” (CABE, 2013, p. 3). Members of CABE’s groups also carry out a number of group projects such as poultry keeping or commercial farming in order to raise the groups’ capital. In summer 2014, the organization has begun distributing so-called top-up loans in order to infuse more capital in the groups as groups often lack enough capital. In its 2012 report CABE states that table banking members use their loans for a variety of purposes such as paying medical bills, school fees or investing in IGAs – which appears to be in line with the literature introduced in Chapter 2.2.1. Although CABE’s table

4

Rachel Ruto initiated JOYWO in 2009. The organization has 159.271 members in over 10.000 table banking groups mainly in rural Kenya and a total revolving fund of over 1 billion KES as of 2014 (JOYWO, 2014). The organization has various donors such as HIVOS International, UN WOMEN, NHIF and Unilever and has received much support from the Government of Kenya (GoK).

5

Chapter 5.2 describes in detail how table banking works in the case of CABE’s groups. 6

Although many table banking groups, like many MGRs, are registered with Social Services of the Government of Kenya (GoK), they can be classified as an informal form of MF as the aspect of banking is based on group-rules and not a legal enforceable framework.

7

(23)

15

banking group members regularly take loans and received capacity building training on saving, keeping group records and business development, investments into IGAs have not yet taken off as expected which CABE attempted to understand through consultation with community members and this research. The following sub-chapter presents factors that could possibly influence members loan use decision-making.

2.3 Factors influencing loan use decision-making

By drawing on the main concepts of (informal) microfinance, RIGAs and gender as well as the theoretical grounding of the SLA this sub-chapter establishes a number of factors that possibly influence table banking members’8

decision-making in using loans for IGAs. Linked to the units of analysis (see Chapter 4.4) factors are examined at various levels: the individual, group and structural. Furthermore, factors are categorized into internal and external factors.9 Rather than introducing gender as another factor, which has already been introduced in terms of loan use in Chapter 2.2.2, gender is the intrinsic component underlying the analysis of all outlined factors.

2.3.1 Internal factors at individual level

Personal priorities

As argued in Chapter 2.1.1 the livelihoods approach attempts to incorporate people’s choices and strategies which are also linked to the priorities they define for themselves. Ashley and Carney (1999) argue that “placing people and the priorities they define firmly at the centre of analysis” is crucial as it reveals what people themselves value most (p.1) This in effect can make policy decision more targeted and assess whether “external support is congruent with people’s livelihood strategies and priorities” (Ibid., p.7). Although people’s priorities are subjective and individual to a certain extent, McGregor (2004) argues that incorporating subjectivity in wellbeing research is very important as it better informs policy-makers and pays more attention to the specific context of the local level than universalist analyses have done. However, it also has to be noted that as dynamic poverty itself is (Narayan & Pritchett, 2009), people’s priorities underlie the same logic. Pouw and McGregor (2014) hence argue “preferences adapt according to changes over people’s life-cycle” due to alterations in influencing factors (p.10). Priorities or preferences hence have, according to these authors, emergent properties (Ibid.) that need to be taken into account.

In line with these arguments my own research attempted to incorporate subjective elements through researching members’ priorities in loan use as this can be considered one of the driving factors

8

As an abbreviation this thesis will state members/groups whenever it refers to CABE’s table banking members/groups.

9

However, internal and external factors do no equal agency and structure as argued in Chapter 2.1.2 people’s background and strategies are also largely shaped by structural issues.

(24)

16

influencing their decision-making. Pouw and McGregor’s (2014) argument was incorporated by comparing intended and actual loan use in order to see whether people’s priorities have changed over time. This ‘personal perspective’ is also consistent with the underlying research approach that included participatory methods and aimed at presenting the findings from the perspective of the people this thesis is about.

Perceptions of risk

As outlined in Chapter 2.1.1 one aspect of the SLA is the vulnerability context, encompassing shocks, trends and risk (DFID, 1999). Banerjee and Duflo (2011) recite the following risk factors in the lives of the poor: income, food, health, political violence, corruption, crime and insecurity (p. 136). Given that taking a loan and investing a loan into an IGA bears, as every financial decision, certain risks it is important to examine what role risk plays in relation to loan use decision-making. Much livelihoods research such as by Ellis (2000) or Moser and Satterthwaite (2010) has focused on risk mitigation, coping and adaptation strategies – how people prepare, deal and adapt to these factors. However, how poor people themselves define and perceive risk in relation to MF in general, and informal MF schemes in particular, has largely been ignored. Given that people’s own perceptions of risk are likely to influence their livelihood strategies, this thesis aims to contribute knowledge to this topic by bringing in a more subjective perspective on risk. It hence rather focuses on members’ psychological reasoning such as the importance of past experiences, which has been pointed about by cognitive psychology (Dietrich, 2010), rather than their strategies. It examines risk in relation to loan taking, which is important since table banking groups exist through members taking loans, and risk in relation to business investment (loan use).

Socio-economic background: human and financial capital

In order to understand what influences a member to use a loan in a certain way, it is also important to consider the individual’s socio-economic background. Linking this to the SLA, financial capital, the savings, credit, remittances an individual has or does not have, is examined (DfID, 1999). Members’ financial capital is also related to their incomes. Hence, this research tests whether Zaman (2004) and Halder and Husain’s (1999) argument that the very poor often use a larger share of their loan for consumption rather than investment compared to better-off loanees, holds true in the case of CABE’s members. At the same time, as emphasized by Collins et al. (2009) it is not just the size of income that matter but also its regularity or irregularity, in particular in rural areas where income depends on the seasonality of agriculture. One of way of dealing with income irregularity is that people employ livelihood diversification strategies as reported in Chapter 2.1.1 (Ellis, 2000; Dorward et al., 2009). Hence, this is thesis also looks at how members, according to their socio-economic background, are able to employ livelihood diversification strategies. Since savings are another crucial aspect of financial capital and of the table banking scheme itself (in the form of members’ shares), my research also looks at how people save and what constrains the in saving.

(25)

17

Various authors such as Jones and Jajawarna (2011) and Davidsson and Honig (2003) point out the importance of human capital, defined as skills, knowledge, labor and health, in taking up new economic opportunities. Hence, this thesis also examines how skills and knowledge in terms of financial literacy and business development shape loan use decision-making. It attempts to see whether human capital theory, which argues that the higher a person’s human capital is the higher is his/her ability to lead a successful business (Honig, 1998) applies to the case of CABE’s table banking groups.

2.3.2 Internal factors at group-level

Since the table banking groups themselves constitute one form for members’ access to financial capital, assessing the capacities of these groups in terms of organization and group relations is another important factor as they influence overall group performance and in turn loan availability for their individual members. Nair (2005) argues community groups’ organizational capacity is shaped by member’ literacy levels, organizational experience, skills and knowledge-sharing. According to her these factors influence overall group performance. Hence, this thesis examines how the level of these factors is within the respective groups and also compares capacity among them, in order to understand whether groups face differing challenges which ultimately impact loan availability. In the Kenyan context this is particularly important as Mwobobia (2012) argues that there is a strong need for capacity building for women organizations in Kenya in the field of finance, literacy skills, marketing, production and managerial skills.

Intra-group relations can be approached from a social capital perspective. Grootaert and van Bastelaer differentiate between structural and cognitive social capital. The former is defined as ““externally observable social structures […] and “the rules and procedures they embody” (Uphoff, 2000 cited in Grootaert and van Bastelaer, 2002, p. 3), while the latter relates to “generally accepted attitudes and norms of behavior, shared values, reciprocity, and trust” (Grootaert and van Bastelaer, 2002, p. 3). Literature on (informal) MF groups remarks that groups’ governance structure including rules and regulations and deep-rooted power relations within a group are of importance when analyzing the success or failure of microfinance groups (Hermes et al., 2005; Wenner, 1995). The latter is especially important given the fact that even though on paper group members are often equal Marr (2002) states, “Original deep-rooted power structures, traditional norms and cultural attitudes may clash with the group’s internal regulations”(p.522). Hence, this thesis also looks at the cognitive dimension of social capital to see how norms, trust and reciprocity affect groups’ formal rules and group relations.

In that context the characteristics of a MF group, relating to its homogeneity or heterogeneity in terms of ethnicity, occupation, income, gender, is also said to affect the repayment rate and availability of further loans (Hermes & Lensick, 2007). Authors such as Paxton et al. (2000) argue that homogenous microfinance groups with strong social ties negatively affects loan repayment rates as they have less incentives for peer-monitoring and social sanctioning, an important part of the joint liability principle.

(26)

18

Others such as Karlan (2007) assert that in a homogenous group social capital, including trust and acquaintanceship to group members, tends to be higher, which they argue relates positively to better peer-monitoring increasing savings and repayment rates. Hence, this thesis looks at members’ socio-economic class, gender and how the groups were founded (neighbors, friends or strangers) in order to assess the homogeneity or heterogeneity of groups and how that impacts on loan repayment.

2.3.3 External factors at structural level

Given that “[people’s] options are also determined by the structures (such as the roles of government or of the private sector) and processes (such as institutional, policy and cultural factors), as Farrington et al. (1999, p.1) state, members’ decision-making must also be considered in light of structural factors affecting their livelihoods. This is in line with the relational approach of Chapter 2.1.2. Although the analysis of structural factors is limited due to the scope of this thesis, a number of external factors, which according to Polanyi (1944) can be of “political, social, cultural and historical nature” (in Pouw & McGregor, 2014) are examined. In that regard, Doward et al. (2003) emphasize the role of the local economy and markets in livelihoods analysis. Given that the official aim of table banking is to initiate IGAs it is also important to look at the state of infrastructure as pointed out by Narayan and Pritchett (2009) and the type of businesses people carry out (Banerjee & Duflo, 2011). Furthermore, the role of local politics and their possibly obstructing role in development (Narayan & Pritchett, 2009) is another factor to consider.

2.4 Conclusion

This chapter described the theoretical framework of this thesis. The SLA is a suitable framework to analyze what factors either constrain or enable people’s ability to make use of economic opportunities as it asserts that not only economic growth but also poor people’s capabilities are crucial to examine. However, as the relational approach asserts attention must be paid to the underlying structured guiding people’s choices and strategies. Similarly, given women’s vulnerabilities and that gender inequalities hamper socio-economic growth analyzing how gender influences loan use is very important. The chapter presented a number of internal and external factors that possibly influence members’ loan use decision-making, whereby it emphasized the importance of integrating subjective elements into the analysis.

(27)

19

Chapter 3. Context

Having established the theoretical framework of this thesis, this chapter provides the national and local context in which table banking groups operate. After outlining Kenya’s main political and socio-economic features, including gender inequalities, the chapter then turns to the specific research location of Samia District in which CABE’s table banking groups are located. After an overview over the research location and its physical resources local livelihoods of the area are presented. It also briefly introduces CABE.

3.1 Political, economic and social characteristics of Kenya

3.1.1 Political characteristics

Compared to some of its troubled neighbors the east-African nation of Kenya has experienced relative political stability since gaining independence from the United Kingdom in 1963. 51 years after independence the presidential republic is led by President Uhuru Kenyatta of the Jubilee Alliance/TAN party. Kenyatta succeeds the Grand Coalition Government of President Kibaki and Prime Minister Odinga which resulted from a power sharing agreement after Kenya’s 2007/2008 post-election crisis. In 2010, Kenyans approved of a new constitution that aims, among other aspects, at greater control over the executive, political decentralization and reforming the judiciary (Glinz, 2010). Regional and national security issues, regional integration, land reform, corruption scandals, the reintegration of internally displaced people, politicization of ethnicity and proceedings by the International Criminal Court (ICC) against prominent political figures are further issues dominating the country’s political scene in the last years (BTI, 2012; Booth et al., 2014; Mwangi, 2008).

3.1.2 Socio-economic characteristics

Kenya is a young and diverse nation with over 80% of the population under the age of 35 years and 42 ethnic groups (UNDP, 2013a; CIA, 2014). In the late 1980s the World Bank and International Monetary Fund (IMF) initiated structural adjustment programs (SAPs) in Kenya, encompassing wide political and economic reforms aiming at democratization and market liberalization. Critics emphasize their negative impact on income inequality, poverty, employment and public service provision for Kenyans (Radeny et al., 2012; Rono, 2002). With an annual GDP growth of 4.7% in 2013, Kenya’s economy continues to grow despite a sharp decline in GDP during the post-election crisis in 2008 (Booth et al., 2014). Main drivers of the economy are agriculture, which contributed 25.3% to Kenya’s GDP in 2013, financial services, manufacturing, trade, tourism, transport and communication (KNBS, 2014). The informal sector in which nine out of ten people are employed is a major livelihood source to urban and rural inhabitants (ADBG, 2013).

Despite Kenya’s fame as a regional finance, communication and transport hub, the country still counts as a low-income country ranked 145 among 187 countries in the United Nations Development

(28)

20

Programme’s Human Development Index 2013 (UNDP, 2013b) with a GNI of US$ 860 per capita10 in 2012 (World Bank, 2012). About 45.5% of Kenya’s population of 43.18 million lives below the national poverty line (Ibid.) whereby poverty is higher in rural areas (SID, 2014). Uneven distribution of income within the country led to high inequality in Kenya with a GINI coefficient of 0.44511 (Ibid.). Notwithstanding declining population growth rates (3% in 2012) and rising life expectancy (61.1 years at birth in 2012) overpopulation and the prevalence of diseases such as HIV/AIDS continue to pose major development challenges (UNDP 2013a; UNICEF, 2013). In terms of education the country ranks better than its neighbors with an adult literacy rate of 79%, although there is a literacy gender gap with 84.8% of males and 73.6% of females being able to read and respectively (KIHBS, 2005/2006).

Although urbanization has increased in the past twenty years, more than three quarters of the Kenyan population inhabits rural areas (IFAD, 2013b) whereby 11% of those live in Western Province where this research was carried out (KNBS, 2009). The rural economy is mainly characterized by agriculture which is a livelihood source for more than 70% of the rural population. Main cash crops are coffee, tea and horticulture produce for export. In order to reduce poverty and tackle the country’s development issues the Kenyan Government launched its development blueprint ‘Kenya Vision 2030’ in 2008 which was developed in a multi-stakeholder approach (Booth el al, 2014).

3.1.3 Gender inequality in Kenya

Although Kenya has improved its scoring in the Global Gender Inequality Index considerably, from place 78 in 2013 to rank 37 in 2014, gender disparities continue to persist (Global Gender Gap Report, 2014). In the fields of political representation, education, economic opportunities, health, rights and social institutions Kenyan women are not on equal footing compared to men (Ibid.; Foulds, 2014) Gender relations in Kenya are mainly shaped by customs and cultural practices, economic conditions, education levels as well as traditional and modern laws (Mwatha, n.d.). Women are also discriminated against socio-economically as poverty severity levels tend to be higher among Kenyan women than men and women-headed households are more often classified as very poor than male-headed households (Ibid.; Foulds, 2014). Rural women in Kenya contribute 89% to subsistence farming but only 32% to off-farm paid labor (Global Gender Report, 2014; Federation of Women Lawyers, 2013). Although women run almost half of all SMEs in Kenya, they have less access to financial services than men which constrains their businesses development (Njoroge & Muturi, 2013). In rural areas in particular women mainly rely on informal savings and credit associations such as MGR to access capital (Karanja et al., 2014). Although the 2010 constitution eliminated gender discrimination with regards to land ownership and property rights, customary law often undermines its application as only 1% of land titles are held by women only (Federation of Women Lawyers Kenya, 2013) which

10

Based on Atlas method. 11

How the unbanked bank themselves in rural western Kenya : understanding small-scale farmers' and producers' decision-making in using microloans for income-generating activities

Master’s Thesis

How the Unbanked Bank Themselves In Rural Western Kenya:

Understanding small-scale farmers’ and producers’ decision-making

in using microloans for income-generating activities

In collaboration with:

Supervisor

Second Reader

Local Supervisor

Abstract

Table of Contents

List of Tables ... 1

List of Figures ... 1

List of Photos ... 1

List of Abbreviations ... 2

List of Words ... 2

Exchange Rate ... 2

Chapter 1. Introduction ... 3

Chapter 2. Theoretical Framework ... 5

2.1 The Sustainable Livelihoods Approach ... 5

2.1.2 Origins, purpose and framework of the Sustainable Livelihoods Approach ... 5

2.1.2 Critique of the Sustainable Livelihoods Approach: gender and relational approach .... 7

2.2 Microfinance, gender and table banking ... 8

2.2.1 Microfinance and rural income generating activities ... 8

2.2.2 Gender and use of microloans ... 10

2.2.3 Critique of microfinance from a post-structuralist perspective ... 12

2.2.4 Informal microfinance institutions and table banking in Kenya ... 12

2.3 Factors influencing loan use decision-making ... 15

2.3.1 Internal factors at individual level ... 15

2.3.2 Internal factors at group-level ... 17

2.3.3 External factors at structural level ... 18

2.4 Conclusion ... 18

Chapter 3. Context ... 19

3.1 Political, economic and social characteristics of Kenya ... 19

3.1.1 Political characteristics ... 19

3.1.2 Socio-economic characteristics ... 19

3.1.3 Gender inequality in Kenya ... 20

3.2 Research location ... 21

3.2.1 Location and administration ... 21

3.2.2 Physical resources ... 21

3.2.3 Local livelihoods and the local economy ... 21

3.2.4. Local microfinance governmental programs in line with national policies ... 22

3.2.5 Introduction of Centre for African Bio-Entrepreneurship ... 23

3.3. Conclusion ... 23

Chapter 4. Methods ... 24

4.1 Research questions ... 24

4.2 Epistemology and research methodologies ... 24

4.3 Conceptual scheme ... 26

4.4 Units of analysis ... 27

4.5 Methods ... 28

4.5.1 Data collection ... 28

4.5.3 Data analysis ... 33

4.6 Ethical considerations ... 33

4.7 Conclusion ... 34

Chapter 5. Overview over table banking groups from a gender perspective ... 35

5.1 Characteristics of table banking groups ... 35

5.1.1 Members’ heterogeneous socio-economic backgrounds ... 35

5.1.2 Inter- and intra-group socio-economic differences ... 38

5.1.3 Strong social ties in the groups ... 40

5.2 Table banking model ... 40

5.2.1 Income sources of a group ... 41

5.2.2 Loan conditions and lending considerations ... 42

5.2.3 Loans borrowed ... 42

5.2.4 Repayment record ... 44

5.3 Governance structure of table banking groups ... 45

5.3.1 Group leadership ... 46

5.3.2 Decision-making processes and general rules ... 46

5.4 Conclusion ... 47

Chapter 6. Use of table banking loans from a gender perspective ... 47

6.1 Intended use of loans ... 47

6.2 Actual use of loans ... 48

6.2.1 Comparing intended and actual loan use ... 48

6.2.2 Actual loan use ... 49

6.2.3 Members’ perceptions of the purpose of table banking ... 52

6.2.4 Women and loan use ... 53

6.2.5 Loan use in relation to existing microfinance studies ... 54

6.3 Conclusion ... 55

Chapter 7. Factors influencing loan use from a gender perspective ... 55

7.1 Internal factors influencing loan use at individual level ... 55

7.1.1. Personal priorities in loan use ... 55