Introduction

Inclusive development has become a mantra of international development organisations. It is reflected in the 17 UN Sustainable Development Goals (SDGs) and informs the policies of organisations in the UN system and world economic fora,Footnote 1 as well as many other international and non-governmental organisations in the Global South. There has also been growing interest in organised action by the subjects of inclusion, particularly within the social and solidarity economy (SSE) of which co-operatives are an important component (Utting 2014). The International Labour Organization (ILO) (2014a, p. 1) states that ‘the promotion of the SSE (i.e. co-operatives, mutual benefit societies, associations and social enterprises) is an efficient way to promote social justice and social inclusion for all members of society’. An ILO and International Co-operative Alliance (ICA) (2014, pp. 1–10) report on co-operatives and SDGs notes that ‘co-operatives foster democratic knowledge and practices and social inclusion, making them well-placed to support the achievements of sustainable development’. The UN International Day of Co-operatives in 2017 was held under the banner of inclusion: ‘Co-operatives ensure no-one is left behind’.Footnote 2 Researchers in the European Research Institute on Co-operative and Social Enterprise argue that SSE organisations in Africa can promote inclusive and sustainable growth (Borzaga and Galera 2014), and in 2018, the FAO Director-General José Graziano da Silva asserted that ‘co-operatives and other associations—in Africa—are the only way for providing family farmers with technical assistance, capacity building, financial resources and access to modern technologies’.Footnote 3

It has therefore been argued that co-operatives offer low-income and disadvantaged people a social and business model based on participatory, sustainable and inclusive practices. The particular interest in co-operatives as sites for inclusive development is based on the values and principles that drive and inform them. These were laid out in ILO recommendation 193 on the promotion of co-operatives,Footnote 4 which stated that they are ‘an autonomous association of persons united voluntarily to meet their common economic, social and cultural needs and aspirations through a jointly owned and democratically controlled enterprise’, involving ‘(a) co-operative values of self-help, self-responsibility, democracy, equality, equity and solidarity; as well as ethical values of honesty, openness, social responsibility and caring for others; and (b) cooperative principles as developed by the international cooperative movement. These principles are: voluntary and open membership; democratic member control; member economic participation; autonomy and independence; education, training and information; cooperation among co-operatives; and concern for community’.

Co-operatives are therefore multi-dimensional and solidaristic organisations that seek to improve the social and economic conditions of their members while grappling with the wider economical, social and political context. From small community groups to major companies, they straddle many sectors and functions. As enterprises, they are part of the economy (whether under capitalist or non-capitalist regimes), have to engage with the regulatory and policy environment (and therefore the state), and are informed by and reflect wider social relations. In a development context, Ecuador and Bolivia are amongst the most prominent examples where governments have actively sought to improve the wellbeing of low-income people by putting in place robust policies to promote SSE and co-operatives (Hirschman and Denburg 2013; Coraggio 2017). The ICA (the global network of co-operatives set up in 1895, now with 1.2 billion members)Footnote 5 also strongly promotes co-operatives in development as well as South–South collaborations: ‘Co-operatives are a tried-and-tested model in international development, enabling people around the world to take control over their livelihoods. Supporting the growth of co-operatives is, therefore, an established way to empower people and local communities to take charge of their own development, putting people before profit.’Footnote 6

The purpose of this article, then, is to examine the proposition that co-operatives promote inclusive development and the kinds of challenges they face. The next section reviews academic debates on inclusive development, locating the place of co-operatives within them. It identifies three interrelated dimensions which emerge from literature and characterises some key dimensions of the co-operative model: economic control, voice and agency. The subsequent section explains the background to the study and how it was carried out. The empirical focus is on low-income settings in poor capitalist economies, specifically in this case, Malawi, a country that has been promoting new co-operative development in recent years. While co-operatives have had a mixed history in Africa, with experiences of state control, corruption and then decline during structural adjustment, there has been a renaissance in the continent during the last 20 years (Develtere et al. 2008; ILO 2014b). The article is based on qualitative fieldwork with two of the largest co-operative unionsFootnote 7 in Malawi, which took place in 2013, with follow-up in-depth interviews in 2017. This is followed by an analysis of the results. The data analysis illustrates the lived experience of being a co-operative member and leader, which will resonate with many other challenging social and economic settings across the Global South. There are therefore wider lessons about co-operatives and inclusive development as a process, explored in the last section of the article.

Inclusive Development and Co-operatives

Van Gent (2017, p. 2) argues that a ‘clear-cut definition of inclusive development is non-existent, and as a result the concept has led to significant ambiguity in both the literature as well as policy frameworks’. Current conceptualisations of inclusive development (Gupta et al. 2015; Hickey et al. 2015; Teichman 2016) emphasise meta narratives of development, or in the words of Gutpa and Pouw (2017, p. 97), theories and debates about inclusive development are ‘locked into big-D development’. In this article, we are not concerned with all the logical ways by which inclusive development can be realised. Our interest is to understand how organisations of low-income people such as co-operatives can achieve or contribute to inclusive development. Therefore, our focus is on the ‘little-d development’ (ibid, p. 96) driven by bottom-up processes.

Understandings of inclusiveness have been appropriated by different perspectives on poverty and the position of ‘the poor’ in society. In particular, there is a tension over the relative roles of the state, the private sector and civil society in addressing poverty. While neoliberal thinkers have seen market forces and a strong private sector as the best way to distribute goods and services, particularly during the 1980s, growing economic inequality and continuing poverty have underlined the necessity for a redistributive state to meet social and economic needs. This has given rise to proposals for inclusive growth on one hand and to ideas about the need for public action and citizen-led change on the other.

In the early 1990s and 2000s, the World Bank (WB) and the International Monetary Fund (IMF) introduced and promoted ‘inclusive neoliberalism’ in the form of Poverty Reduction Strategy Papers (PRSPs) which, according to Ruckert (2006, p. 7), were the ‘most visible policy tool of the inclusive neoliberal development’ regime during the 1990s and early 2000s. Klak et al. (2011) argue that the type of ‘inclusive neoliberalism’ promoted by the WB and IMF was an attempt to make development policy ‘country driven’ and ‘indigenously owned’ (Fine 2001, p. 12; Craig and Porter 2006, p. 53). However, inclusive neoliberalism was criticised for being top-down rather than driven by developing countries (Plehwe 2007). Others have questioned the related pro-poor growth agenda (Cook 2006; Chibba 2008; Fernando 2008) on the grounds that benefits to the poor may simply be an indirect outcome rather than a deliberate target of policy (Rauniyar and Kanbur 2009, p. 3). Although inclusive growth may include deliberate welfare dimensions and redistributive policies, such as education and health measures that are directed at the poor, it is argued that inclusive development also requires public action and advocacy, as well as (or to prompt) state intervention.

In the context of these debates about the nature of growth, inclusion can easily become a statistic rather than a comment on the quality of life (rather like employment figures that ignore the conditions of work). Inclusion per se is a necessary but not sufficient condition for improving the quality of life and wellbeing, which also involves the active production of the conditions required to meet social needs (Hickey and du Toit 2007). Gupta et al. (2015, p. 546) define inclusive development as ‘development that includes marginalized people, sectors and countries in social, political and economic processes for increased human well-being, social and environmental sustainability, and empowerment’. In a similar vein, Hickey et al. (2015, p. 6) conceptualise it as a ‘process that occurs when social and material benefits are equitably distributed across divides within societies, across income groups, genders, ethnicities, regions, religious groups.’ Equally, Teichman (2016, p. 2) argues that inclusive development is the ‘development that provides basic physical security for the population, achieves the elimination of poverty, and mitigates the deprivations that prevent citizens from participating fully in society.’ In related literature on inclusive innovation, Papaioannou (2014a, p. 2) argues that inclusive innovation requires ‘a multi-dimensional theory of justice…related to social equality, equality of opportunity and democratic participation’, and public action and campaigns ‘promoting bottom-up changes to global policies and institutions’ (Papaioannou 2014b, p. 19). This dynamic view evokes Young’s (2000, p. 23) idea of an inclusive and democratic process where ‘all those affected … are included in the process of discussion and decision-making’ and Fraser’s conceptualization of inclusion as a deliberative process rather than an ideal, a state or a set of policy prescriptions (Fraser and Honneth 2003). Such views, then, question how low-income people themselves are involved in such action. These are the kinds of questions raised, for example, by Mohan’s (2006, 2007) discussion of active citizenship, Cornwall’s (2004, p. 76) ‘popular’ spaces of participation and Webster’s and Engberg-Pedersen’s (2002) concept of ‘political space’ that can be appropriated by organisations of the poor. Such ideas resonate too with those of Gibson-Graham and Roelvink (2013), whose work on community economies includes control (‘something we do, rather than something that does things to us’ [ibid, p. 459]) and appropriate frameworks of governance (ibid, p. 461).

Co-operatives straddle the spaces of private enterprise and popular participation. They are organisations that aim to address the social and economic needs of their members through democratic and deliberative governance and decision-making, as well as engage with other co-operatives and the wider community. They also have to engage with the economy and, as legal personalities,Footnote 8 with the regulatory and policy environment. This multi-dimensional role and its potential contradictions have been an object of ongoing debate; For example an earlier report for the FAO on smallholder organisations, including co-operatives, concludes that smallholder organisations are directed at the agricultural middle class rather than the really poor (Poole and de Frece 2010). The argument that co-operatives are challenged to include the really poor has been made in other literature; For example Münkner (2012, p. 30) argues that co-operatives are not able to benefit extremely poor people, because extremely poor people have no resources they can pool (i.e. labour, production, capital). We return to this point when analysing economic control. On the other hand, Münkner also argues that, although co-operatives do not help the poor directly, poor people can benefit from organised self-help and mutual aid provided by co-operatives (Birchall 2003, p. 13). Others argue that, although co-operatives can provide organisational structures to empower and provide services for the most vulnerable, they have limitations in making the most vulnerable part of their organisations (Lakshmi and Visalakshmi (2013), while Hannan’s (2014) research indicates that whether co-operatives reach poor people depends on organisational governance and how co-operative activities are carried out, a point to which we also return. Yet others point out the importance of the policy context; For example a literature review by Kwapong and Hanisch (2013, p. 127), which analysed four different perspectives, concludes: ‘Co-operatives provide a chance for the poor to work themselves out of poverty. However, a favorable environment needs to be created to support the development of co-operatives.’ Other contextual elements can create further challenges to inclusion, while also resulting in unexpected benefits for co-operative members; For example Verhofstadt and Maertens’ (2015, pp. 101–102) analysis of the propensity to become members of rural co-operatives in Rwanda notes that remoteness is not conducive to membership. However, if remote farmers do become members, the income effects are proportionately greater than for those farmers who are less remote. Nonetheless, while arguing that ‘agricultural co-operatives are not a solution for near-landless or land-poor farm-households’ because ‘the income effect of co-operative membership for these households is too low to get them out of poverty’, Verhofstadt and Maertens also note that there are uniformly positive effects for both male- and female-headed households, and for educated and less-educated members, suggesting that policies should aim towards including all forms in co-operatives (ibid).

There is considerable literature on co-operatives, poverty reduction and co-operatives’ capacity to include (or not) the poor and very poor. However, there are other dimensions to the nature of and challenges for inclusivity in co-operatives; For example a report for the World Bank (Agrawal and Perrin 2009) found tensions between inclusion and representation, and between social hierarchies and equitable and efficient service delivery. Such findings are echoed in Bijman et al.’s (2011) discussion of co-operatives engaged in global value chains (GVCs), where trade-offs are observed between hierarchy and democracy, and between being both business/market centred and community centred, in aiming to run a business that is both efficient and inclusive (Bijman et al. 2016).Footnote 9 Such trade-offs may also be influenced by the relative homogeneity or heterogeneity of the membership, where, it is argued, the former is more inclined to community and democracy, and the latter more to hierarchy (Francesconi and Wouterse 2011).

There is also considerable tension in literature about the economic performance of co-operatives—their failure in some cases, as well as success in others (Develtere et al. 2008). Literature on co-operatives in the African context has pointed to problems with state control, elite capture/corruption, poor leadership and management skills, lack of adequate service delivery leading to members abandoning the co-operative or selling on the side (a ‘loyalty dilemma’), restricted entry of women and youth, and obstacles to women’s leadership (Develtere et al. 2008; Majurin 2012). Similar tensions have been identified by Wanderley (2015) and Coraggio (2016) in Latin America. On the other hand, there has been a general increase in female participation in savings and credit co-operative organisations (SACCOs) because SACCOs do not require exclusive ownership of assets for membership. It has also been observed that women can gain non-economic benefits such as greater self-confidence, better negotiation skills, better gender relations in households, and taking more control of household decisions, particularly in women-only enterprises (Majurin 2012; Dash 2011; Ferguson and Kepe 2011; Burchi and Vicari 2014).

From this, somewhat inconclusive, literature, we can take away two main points: One is that co-operatives are multi-dimensional organisations with social purposes that go beyond the economic, narrowly defined, and therefore need to be analysed with a broader framework in mind (Borda-Rodriguez and Vicari 2014). Two, co-operatives are situated in national and global contexts of unequal social, economic and power relations, and therefore cannot be expected to provide perfect bottom-up models of inclusive development. However, extrapolating from the wider literature above, we can see that there are some key dimensions that map on to co-operatives as social and business models within the ‘little-d development’ framing of Gupta and Pouw (2017).

First is the community economies approach to economic control: ‘something we do, rather than something that does things to us’ (Gibson-Graham and Roelvink 2013, p. 459). In this conception, economic control is concerned with economic participation and education, training and information, all of which have the potential to enable low-income and disadvantaged people to have control of different aspects of production and distribution. Economic control provides an opportunity for greater influence over forms of livelihood and their outcomes, and have positive distributional effects if they improve low-income people access to goods and services (reducing adverse inclusion).

Second, by extension, is the need for deliberative and democratic processes (Young 2000; Fraser and Honneth 2003; Papaioannou 2014b), which we encapsulate in the term ‘voice’. Co-operative governance is based on one person one vote, which in principle, enables low-income people to express their views and be heard.

Third is the capability to act or ‘agency’, underpinning Mohan’s (2006, 2007) conceptualisation of active citizenship and Gibson-Graham and Roelvink’s (2013) idea of doing rather than being done to, and most particularly, the ability to take action and exert influence within the building blocks of community life.

Economic control, and particularly voice and agency, are closely inter-related in process and practice, which will become evident in the discussion of the research results below. In turn, inclusive development is seen as an on-going process involving many contradictions not simply an end state, either within co-operatives or more widely. With this conceptual framing in mind, the following sections analyse the contributions and contradictions of two co-operative unions in Malawi, and then reflect on the implications of co-operatives as a vehicle for bottom-up inclusive development.

Context, Data and Methods

Malawi is one of the poorest countries in Africa (UNDP 2014).Footnote 10 Since the 1990s, it has experienced floods and droughts, severe famine and an HIV/AIDS pandemic, while fuel and electricity shortages, a lack of adequate skilled labour, poor government regulation and lack of finance have created an uncertain investment climate for the private sector. Malawi’s agricultural sector employs nearly 90% of the working-age population, with the majority engaging in small scale agriculture. Access to land and agricultural policies in Malawi tend to favour large-scale production at the expense of smallholder farmers who already face landlessness and declining agricultural productivity (Chirwa 2004; Søreide and Williams 2014).

In recent times, consecutive droughts have reduced improved growth rates, and a combination of drought and floods has threatened food security, while poverty is said to be increasing in rural areas in particular.Footnote 11 Since 1994, primary education has been free for everyone; however, in 2015, only 35% of children finished school, and 73% of people between the ages of 15 and 24 years old were literate (UNESCO 2019). Macroeconomic instability and high-inflation rates (23% in 2016, 8% in 2017 and steadily increasing in 2018)Footnote 12 have severely affected people who live below the poverty line (US $1.90 per day)—estimated at about 50% of the population.Footnote 13

Malawi is also a country with a previously declining co-operative sector, which has been renewing itself in recent times as part of government policies to promote development (Chambo and Smith 2010). In Malawi, the Ministry of Trade, Industry and Tourism is responsible for promoting and overseeing co-operatives, and for channelling international development assistance to them. International development organisations and NGOs that support Malawian co-operatives include Twin Trading, Co-operative Development Foundation of Canada, Japan International Cooperation Agency (‘One Village, One Product’ movement), Land O’Lakes International Development, The World Bank and the International Development and Co-operation Department of the Commission of the European Union. Through a wide range of training programmes and strategic funding, these international organisations have to an extent improved the agricultural practices, governance, business planning and marketing of co-operatives and other forms of social enterprises in the country.

As in many other African countries, co-operatives in Malawi (particularly in agriculture) were established under colonialism, and agricultural marketing co-operatives became absorbed into the state or parastatal sector after independence (ibid: 9). In the 1970s, savings and credit cooperatives grew, eventually becoming part of the Malawi Union of Savings and Credits Co-operatives (MUSCCO)—one of the case study unions in this article. With the decline of parastatals and the state’s direct involvement in the economy during the 1990s, the co-operative sector was again encouraged to grow, and a small government department was established to support the co-operative movement (ibid.). The Co-operative Societies Act was passed in 1998, and the Financial Co-operatives Act in 2011. At the time of the fieldwork, records indicated that there were 681 registered co-operatives in Malawi, with 382 being active in the agricultural sectors, 107 being saving and investment promotion co-operatives and 192 being SACCOs.Footnote 14 Amongst them, the Malawian Department of Co-operatives reported that only 234 were active at that point, and there was then no apex body. However, data from 2017 report eight co-operative unions and 1200 co-operatives (but not the extent of their activity nor whether they are registered) (MA1 2017),Footnote 15 while a national co-operative apex organisation was established in 2015.

There are a few studies that have looked at individual agricultural co-operatives and concluded that lack of market access, managerial skills and poor governance have undermined individual co-operatives’ capacity to survive (Nkhoma 2011; Nkoma and Conforte 2011; Matabi 2012). By contrast, the research behind this article focussed primarily on the role played by co-operative unions which are expected to promote co-operative values with their affiliated primary co-operative members, and provide a number of services (i.e. capacity building, specialised training, credit, market outlets). Unions also represent the interests of members, and engage and negotiate with national and international organisations, such as development agencies and/or international buyers.

In 2013, those working on the research project identified and established a collaborative relationship with four of the biggest co-operative unions in Malawi, conducting 21 in-depth semi-structured interviews and 7 focus groups with co-operative leaders and members in the capital city of Lilongwe and in the rural and agricultural areas of Mzuzu. Focus groups and in-depth interviews (lasting 1–2 h) were designed to capture the views, experiences and voice of co-operative members. Interviewees were fluent in English, which allowed a rich dialogue on different aspects of inclusive development and co-operatives, while some translation assistance was required with the focus groups. The project also carried out 14 in-depth semi-structured interviews with policy-makers and officers from development aid organisations that work with co-operatives. In mid-2017, a further field visit was conducted, and eight further in-depth interviews were carried out, including five with union co-operative leaders.

For the purposes of this article, we focus on two large unions: Mzuzu Coffee Planters Co-operative Union (MZCPCU) and the Malawian Savings and Credit Co-operative Union (MUSCCO) (Table 1). MZCPCU and MUSCCO were selected because they are well established and have very different functions and composition. MZCPCU promotes the production and marketing of coffee by smallholders, and MUSCCO supports savings and credit in both urban and rural areas across Malawi. In MZCPCU, members are relatively homogeneous in terms of their goals and expectations. In the case of MUSCCO, primary co-operatives (SACCOs) are heterogeneous with different interests (for example farmers and teachers). Within SACCOs, interactions (taking out loans and depositing savings) are also individualised, although the membership as a whole is represented through the board.

Table 1 Key characteristics of MZCPCU and MUSCCO
Full size table

To carry out the analysis for this article, the interviews and focus group data were reviewed and recoded to identify connections amongst and between the core themes of economic control, voice and agency. The following sections first look at economic control (specifically the dimensions of co-operative membership and its education and training, as well as perceived challenges and contradictions), and then voice and agency (mechanisms for voice, voice in practice and means of exerting agency, also examining challenges and contradictions). It will be evident that there are considerable differences between the two co-operative unions in terms of the different types of inclusion they are able to promote, partly linked to the homogeneity/heterogeneity dimension and also to the dimensions of hierarchy/democracy theorised by Bijman et al. (2011). This does not mean that inclusive development is only possible with certain types of co-operative—rather that they experience different challenges. Given that inclusive development is a process embedded in wider social relations, contradictions are inevitably involved.

Results: The Cases of MZCPCU and MUSCCO

The origins of the MZCPCU are in the former parastatal Smallholder Coffee Authority (SCA), established in 1971 as a marketing mechanism for small coffee producers in Northern Malawi. However, its performance was poor (Chirwa et al. 2008),Footnote 16 resulting in serious debts. With structural adjustment and reduction of state involvement in the economy during the 1980s and 1990s, the Smallholder Coffee Farmers Trust (SCFT) was set up in 1999 as part of the transition to privatisation and thence to a co-operative. Farmer members owned and managed the SCFT, and the commercialisation of coffee was put on a better footing, restoring confidence amongst farmers (ibid). In 2007, after consultation with members, the SCFT became the fully fledged co-operative union MZCPCU. Up to 2013, MZCPCU had consolidated 3000 members in 6 highland co-operatives, of whom 25% were women, and with the help of Twin Trading, had achieved Fair Trade certification, 4C (corporate responsibility) verificationFootnote 17 and was seeking organic production status. MZCPCU’s web-site notes that it aims to enable every member to achieve decent accommodation, food security, adequate warmth and education for members’ children.Footnote 18

MUSCCO has a somewhat longer history (starting business in 1980) and had about 98,000 members in SACCOs in 2014.Footnote 19 MUSCCO states that its overall mission is ‘To facilitate and promote on a sustainable basis, the growth and development of Savings and Credit Cooperatives (SACCOs) through provision of a diversified range of financial, technical, administrative and other support services and products.’Footnote 20 SACCOs can be community or occupation based (for example farmers, teachers, civil servants, police).Footnote 21 The main purposes of loans include: seed, fertilisers, school fees, small business investments in both rural and urban areas, and gaps between spending needs and pay day (M5, 2013). MUSCCO also has a range of services: marketing, administrative support, advocacy, capacity building and financial intermediation.Footnote 22 Since 2013, MUSCCO has also had a very strong gender policy, although there are considerably more male than female members. In 1985, 22% of the membership were women, in 2002 their share increased to 25%, and in 2013, 35% of members were women. The following sections first look at economic control, then voice and agency, which, because of their close interrelationship, are taken together.

Economic Control

One of the purposes of co-operatives is to provide economic control to the membership. The experiences and perceptions of members are therefore a key dimension of their sense of inclusion and control. However, membership involves a financial contribution (a share in the co-operative) which is not always feasible for extremely poor people, and can pose an immediate dilemma for co-operatives, as noted in the discussion of literature above, particularly if they are embedded in commercial value chains. On the one hand, co-operatives need to be on a sustainable financial footing and to be able to provide services to members; not doing so can result in members selling their products elsewhere. On the other hand, a restriction on membership means that co-operatives are less inclusive.

For example, in the case of MZCPCU, the commercialization of coffee led to the co-operative union restricting membership to farmers with a minimum of 500 coffee trees (MZ1) to promote high outputs of coffee beans. However, for those who were members, the level of economic control was highly valued, as these statements from a mixed (male and female) focus group suggest (MZ9 2013):

‘what makes me to join co-operative is that I become the owner of the co-operative…it’s not controlled by other people, but you are the ones running it’

‘what made me interested in co-operative, is transparency, where you see right away from where we producing to where it sold’.

Other members noted the financial benefits:

‘when you are in a group, there is much security of your resources…it’s like insurance because you feel that you are not alone’;

‘when you are in a co-operative, you are guaranteed market, when on your own with your small quantities, it becomes difficult to sell your crop’.

Moreover, MZCPCU pursued a policy of enabling increased economic control by women coffee farmers by ensuring they had access to land, and by niche marketing of ‘women’s coffee’. Female membership rose from 368 in 2013, when interviews were first carried out, to 601 in 2017 (MZ13):

‘women are benefitting, at least now their homes are being changed. They have started helping in their families. They are having their own money…They can make decisions because they are being economically empowered so men after looking at this they also started sharing their land with their wives’ (MZ13 2017).

These women were able to diversify their sources of income and improve living conditions, and also had access to a micro-credit scheme set up by the co-operative union. Even so, the union still faced a common phenomenon in terms of female inclusion: ‘If we look at the percentage of women [in Mzuzu] who are economically empowered, it is very few; out of all the women maybe just 30%’ (ibid). The inclusion of youth has also proved challenging. It was observed that coffee is a slow crop which requires 3 years’ maturation before producing, while the younger generation want ‘quick results’ (MZ2 2017). A combined approach proposing a mix of activities, supported by access to credit, has had some success: ‘most of them have started coming in. So we are using those that are doing something, that are already benefitting…as role models to explain how they are faring’ (ibid).

The role of the co-operatives in supporting member knowledge and skills is an important aspect of the greater economic control experienced by members. Some members reported that, prior to being in the co-operative, they did not know if they were making a profit or loss. The lack of skills undermined their capacity even to sell their coffee to middle-men. Participants in the focus group noted (MZ9 2013):

‘I noticed that when you are in a co-operative, people get extension services especially technical knowledge, so I thought that if I join then I will have access to that knowledge’

‘I have been trained on sustainable way of producing my crop like using local pesticides that I can easily make from shrubs’

‘At first, we were just planting, now we have [been] taught the best agronomical practices’.

Low levels of literacy also triggered the co-operative union to develop long-term training processes. According to one co-operative union leader, training needed to be designed in a way that members were closely involved with the different activities conducted by the co-operative union and primary co-operative (i.e. marketing, budgeting, farming techniques, costs and so on). These aspects were in turn expected to reinforce commitment:

‘we try to let them understand that in bringing their produce to the co-operative they are actually helping their co-operative grow, whilst if they sell their produce to the middle men that’s the end of it’ (MK12 2013).

In part, the training and skills development were designed to meet the demands of contracts with buyers (Twin Trading) to upgrade production and processing to meet new standards. The pros and cons of certification have been documented (Blowfield 2004; Pirotte et al. 2006; Vanderhoff Boersma 2009; Humphrey 2008; 2009). In this case, the co-operative union was seeking to tap into niche export markets and gain premiums for ‘Mzuzu Coffee’; it was also aiming to address environmental issues through organic means (MZ3). Having to cover the co-operative administrative costs resulted in 60% of the price going to farmers and 40% going to operations. However, the 40% for operations was not sufficient to cover operational costs and permit further innovation (MZ5 2013). As noted by one manager (MZ2 2013): ‘those standards are biased towards social, environmental and economical…You’ve got to be really on top, so that you maintain those codes and standards’, while the CEO observed: ‘you really have to find a market that is good to make sure the farmers are happy…what makes people cling, the growers, is to see that every year they are getting a good price’ (MZ1 2013).

There were therefore contradictory aspects to economic control in MZCPCU. First, there was the balancing of member and management interests. Second, there was the demand on the extension work of the co-operative, and meeting the volume of demand for speciality coffee. Assistance with extension work was supported by the buyers and by donor agencies (MZ6 2013), while the co-operative union and Twin Trading aimed to provide farmers with technical advice on environmental issues and crop diversification (MZ3 2013). Third, however, engaging in commercial contracts was challenging for both management and members. As noted by the operations director: ‘We don’t pay cash at the same time we get their coffee…Those people who’ve got small productions look at that waiting as taking too long for them because they got immediate needs’ (MZ5 2013). However, by 2017, the co-operative union was able to work with forward contracts and pay an advance to producers when they most needed the cash to pay for fertilizers, school fees and help with harvesting (MZ5 2017).

Economic control by the membership is also a key dimension of savings and credit co-operatives (SACCOs) such as MUSCCO. The basic functions of SACCOs are that members can both save and take out loans at affordable rates. Saving is challenging for low-income populations, although it is certainly possible to build up savings from small contributions: ‘if provided proper structure, a proper mechanism for mobilization of savings, it can be done’ (M5 2013). The second function—lending—is feasible if a SACCO is adequately capitalised. Capitalisation depends on members’ savings, members buying shares in the SACCO, interest on loans and other sources of revenue.

SACCOs have other ways of enabling members to have economic control; For example MUSCCO provided training and assistance with business development, as well as leadership training and upgrading in ICTs. To build member networks, with the help of European co-operative support agencies, MUSCCO had explored the possibility of linking savings and credit facilities with activities of the National Farmers’ Union of Malawi (FUM) and National Smallholder Farmers’ Association of Malawi (NASFAM): ‘So we are trying to link all those…so that maybe there is a kind of chain, or we call it a triangle’ (M4 2013).

SACCOs may also include low-income professionals; For example one SACCO, identified here as SACCO 1, started in a research station and had about 3000 community-based members from different organisations (schools, churches, health sector, women’s clubs and so on). By contrast, another (SACCO 2) had donor-finance and grew to a membership of over 40,000 over a period of 5 or 6 years, with 20–30% tobacco farmers and 50% female participation either as individuals or in women’s groups (M3 2013). However, there were contradictions for both SACCOs with respect to economic control. As a SACCO 1 leader noted:

‘our culture gives more power to the husband. Now the woman accesses a loan…the one who will be in charge of managing that loan…will be the husband…in most cases it has contributed to defaulting…because the husband says no, I am the boss in this house so I can do anything’ (M6 2013).

In the case of SACCO 2, the tobacco farmers experienced a decline in prices over a period of 4 or 5 years (M3 2013). Moreover, the rapid expansion based on donor funds meant that its internally generated capital was weak. With this under-capitalisation, SACCO 2 fell into debt, and 50% of the members left:

‘It was the biggest SACCO in the country. Over 40,000 members, but its balance sheet had to be cut in half because of loan default… So you can imagine how painful it is to close a [sic] biggest co-op in terms of financial co-op. Creates a very big risk to the whole movement’ (M4 2013).

Economic control can be threatened in other ways; For example for the co-operative union to function, SACCOs need to pay for its services, but this in turn can create tensions. Other contextual issues can also intervene, for example: high inflation, loans for the chronically sick or death claims (M5 2013), malfunctioning of automatic deductions in savings schemes in the public sector and under-resourcing of government support. As with production and marketing co-operatives, good management and skills are crucial:

‘most of the co-operatives are small and for them to generate enough income so that they can implement…risk management processes becomes a very big problem…They cannot recruit a manager. Some of them cannot even computerise’ (M4 2013).

Economic control and inclusion can, in fact, work in opposition to each other; For example to maintain its capitalisation, in 2013, MUSCCO still depended for part of its administrative costs on external funding (M3 2013). By 2017, savings had increased but, paradoxically, the membership was declining (M4), possibly indicating that those who had been able to sustain their membership and save may have been better-off to start with. The challenge facing the co-operative was therefore how both to grow and diversify the membership.

Voice and Agency

There is a close interrelationship between the democratic processes in co-operatives and members’ agency, so we take both voice and agency together in this section, in particular looking at members’ roles in governance, leadership and agentive behaviours. An important difference between co-operatives and other forms of private enterprise is that co-operatives are owned and governed by their members, with each having a vote. Formally, the channel for voice is through members’ meetings and the annual general meeting, which also elects the board. However, contextual dimensions also come into play; For example in the case of MZCPCU, material success was considered an important factor in electing leaders, as well as literacy and being an upstanding citizen. A leader of a primary co-operative in MZCPCU observed (MZ12 2013):

‘somebody who wants to be a leader is to be having not less than 3000 trees as a man. Not less than 1500 as a lady…you cannot be just a leader and don’t have production…Secondly you have to be a member of that area…The third one, he should be literate…the fourth is that you should have no record of criminal cases…should be a person of good reputation in the village.’

As noted above, MZCPCU has an increasingly strong policy of inclusion of women members at all levels. The growth in women’s membership has led to greater respect on the part of management and within communities based on women’s abilities as farmers, their willingness to innovate and their increased incomes (MZ3, MZ5 2013). This growing recognition of women led to the appointment of a women’s co-ordinator, as well as a small number of other women in management (MZ13 2017), and to creating female role models who could recruit other women members (MZ3, MZ5 2013). A union leader noted:

‘We’ve got women on the board of the Co-operative Union and we’ve got women on the co-operative boards. These women…are successful coffee growers. Some of them have even bought cars so they are like role models to other women’ (MZ5 2013).

Another leader added:

‘We are taking those women now as models and we go out to show them these are womens [sic] that are doing better because they are now economically empowered…once a woman is empowered she is able even in the house to be able to say something, to voice out, even in the groups to make decisions’ (MZ13 2017).

In turn, this has reinforced women’s agency, with women making joint decisions with spouses to build houses and send children to school, even though this was also leading to some gender conflict: ‘these men are like “now these women are becoming rude, they can’t respect the husband” ’ (ibid). However, although the growing material success of women farmers had the potential to enhance their voice, there was still a relatively small number of women in management within the co-operative union (MZ13 2017).

Just as economic control within co-operatives requires education and training, so is the case of voice and agency when it comes to negotiating co-operative development strategy. MZCPCU managers noted that a gap in the experience, education and training of members compared with that of the co-operative union leadership could give rise to anxieties on the part of members (reinforcing the hierarchy/democracy tension alluded to by Bijman et al. 2016). Although the union leaders had considerable agentive experience through managing organisations, negotiating with government and ministry officials, and dealing with external organisations such as Twin Trading, the agency of farmers presented some challenges:

‘if we realise a lot of money from this, it will be used to run this organisation for the overheads…so that the farmer gets most of what is realised on the market….however if you go to the farmers, they are afraid – they think we can abandon them – but that’s not the idea…it’s just a matter of training them to change their attitude’ (MZ8 2013).

Communication between the co-operative union leadership and the primary co-operatives underpins the democratic structure and the ability to exert both voice and agency (Simmons and Birchall, 2003). In this case, one leader observed:

‘somehow you will find that there is a break in communication…Maybe we haven’t done much of our homework…they don’t bring back the feedback to the lower level institutions. So it demands a bit of extensive training’ (MZ2 2013).

In the case of MUSCCO, there were different issues for homogeneous and heterogeneous SACCOs, in terms of shared voice. As noted by one of the leaders of (heterogeneous) SACCO 1:

‘there is open membership. We don’t say, we have enough from the teaching profession, or we have enough from the farmers, no. Everything is done on [an] open basis’ (M6 2013).

However, it is not clear whether and how the composition of membership is reflected in governance and leadership (taken as proxies for voice and agency). First, of its nearly 98,000 SACCO members, approximately 33% were women,Footnote 23 with a greater proportion of women in community-based SACCOs in rural areas, while urban SACCOs tended to be employment-based with greater male participation. MUSCCO has also been targeting youth in schools to become members of Village Savings and Loans Associations (M4 2017). However, from interviews in 2013, it was unclear to what extent women had a role in leadership either in specific SACCOs or in the co-operative union. There was also unevenness in the quality of leadership, which in turn affects democratic processes and member participation. As noted by one MUSCCO manager:

‘leaders when they come into leading the SACCO, sometimes they might not understand the concept, they would think that maybe when they become leader, then they…will be getting more loans… When they get in, they find that is not the case’ (M4 2013).

Such issues were compounded by levels of literacy and understandings of the purpose of co-operatives: ‘it means you have to take longer time to prepare them …they have to work themselves to make it work’ (M5 2013). Furthermore, SACCO leaders were not necessarily equipped to keep track of the changing regulatory and policy environment (M2 2013). Although MUSCCO provided training, the democratic governance structure of co-operatives meant that ‘by the time a director fully understands the operations of the co-operative, they are ready to go’ (M4 2013). Training and practical experience have the potential to enhance the elements of voice and agency in the SACCOs, however, leaders were very aware that ‘capacity building is a gradual and permanent process’ (M5, 2013).

Reflections and Conclusions

We have put forward a notion of inclusive development as a bottom-up process built on economic control, voice and agency. Such a process, we argue, can be supported by organisational platforms such as co-operatives, although not without contradictions and challenges. We are adding to literature that focusses primarily on addressing inclusive development through social and economic policy by examining more closely the kinds of platforms that can actively promote deliberative and participatory action by low-income and disadvantaged people. We are also adding to the debates about inclusion in the co-operatives and development literature by demonstrating that inclusion is not a once and for all outcome but an ongoing process of change in the development of co-operatives themselves and their relationship to the social and economic contexts in which they work. The above observations from fieldwork, therefore, present a somewhat complex and contradictory picture, while simultaneously underlining the processual nature of inclusive development through platforms such as co-operatives in a developing capitalist economy. There are several ways in which the co-operatives reflect challenges in the wider social context in Malawi: issues around gender relations, youth inclusion, levels of literacy and skills needs for organising and management, and the tension between needing to be financially viable and the inclusion of the very poor or vulnerable sectors of the economy. These are commonly cited issues with respect to co-operatives in development; however, the data from the two case studies also affirm the leadership’s awareness of inclusive development as a process and of the need to make and support changes over time.

Many analyses suggest policies that could be pursued by governments to create an enabling environment both for co-operatives and for greater inclusion of the very poor. While acknowledging that this is an ongoing challenge, our concern in this article is to demonstrate that inclusion is a process which requires both bottom-up collective action (‘little-d development’) and enabling policies. There are inevitable tensions between sustaining the economic viability of co-operatively based business and the aim of including the most disadvantaged. Addressing these tensions is a task for both co-operative leaders and government policy. In the Malawian case, a new national apex organisation, established in 2015, aims to ‘facilitate and co-ordinate [the] promotion and of development and growth of sustainable, independent and autonomous co-operatives… [T]he co-operatives allow the indigenous Malawians, those who have no say, to participate in our economy…We are a link between the movement and government, and even other players’ (MA1 2017). Such linkages are a critical and evolving element of the wider context. However, as noted by a MUSCCO leader in 2013, a strong apex organisation able to both negotiate with government and help shape inclusive development in Malawi requires strong co-operative unions within it. A question then is how to combine strength in business terms and economic control with strength in inclusive membership, voice and agency.

Inclusive development as a bottom-up process driven by low-income people has the potential to enable disadvantaged individuals and communities to meet social needs as well as create avenues for participatory decision-making and collective action. This is in line with arguments made by Münkner, Birchall, Hannan and others: In other words, addressing inclusion lies with how members use and shape their co-operatives, and how they address their governance issues, rules and processes for inclusion and exclusion, while pressing and working with governments to support co-operatives in challenging settings. There is therefore no simple or straightforward solution to greater inclusion. Our case studies show how economic control, voice and agency in co-operatives rely on access to knowledge, expertise and adequate communication across all levels of membership and management. Co-operative unions and apex bodies can offer avenues of communication, particularly where poor and low-income people have little leverage, and in this sense, can create bridges between communities and supporting agencies. Co-operatives (primary, secondary and apex organisations) can therefore contribute to better understanding inclusive development as an ongoing, deliberative process, where higher-level policies are also needed. Such a process is not an ideal one: it is based in everyday experience, involves contradictions and challenges, and requires a long-term perspective.