Microfinance and the business of poverty reduction: Critical perspectives from rural Bangladesh

Introduction

Poverty is big business. Even in the United States, one of the richest countries in the world, the poverty industry is worth about $33 billion a year comprising payday loan centers, pawnshops, credit card companies and microfinance providers who generate business from the poorer segments of the population (Rivlin, 2010). Among the so-called developing and least developed countries millions of people continue to face crippling poverty. ‘Ending poverty in all its forms everywhere’ is the first of 17 Sustainable Development Goals set by the United Nations. In absolute terms at the global level there are currently between 1.2 and 1.5 billion people still living in extreme poverty and 162 million children still suffering from chronic under-nutrition, a figure the UN deems ‘unacceptable’ (United Nations Development Programme, 2014).

Microfinance, or the provision of small loans to the poor with the aim of lifting them out of poverty, is a key poverty reduction strategy that has spread rapidly and widely over the last 20 years, currently operating in more than 60 countries (Bateman, 2010). According to many researchers and policy makers, microfinance encourages entrepreneurship, increases income generating activity thus reducing poverty, empowers the poor (especially women in developing countries), increases access to health and education, and builds social capital among poor and vulnerable communities (Khandker, 2005; Westover, 2008). Studies of market-based measures to alleviate poverty are also gaining considerable traction in the management literature where scholars have developed concepts like ‘base-of-pyramid’ and ‘creating shared value’ to address what businesses can do to alleviate poverty and enhance social welfare (Porter and Kramer, 2011; Prahalad, 2004).

However, more recently concerns have been raised about the real value and impact of microfinance. In the last few years ‘microfinance meltdowns’ have been reported in Morocco, Nicaragua, Pakistan, Bosnia, Mexico and Lebanon, and most dramatically in the Indian state of Andhra Pradesh when the entire microfinance industry collapsed in late 2010, which was the context of the quote by the then Indian minister mentioned above (Bateman and Chang, 2012). More disturbingly, inability to repay microfinance loans has also been linked to ‘hundreds of suicides’ among borrowers in India (Associated Press, 2012) and organ trafficking in Bangladesh (BBC, 2013). Such concerns raise important questions: does microfinance enable entrepreneurship among impoverished communities that can lift them out of poverty? Is it possible, as some critics claim, that microfinance instead of alleviating poverty actually serves to exacerbate poverty in particular contexts (Bateman, 2010; Karim, 2008)? If so, how? How do the receivers or ‘clients’ of microfinance cope with rising debts that result from incurring microfinance loans? And how does group borrowing influence social relations between individuals in the group?

To help answer these questions, we report the results of an ethnographic study of microfinance in three villages in rural Bangladesh that have been targeted by microfinance organizations aiming to reduce poverty in the region by promoting entrepreneurial activity. Our results challenge existing theory and research regarding the role and impact of microfinance – that it generates income through entrepreneurial activity, empowers women and builds social capital in poor communities. Our aim in this article is to change the conversation about microfinance and market-based development by changing the lens through which the problem of poverty reduction programmes is seen: not from the perspective of the providers of microfinance institutions (MFIs) or the many government and non-government organizations (NGOs) that develop and implement microfinance initiatives, but from the perspective of the receivers of microfinance, especially those that live in extreme poverty.

Our study makes three contributions to the literature. First, at the individual level we provide an empirically grounded narrative about the lived realities of poverty and describe the experience of poor communities with microfinance. Our ethnographic study enables us to develop a grounded theory of the vulnerability dimensions of poverty. The majority of the literature on microfinance has focused on the ‘supply side’ of the equation. Research questions have tended to be donor driven and the emphasis has been on the reach of microfinance initiatives (Mosley and Hulme, 1998), the high repayment rates on microloans (Matin et al., 2002), the role of NGOs in implementing microfinance projects (Baruah, 2010), the kinds of microenterprises that emerge (Datar et al., 2008) and the organizational processes of microfinance providers (Galema et al., 2012). This article provides a different account: how microfinance affects the daily grind of the rural poor, the choices they have to make to stay financially afloat and the economic, environmental and social consequences of these choices. Such rich empirical accounts are rarely found in the literature, especially from developing regions such as Bangladesh where access can be difficult. Much of the rhetoric of ‘empowerment’ is from a top down perspective of those that do the empowering with little or no attention being paid to the increased vulnerabilities faced by those living in extreme poverty arising from market-based poverty reduction initiatives like microfinance. Our study provides such a bottom up perspective.

Second, findings from our study contribute to the literature on the role of social capital in poverty reduction. Emerging research on base-of-pyramid (BoP) approaches to poverty reduction and empowerment of poor populations suggests that market-based initiatives directed at impoverished communities can leverage their social capital to develop capabilities that could lift them out of poverty (Ansari et al., 2012). Our study identifies the boundary conditions of social capital creation through market-based initiatives by highlighting processes whereby social capital can be undermined by market-based measures like microfinance. Third, our study contributes to the literature on building inclusive markets by describing accounts from voices that tend to be excluded in the debate on inclusive growth at the institutional level. NGOs are key actors that fill the ‘institutional voids’ in rural areas of developing countries (Mair et al., 2012) and our study complements emerging research in the area by problematizing the role of NGOs as institutional agents of poverty alleviation (Khan et al., 2010). In the sections that follow, we discuss the emergence of microfinance and examine its theoretical basis as a poverty reduction strategy. We then describe our ethnographic study of communities in three Bangladeshi villages and analyse their experiences of microfinance. We conclude by discussing implications of our findings and providing directions for future research.

Poverty and vulnerability

While there is no universally accepted definition of poverty, economic dimensions of poverty based on income and consumption data have generally been used to measure poverty levels. For instance, the World Bank defines two thresholds of poverty – the ‘extreme poor’ who live on less than $1.25 a day and the merely ‘poor’ who live on less than $2 a day based on consumption per capita (Banerjee and Duflo, 2007). While the dollar-a-day figure may be a useful heuristic for researchers and policy makers it does not capture the lived realities of the poor – feelings of powerlessness and vulnerability risks, for example, or poor nutrition and health arising from sustained deprivation, or gender differences in poverty (Chakravarti, 2006; Ravallion, 2003). Economic measures of poverty may reflect the structural aspects of poverty but do not capture the cultural, social and psychological dimensions of poverty and more importantly precludes any kind of agency to the poor by ignoring their survival strategies, social relations and practices of resistance (Arora and Romijn, 2012).

Some studies have identified qualitative indicators of poverty such as vulnerability, deprivation, helplessness and deficiency that arise from income poverty and the inability of the poor to leverage resources required to fulfill their basic needs (Bradshaw, 2007; Chambers, 1995; Chakravarti, 2006). However, apart from offering estimates of the number of people living on $1 or $2 a day, research assessing the effectiveness of poverty alleviation measures does not provide useful insights into the lived experiences of the chronic poor or the effectiveness of poverty alleviation strategies in reducing the qualitative aspects of poverty such as vulnerability, deprivation and helplessness. Our study attempts to address this gap. Our analysis provides an expansive concept of chronic poverty that takes into account feelings of vulnerability and associated risks experienced by the extreme poor. Drawing from the extant literature, we define poverty as a process whereby people are subject to sustained physical, social, economic, political, psychological and/or spiritual deprivation that gives rise to any combination of physical weakness, perceived isolation, and feelings of ill-being, vulnerability and powerlessness (Banerjee and Duflo, 2007; Chakravarti, 2006; Chambers, 1995; Ravallion, 2002).

Vulnerability has been conceptualized in the literature in a number of ways. In the economics and development literatures vulnerability is defined as the ‘probability of risk today of being in poverty or to fall into deeper poverty’ (World Bank, 2012). Thus, vulnerability is the probability of experiencing a future loss in welfare and the prospect of individuals or households becoming poor in the future or the prospect of continuing to be poor if they are currently living in poverty (Christiaensen and Subbarao, 2005; Zhang and Wan, 2006). Vulnerability to income shocks along with deprivations in health and nutrition can be considered to be part of an expanded poverty concept (Morduch, 1994). The World Bank identified vulnerability as an ‘important consideration for poverty reduction policies’ because vulnerability risks influence household behaviour and coping strategies.

Vulnerability is a more dynamic concept than poverty in the sense that it describes processes and events that lead populations to fall in and out of poverty. Moser (1998: 3) defines vulnerability as ‘insecurity and sensitivity in the well-being of individuals, households and communities in the face of a changing environment’. The decline in welfare can result from several causes: natural disasters, environmental damage, economic shocks or social and political exclusion. Vulnerable populations also differ in their resilience to risks and their capabilities to manage risks are constrained by their inability to earn a living as well as by the social and psychological effects of deprivation and exclusion (Moser, 1998: 4). In the context of poverty, vulnerability can be policy induced (e.g. increased risks arising from government imposed austerity measures) or market induced (e.g. rising indebtedness as a result of increased borrowing; Glewwe and Hall, 1998). Vulnerability is also related to asset ownership: people with more assets are less vulnerable and loss of assets leads to greater vulnerabilities.

In his influential work on development, Sen (1983, 1985) argued that poverty reduction strategies should focus on developing capabilities among the poor to enable them to leverage economic opportunities. Poverty has less to do with utility or choice but is seen as deprivation of capabilities to participate in economic activity or political processes. How the poor can develop the capabilities required to escape poverty is of course a significant challenge. Prevailing economic wisdom argues that asset accumulation and access to capital are key factors in developing capabilities among the poor (De Soto, 2003). The popularity of market led approaches to poverty reduction such as microfinance and BoP strategies rests on the assumption that these strategies can deliver the required capabilities. Ansari et al. (2012) argue that leveraging the social capital that exists in poor communities may enable them to build the capabilities needed to access resources from external groups or institutions. They argue that a BoP approach has the potential to both retain existing social capital in impoverished communities while enhancing their social capital through accessing resources from external networks and groups. However, there is little empirical research that supports this assumption.

Poverty and social capital

While poor communities lack economic assets and financial capital, their social relations play a key role in sustaining their livelihoods. Rural communities in subsistence economies are often characterized by norms of collectivity, reciprocity, sharing of community resources and extended kinship ties that are essential for their survival (Scott, 1976). There is some evidence that suggests communities with strong social networks are better able to deal with poverty and vulnerability (Moser, 1998; Woolcock and Narayan, 2000). These networks generate social capital, which reflects the general goodwill and resources that arise from networks of relationships in a community (Adler and Kwon, 2002; Putnam, 1993). Social capital is a multidimensional concept comprising of structural, relational and cognitive components (Nahapiet and Ghoshal, 1998). Network configurations and associations constitute structural social capital. Networks and ties are fostered through communication and shared meanings, which constitute cognitive social capital. Relational social capital refers to the extent of trust, reciprocity and cooperation between individuals in a network.

Putnam (1993) also distinguished between two types of social capital – bonding social capital, which is characterized by horizontal relationships based on reciprocity, trust, shared norms, values and beliefs that promote solidarity between individuals within a network enabling them to ‘get by’; and bridging social capital, which reflects the ability of individuals in a network to gain privileged access to resources and information from external networks in an attempt to ‘get ahead’ (Woolcock and Narayan, 2000). Many poor rural communities face difficulties in accessing resources from external groups because they do not possess adequate bridging social capital. The problem is compounded in developing countries where the state is often unable to provide resources and opportunities to impoverished populations. Emerging research suggests that non-state actors and institutions can fill the void left by the state to promote business ventures and entrepreneurship that would empower these ‘BoP communities’ by enhancing their social capital and enabling them to escape the poverty trap (Ansari et al., 2012; London, 2009; Mair and Martí, 2009).

BoP advocates argue that microfinance can deliver economic development and social empowerment by creating bridging social capital that allows impoverished individuals to access external resources and networks. However, owing to structurally unequal power relationships between finance providers and borrowers, microfinance can also create new dependencies on external institutions while adversely impacting social relationships of trust and reciprocity thus eroding bonding social capital (Ansari et al., 2012). Moreover, BoP approaches to poverty alleviation lack sufficient theoretical development and empirical support. Critics have questioned the role of business in poverty alleviation arguing that BoP approaches continue to be informed by win-win assumptions that privilege business rather than enhance social welfare of BoP communities (Karnani, 2007), obscure unequal power relations (Arora and Romijn, 2012) and serve to depoliticize the economic sphere by advocating solely market-based measures to alleviate poverty (Banerjee, 2008). Market-based approaches such as entrepreneurship and BoP ventures are a reflection of particular rationalities that are based on ideological assumptions of individualism and choice that are sometimes incompatible in communities characterized by sharing, reciprocity, kinship ties and collectivism (Adler and Kwon, 2002; Ansari et al., 2012).

From its humble beginnings in the late 1970s and early 1980s in rural Bangladesh, microfinance today is a global multi-billion dollar industry. The United Nations declared 2005 as the ‘International Year of Microcredit’ calling for ‘constructing inclusive financial sectors that strengthen the powerful, but often untapped, entrepreneurial spirit that exists all over the world and a new wave of micro entrepreneurship, giving poor and low-income people a chance to build better lives’ (United Nations, 2004). The Grameen Bank, formally established in 1983, was the first organizational entity to offer collateral-free microcredit to the poor based on early experiments with providing small low-interest loans by its founder Dr Muhammad Yunus. These early experiments of providing microcredit to the poor resulted in two intriguing findings: first, the repayment rate was exceptionally high despite the cash poor clientele and, second, women proved to be significantly better at repaying than men (Yunus, 1999). Further investigation revealed that one of the reasons for high repayment was access to family and community networks to repay loans. A key driver in repayment was reputational damage and bringing ‘shame’ to the family (Bateman, 2010). Reliance on family and community networks for repayment led to another innovation – the creation of ‘solidarity circles’ (kendra) where groups of women rather than individuals would be responsible for ‘helping’ an individual to repay loans if the borrower was facing financial hardships. Thus, the ‘social’ entered the microfinance discourse in the form of social capital, or more accurately ‘social collateral’ (Bateman, 2010).

Proponents of microfinance claimed that offering credit to poor communities would provide a source of additional income and employment as well as access to low-interest loans, enabling poor communities to escape from the clutches of local moneylenders and loan sharks and their exorbitant interest rates. The availability of financial services to poor segments of the population could help them deal with vulnerabilities arising from poverty, while empowering women who could find few business opportunities because of patriarchal systems of control. Finally, microfinance could help build social capital and solidarity in impoverished communities because MFIs promoted group lending and were willing to accept ‘social solidarity’ as collateral (Bateman, 2010; Matin et al., 2002).

Despite scores of reports and scholarly papers addressing the impact of microfinance, no clear picture emerges about either the sustainability of MFIs or its impact on poverty alleviation (Armendáriz and Morduch, 2005). While some studies claim that microfinance increased disposable income and enabled poor families to move out of poverty (Khandker, 2005), other studies found no evidence of such a relationship (Kah et al., 2005; Morris and Barnes, 2005) and some even found a negative impact (Bateman, 2010; Dichter and Harper, 2007; Karim, 2011; Roodman, 2011). Even the World Bank, a powerful proponent of microfinance, appears to take a more cautionary stance in recent years, concluding that ‘more research is needed to assert whether there is a robust and positive relationship between the use of credit and household welfare, including moving out of poverty’ (World Bank, 2007: 104).

Assessment of the impacts of microfinance reveal mixed findings – both about its effectiveness in poverty reduction and about its sustainability as a financial model. What is missing from extant accounts of microfinance is the subjective experience of the poor, especially the extreme poor and how they negotiate the everyday grind of poverty, their financial decision-making process and outcomes, their accounts of vulnerability and disempowerment, their experience as clients and users of microfinance, their interactions with microfinance providers, and the social and economic outcomes that result. It is to these subjective formations of microfinance that we now turn in an attempt to develop a more grounded theoretical approach.

Methods

We adopted a micro-level ethnographic approach to understand subjective experiences of poverty focusing on individual and household narratives. Ethnographic accounts enable us to develop narratives that ground key characters spatio-temporally while also highlighting their interactions with other protagonists and antagonists, thus revealing how dominant narratives and their counternarratives emerge and evolve (Barnberg and Andrews, 2004; Boje et al., 2016). Our data collection focused on understanding subjective experiences about the qualitative aspects of poverty, such as feelings of vulnerability, deprivation and helplessness. We also wanted to understand how the availability of microfinance influenced the lives of chronically poor individuals and households. By focusing on the lives of the receivers of microfinance we provide a more complex analysis of the experiences and lived realities of poverty than what can be understood from household consumption figures, interest rates, repayment rates and loan disbursement figures. Our ethnography involved observations (of borrower meetings), focus groups and in-depth interviews and was conducted by two teams of researchers and their locally based associates.

The study setting and sample

Fieldwork was conducted in three villages in Bangladesh in the Matlab district. The region comprises more than 150 villages and has few roads – access is mainly via small boats. Agriculture (mainly rice and jute) and fishing are the two main occupations. Microfinance activity in the region dates back to the mid-1990s, although there are very few reliable sources that document the extent and use of microfinance. Chronic poverty in the region meant that Matlab became a focal point for microfinance programmes operated by key NGOs such as the Bangladesh Rural Advancement Committee (BRAC), Grameen Bank, Association for Social Advancement (ASA) and Krishi Bank. All the individuals and households that constituted our sample can be classified as ‘chronically and extreme poor’ based on the $1 and $2 a day consumption thresholds.

The cultural, social, political and economic landscape of Bangladesh posed significant challenges to us as western researchers and we did our best to ensure that culturally constructed rules around distinctive patterns of behaviour, norms, values, traditions, laws and customs were taken into account before the sampling process commenced. For example, we engaged with the Matlab district government official and village elders to gain permission to visit the villages and collect data. We visited a number of villages accompanied by local leaders who introduced us to the village elders from whom we subsequently obtained permission to interview community members to collect research data. The sampling frame adopted for the research included people from three villages in the Matlab district. Following Glaser and Strauss (1967), we selected a theoretical sample to understand subjective experiences of microfinance clients. This process of data collection is ‘controlled by the emerging theory’ (Glaser and Strauss, 1967: 45), rather than the requirements of statistical sampling and is consistent with the exploratory and descriptive purpose of the research.

Data collection was done in three phases across three villages (denoted as A, B and C in Table 1). This allowed us to document the experience of borrowers with microfinance over time. A total of 56 in-depth interviews and six focus groups were conducted with borrowers across three villages during the first two phases. Consistent with our ethnographic approach, two research associates lived in the area for six months, immersing themselves in and familiarizing with the context while conducting interviews and focus groups with villagers. A reflective, progressive ethnographic diary was kept, where incidents and learnings were recorded. For example, we observed that women were more likely to communicate freely when they were doing their normal daily chores like walking to the well to collect water or washing clothes in the river, or sitting together over tea with only other women present; whereas the men became more communicative when sitting in the local tea shop after finishing work for the day. We arranged our interviews and discussions with participants in accordance with their daily routines. Our field diary had more than 150 pages of observations, reflections and notes, which we included in our data analysis.

OPEN IN VIEWER

Table 1.

Sample groups.

Sample	Focus groups	Interviews
Village A	1 female group with 6 participants 1 male group with 6 participants	20 (5 females and 15 males)
Village B	1 female group with 10 participants	18 (13 females and 5 males)
Village C	1 female group with 8 participants 1 male group with 6 participants 1 male group with 7 participants	16 (12 females and 4 males)
Local research associates		1 female1 male
Follow up field visit	1 focus group with 8 housewives and 3 husbands	17 (11 males and 6 females)
TOTAL	7 focus groups	73 interviews

Data analysis

The primary purpose of the analysis was to make sense of the themes emerging from the ethnographic data, which consisted of transcripts of interviews and focus groups, our reflective diary, as well as personal notes and observations. We followed a two-stage approach in analysing data. The first stage involved coding the data from the transcripts of the interviews and field research notes. The second stage involved describing and interpreting patterns in the data that emerged from the day-to-day lived reality of the research participants and relating the themes that emerged with the literature on vulnerabilities and social capital. Based on our interviews we also developed narratives that described the experiences of borrowers throughout the process of loan procurement to repayment, how the money was spent and the consequences of non-repayment. The software package Leximancer 4 was used for initial coding of the data. Leximancer is a data mining programme that identifies key concepts within the text. Concept mapping does not merely use keywords but represents the data in visual forms by focusing on clusters of related terms and their relationships (Smith and Humphreys, 2006). Leximancer provides a schematic diagram that displays five items of information about the text: the main concepts discussed in the document set, the relative frequency of each concept, how often concepts co-occur within the text, the centrality of each concept, and the similarity in contexts in which the concepts occur. The connection between concepts is measured by examining how often two concepts are discussed within the same passage of text in order to establish relationships and intra-textual interactivity. Some initial codes that emerged from Leximancer coding included concepts like debt, repayment, NGOs, money, family, income, shame, crops. We then made sense of these themes by relating them to the narratives we had constructed as well as to theoretical perspectives from the poverty literature.

We used an iterative process to infer second order codes and aggregate theoretical dimensions (see Figure 1). For example, in arriving at the dimension of economic vulnerability we looked at creating first order codes from an analysis of the transcripts and output from the Leximancer data mining software programme. The concept ‘money’ was related in varying degrees of strength to other concepts like ‘borrow’, ‘loan’, ‘pay’, ‘NGO’, ‘family’, ‘debt’, ‘land’, with ‘borrow’, ‘loan’ and ‘debt’ having the maximum co-occurrences with ‘money’. We then searched in our transcripts for references to these concepts from all respondents, which allowed us to interpret them as elements of a second order theme that reflected rising indebtedness. Based on these second order themes, we then constructed narratives that reflected participants’ experiences of rising indebtedness. We then went to the vulnerability literature where we found references to risks arising from debt, asset loss and non-repayment of loans, which helped to validate our empirically developed codes. We returned to our data to find cases where households lost their assets in order to repay loans. We followed a similar approach in interpreting themes that reflected social relations. Based on this iterative process we were able to construct second order codes that led to three aggregate dimensions of vulnerability: economic vulnerability, social vulnerability and environmental vulnerability.

OPEN IN VIEWER

Figure 1.

Analysis scheme.

Findings

In giving meaning to the data, we drew on theoretical perspectives from the poverty literature, particularly qualitative indicators such as vulnerability and social capital. In reading and analysing the transcripts of our interviews, observations and focus groups, vulnerability and social capital emerged as key themes that described people’s experience of poverty.

Measuring vulnerability as a probability risk function does not reveal the complexities of the phenomenon. The space of vulnerability comprises both internal and external factors. Internal factors include defenselessness, inadequate capacity to mobilize resources to cope with hazards, and potentiality, the risks of severe consequences (Watts and Bohle, 1993). External factors include risks of exposure to hazards (Chambers, 1989). Vulnerability does not just result from poverty: it can also ‘reinforce the income processes which lead to poverty and further diminish the expected welfare of the poor’ (Morduch, 1994: 225). Our analysis reveals the multidimensional nature of vulnerability, in particular the economic, social and environmental dimensions of vulnerability. Our findings indicate that the method of distribution and use of microfinance in our research sites, instead of alleviating poverty actually served to exacerbate poverty and increase vulnerabilities rather than create empowerment for the majority of borrowers. Tables 2, 3 and 4 present data in the form of illustrative quotes that describe the heightened vulnerabilities in these communities. Three dimensions of vulnerability emerged from the data: economic, social and environmental vulnerabilities as we discuss below.

OPEN IN VIEWER

Table 2.

Economic vulnerability.

Rising indebtedness	Loss of land assets
Farmer 1: Cannot afford to have proper meal each day. Need to rely on others for food. Despite hunger, I had to pay instalments NGO will never care why I can’t pay.	Farmer 2: Took loan from Bru-tangail, 20,000 taka, 550/week, 40 weeks. Agrani Bank, gave land document three years back, 12000 taka, still need to pay, interest is increasing, have to sell land, always on loan, no peace to farmers.Farmer 3: I had to sacrifice my land to pay the loan but I can spend few good days if I borrow.Farmer 4: These NGOs are not good for us, our poverty level is increasing but we have no other option. Does not matter if you die, you have to pay the instalment or sell house or land? The poverty is becoming ‘boundary-less’.Farmer 3: My in-law in Daudkandi has lost all his land because he borrowed too much from many sources. Last year, he made huge loss due to bad weather and storm. He had sold all his land to pay the loan. He is still in huge debt. It is impossible to pay loan growing crops.Woman 14: I used to have a house and some land. My husband fell sick. Used the loan for medicine. Couldn’t repay. Now we lost house and land. I live in my sister’s place and work in her house.
Woman 7: Say, you can collect some of the vegetable (leaves) from the roadside and eat (cook) them, that money you pay to the NGO like this.
Woman 11: They ask about: what you do with this money? – We tell them whatever they want to know – We have several loans, we borrow from one and pay to others – We have few loans – When we are in deep trouble, borrow 10–20,000tks, we eat less and pay back.
Woman 9: I had to tell lie to the agents for borrowing. If I tell them about the true reason then they will not give me loan. The true reason of borrowing money is to pay other loan that I had borrowed from other NGOs.
Woman 3: I had borrowed the money for agriculture specifically for cultivating maize. At present, I have no source of income. My husband was a rickshaw puller, his rickshaw got stolen, since then he is unemployed. Whenever we need cash, we need to borrow because our daily or regular income is very low. I have not benefitted much from the loans because the crops were not as good as expected. So, we are not being able to repay the loans. If we want to restart, we will have to loan again. Oh, the loans have become burden to us.
Woman 12: Every baree (household) took loans from NGOs. Not a single household has any earnings to pay it back.

OPEN IN VIEWER

Table 3.

Social vulnerability.

Surveillance	Disempowerment	Shaming
Respondent 1: My cousin died. NGO people came to me and told me to pay the money on behalf of the deceased person as he cannot pay. They watch and try to find out who in the family can pay.Respondent 2: The NGO workers never pay heed to whatever problems you may have . . . they demand and recover the instalment amounts by hook or by crook, if necessary they will make you borrow from relatives to pay back the instalment. Despite hunger, I had to pay instalments. NGO will never care why I can’t pay.Farmer 5: We can trick them (NGOs) by not staying at home when they come to collect money but the female has to stay at home, so they catch them.Woman 12: The men run away when NGOs come to collect money. They know we women are in the house. They stay till it is dark so they can catch the men when they come home.Woman 5: When we can’t pay the NGO enter our house to see if they can take anything. Once they took my only water bucket. And my sheelnodi (mortar and pestle). They say once I repay the loan they will return it. Once they even brought police to my house.	Woman 4: We are dying out of this vice (dowry). We cry for the money spent on the marriage, don’t cry for our daughters being sent off. Suppose I need to get my daughter married. I have to borrow money 10,000 to 20,000tk. Marriage ceremony lasts for two days – need lots of money. Need to spend a lot, don’t cry if mother dies, but cry for daughter’s marriage, give everything to daughter. Sacrifice life for daughter’s marriage.Woman 4: For a granddaughter’s wedding we borrowed, tk 30,000 cash and 10 grams of gold. Another one, tk 20,000 in cash and 5 grams of gold. Have to marry off the girls, otherwise the husbands will harass them, there will be unrest. No peace. Have to pay in the hands of father-in-law, not the groom. Otherwise there is social conflict.Farmer 1: I had to give marriage to my daughter and need money. I borrowed too much. Now I have huge loss. But I had to give marriage to my daughter.Woman 19: So this credit will destroy the culture and country, everybody is concerned about increasing borrowing but never think of earning.	Male, Rickshaw driver: NGO should not use bad words to the housewife in front of other neighbours. They come in every Mondays and Wednesdays to collect weekly payment. If you ask for extension they get very angry and use bad language. This is shame to me and my family, when the toll collectors keep sitting in my home for the whole day and use bad language. I am a human and I cannot drive rickshaw the whole day. I had to borrow from others to pay these immediate payments.Woman 17: My father borrowed money from NGOs to help my husband, and now they are also in trouble. They are losing their face too.Woman 11: Often they take us to their NGO office and make us sit there all day. Showing us in front of villagers. It is kind of insult and shame. We lose face.Woman 7: The NGOs money is must pay money. Even though we don’t give collateral to NGOs against borrowing, but their swearing in front of other villagers makes you to pay the instalment in each week.The NGO people swear using my mother’s name. It is too dirty and shameful.

OPEN IN VIEWER

Table 4.

Environmental vulnerability.

Climate change impacts	Shifting cultivation patterns
Farmer 1: The winter season is becoming shorter and shorter. We have had rain water logging since late October that has remained still till this early November, but look now, it is late November, supposed to be prime winter, but can you feel any cold . . . potato needs chilling cold, otherwise it won’t get enough food.	Farmer 4: The potato-maize is good combination but it requires high level of chemical fertilizer. Yes, we are giving sufficient fertilizer, but long maize root takes much natural nutrition from the soil. This condition is pushing us to apply more fertilizer next year. The great danger of this practice is that there will be no natural nutrition in the soil. We have to fully rely on chemical fertilizer.Retired farmer: The young farmers only know want to make money. They have been growing potato-maize every year. It takes high level of nutrition from the soil and left very little for the next crop. Every year they are applying more fertilizer. God forbids there will be no rice grown in this soil.Farmer 1: There will always be the natural disaster, hot wind and high price of fertilizer. These will cause loss in potato crop, and it will be so high that good maize will not recover the loss. I don’t know why they are doing this. They are not farmers, and I really do wonder why they are taking huge loan.Farmer 2: This year many families have taken new loan from new NGOs. Farmer 3 owns no land, and has rented 1 acre and trialled growing maize. Too expensive at the end, 700 tk kg seed, most of the farmers got lower yield. Last year the price of maize crop was higher than rice so people selected to grow maize for the first time. This simple calculation, many farmers follow that (grew maize for the first time) made a terrible loss and lost everything, due to bad weather and less demand.
Look at me I am wearing only a shirt on a winter morning and it is 9 in the morning. You will never get a good yield of potato if the winter is not cold enough. If I have to plant potato in December then when will I harvest maize? It will be all rainy and cloudy by the time of maize harvest, all too difficult to dry maize then.
Farmer 2 Village C: All we can do is pray to God to stop this devastating air, I have lost huge yields because of it, and it is increasing every year. Agriculture is no longer the simple way of life. You have to have sufficient cash in hand.

Interdependent vulnerabilities, debt spirals and erosion of social capital

For people living in extreme poverty our findings challenge the basic assumptions of microfinance: that it creates a new class of entrepreneurs enabling the poor to move out of poverty; empowers the poor, especially women in developing countries; and strengthens social capital. Instead, we found families faced increased vulnerability owing to rising levels of indebtedness leading to loss of land assets as well as erosion of social capital resulting from diminished bonding social capital. There was very little evidence of entrepreneurial activity taking place in the villages and most loans were used for income and consumption smoothing. Microfinance became an added element of risk in the communities we studied and there were few if any opportunities to diversify sources of income. We will discuss our findings and their theoretical implications by narrating a case of one microloan given to a group of women and show how interdependent vulnerabilities resulted in an increasing spiral of debt while damaging social relations between women borrowers and their families.

The story of Shonali and her family

Shonali lives in Village B and is a respected member of the community. Her husband Ali and two sons Kalu and Mukul are farmers. Shonali is the leader of a group of 20 women borrowers from Villages A, B and C and is the main contact person for NGOs who operate as MFIs in the area. NGO managers stay at her house when they visit the village for loan disbursement and repayment. In addition, maize traders also visit her to promote maize cultivation. The formal documentation and primary borrower-lender relationship is with Shonali and the NGOs who provide the finance for new businesses (in this case maize cultivation). As group leader, Shonali has the authority to recommend individuals in her group for accessing loans from NGOs. One such borrower, Mrs Delwas, borrows some money from the group fund. On closer investigation we found that it was Mrs Delwas’s husband who wanted the money but because the NGO does not lend to male household members, Mrs Delwas was pressured by her husband to take out a loan. The loan would be used for her husband’s cultivation of maize and potato crops. Shonali also used some of her own loan to pay for her husband’s and sons’ maize crop. Unfortunately, the crops failed and there was no income to pay back the loans. The NGOs pressure Shonali to pay back the group loan, who in turn puts pressure on Mrs Delwas. Their relationship becomes increasingly strained. Shonali also uses her son Kalu’s wife to obtain a loan from another NGO by providing a reference for her to the NGO. She uses that money to repay some of the loan she had taken earlier from the first NGO but the family’s overall debt situation is now worse than before they took the first loan. Kalu finds work at a cinema hall and with that income is able to pay back some of the loan. He also pressures his wife to take out additional loans from other NGOs so he can pay some of the original loan back. Kalu’s brother, Mukul, who had also borrowed money from the women’s loan, is also being pressured by NGOs to repay. Although there is no formal relationship between the NGOs and Kalu or Kalu’s brother, the NGOs are aware that the money they loaned to women are being used by the men in the family and start pressuring the men as well. Kalu’s younger brother Mukul is unable to deal with the aggressive loan recovery tactics of the NGOs and runs away from the village. At the end of the day there is no income from any entrepreneurial activity. Instead, Shonali’s family is now deeper in debt, their social relations with other families in the community have deteriorated and tensions within her own family have increased.

Figure 2 describes the processes and outcomes of this particular microfinance transaction.

OPEN IN VIEWER

Figure 2.

Processes and outcomes of a microfinance transaction.

Discussion

Our ethnographic study builds a grounded theory of vulnerability and extends our understanding of the role of social capital in poverty alleviation. In addition, our findings contribute to the literature on building inclusive markets, highlighting in particular the problematic role of NGOs in filling institutional voids. We elaborate on the theoretical implications of our findings below.

Our findings explain how vulnerabilities and powerlessness arising from poverty are exacerbated and why communities living in extreme poverty are unable to escape the poverty trap through microfinance. Figure 3 describes our model of vulnerabilities.

OPEN IN VIEWER

Figure 3.

A grounded model of vulnerability.

In developing an asset vulnerability framework, Moser (1998) identified labor, human capital (health, skills and education), productive assets, household relations and social capital as key assets that determine the extent of vulnerability. Our findings indicate that increased economic, social and environmental vulnerabilities arising from microfinance loans adversely affected both economic and social assets of the communities in the study. Inability to repay loans eroded the assets of the communities and led to poor health outcomes while undermining household relations and social capital. By revealing the economic, social and environmental dimensions of vulnerability our findings provide a richer picture of vulnerability than what can be captured by a probability risk function.

According to Moser (1998: 3), vulnerabilities involve both sensitivity (the magnitude of a system’s response to an external event) and resilience (the ease and rapidity of a system’s recovery from stress). Market-induced vulnerability increased the risks of poor communities falling further into poverty by influencing both sensitivity and resilience in the communities we studied. Inability to repay loans increased insecurity of borrowers as they found themselves in an inescapable debt spiral. In addition, the social capital of these communities, which was a key resource in coping with poverty, was undermined through aggressive loan recovery tactics, thus adversely affecting the resilience of these communities to recover from economic stress. We found that individuals and families experienced a sense of helplessness, which often accompanies economic stress (Moser, 1998) and that reduced the capacity to mobilize resources to cope with hazards (Watts and Bohle, 1993) as well as deprived capabilities of individuals to participate in economic activity or political processes leading to what Amartya Sen (1985) describes as the ‘politics of hope and despair’.

We also found that NGOs who were the main service providers in the region contributed to deteriorating social relations in the community through their aggressive loan recovery tactics that created a ‘culture of shame’ (Karim, 2008) by targeting women who were blamed for bringing shame to their husbands and families. Rather than building social solidarity, group lending had the opposite effect because each woman borrower was responsible for repayment of all other individual loans. Fear of potential defaults by members of their ‘solidarity circle’ created a mechanism of surveillance that diminished relations of reciprocity and trust and broke down group relations and family relations.

Borrowing from microfinance providers eroded social capital in two ways. First, aggressive repayment tactics from lenders involved public shaming of defaulters that adversely affected their social ties both with the community and with their family members. The ‘solidarity groups’ that were the basis of the social collateral of microfinance loans thus led to an erosion of bonding social capital. Second, access to microfinance loans did not create new bridging capital in the sense that while individuals had access to credit, the loans did not result in income generating activities but further increased indebtedness of households. In fact, following Bourdieu (1986) we found that microfinance activity resulted in negative social capital owing to erosion of bonding capital without any creation of bridging capital. In addition, as highlighted earlier bridging social capital contributed to a reproduction of inequality whereby a respected community member and group leader used her relationship with the external microfinance network to obtain loans for other members of her family. Thus, the availability of microfinance did not increase structural social capital from new network configurations but instead eroded relational social capital by undermining existing relations of trust, reciprocity and cooperation (Nahapiet and Ghoshal, 1998).

Our findings also contribute to the emerging literature on building inclusive markets by providing a micro-level account from the perspective of the target population. In many ways we provide the ‘other side’ of the picture in the important work done by Mair et al. (2012) on building inclusive markets in rural Bangladesh by drawing on the experiences of people that are being included in the global financial services system. In their study of BRAC’s operations in rural Bangladesh, Mair et al. (2012: 827) found that the NGO filled an ‘institutional void’ in the region by focusing on ‘local means of issue resolution’ and ‘making use of customary sources of social support’. Using traditional social networks may well allow NGOs to fill institutional voids but they do not address the longer-term consequences of increasing debt burdens and the economic, social and environmental vulnerabilities that we identified. Traditional forms of reciprocal exchange among family and extended family members are being overlaid with an increasing number of market-based financial exchanges with NGOs based not on trust but associated more with expediency and a short-term need for cash. Little consideration is being given by borrowers to longer-term consequences about how the loans will be repaid. NGOs can fill institutional voids in developing countries and build inclusive markets by ‘legitimating new market actors’ like women borrowers (Mair et al., 2012: 827), but we need to be cautious in assuming that inclusion necessarily leads to poverty reduction, especially for those living in extreme poverty, which our findings indicate is a boundary condition for market-led approaches to poverty reduction. Our study can be seen as a counter narrative to the NGO narrative seen from the perspective of the receivers of microfinance.

Implications, limitations and future directions

So how can researchers use our work? Our attempt to build a grounded theory of the vulnerability dimensions of poverty contributes to theories of building inclusive markets and the role of entrepreneurship in poverty alleviation. Entrepreneurship is the fundamental basis of the microfinance approach to poverty alleviation. As Bruton et al. (2013: 688) state, ‘market-based solutions such as entrepreneurship offer the best opportunity to create substantial and significantly positive change within poverty settings’. Promoting entrepreneurial economic activity among groups of women in rural Bangladesh is also the raison d’etre of Grameen Bank. Our study suggests there are boundary conditions for the entrepreneurship approach to poverty reduction. While there may be some potential entrepreneurs among poor rural populations, the vast majority of the poor do not possess the skills and creative visions that are required for successful entrepreneurs. In fact, as Karnani (2008) argues, poor self-employed farmers are mainly engaged in subsistence activities that offer no competitive advantages. If, as in our study of farming households, taking out loans to grow cash crops like maize is evidence of entrepreneurship, then we need to reconsider the concept of entrepreneurship in rural contexts like Bangladesh. Most individuals even in the industrialized countries of the world are not entrepreneurs and use credit for consumption purposes. It would be unreasonable and unrealistic to expect that the rural poor in developing countries are budding entrepreneurs who would use credit to make wise business decisions that generate income (Dichter and Harper, 2007). In fact, our findings indicate that vulnerabilities were exacerbated as a result of taking out loans in the name of entrepreneurship because no income generating opportunities were created. If the first microfinance revolution demonstrated that the ‘poor are bankable’ then the challenge of the second microfinance revolution is to identify what services are needed for the poor to meet their complex livelihood needs rather than ‘pretending they are all small farmers or micro entrepreneurs’ (Matin et al., 2002: 291). One of the criticisms of microfinance is that it reaches the ‘moderately poor’ but excludes those that live in extreme poverty (Prior and Argandona, 2009). Our study advocates a more cautionary approach in promoting microfinance to populations living in extreme poverty.

Much of the focus of the BoP literature has been on level of investments rather than on the social and economic value of these investments to poor communities. In their analysis of BoP initiatives Ansari et al. (2012: 818) call for a ‘grounded understanding of how poverty stricken communities survive’ to understand whether BoP initiatives create or destroy social welfare. Our study provides such a grounded understanding and our findings indicate that microfinance as a BoP strategy increased vulnerability risks of the extreme poor in the three villages that we studied. More research is needed to understand the relationship between microfinance and vulnerability risks in other contexts. ‘Co-creation’, ‘partnerships with local communities’, ‘creating markets for virtue’ are key concepts in the BoP and CSR literatures (Gond et al., 2013). Our findings indicate these partnerships continue to exclude people who are the recipients of poverty alleviation programmes and we argue that if poor communities become real partners in development the value propositions of these programmes will undergo a major transformation because they will be defined based on the needs of the poor, not on markets for microfinance. Collectivity and solidarity, which are crucial for the sustainability of poor communities, cannot be based on a system that excludes or silences the voices of those that are targeted for ‘development’ assistance (Papa et al., 1995).

Our findings also contribute to research that explores the role of social capital in poverty alleviation. For instance, in the theoretical framework developed by Ansari et al. (2012) social capital can develop capabilities in BoP segments by distributing capabilities within groups through bonding social capital and transferring capabilities between groups through bridging social capital. Our study of economic, social and environmental vulnerabilities provides key insights on the processes that hinder the development of capabilities and instead create new vulnerabilities. More research is needed on the negative aspects of social capital – for instance, the conditions that enable individuals to increase their own power and resource access as well as the conditions that can erode bonding social capital.

Another implication of our findings is that in addition to economic indicators such as access and cost effectiveness, microfinance programmes should be assessed by qualitative indicators of vulnerability and risk as identified in our study. Better social and economic outcomes could be achieved if social investment was directed at initiatives aimed at reducing risk and vulnerability such as building hospitals or schools, investing in building and supporting local businesses to provide employment and a steady income all year round for family members, or investing in social movements for fair and equitable access to land for farming. The target populations of poverty reduction programmes must play a key role in developing and running programmes like financial cooperatives, credit unions, local development banks and farmers’ societies. Alternatives to profit-driven, ‘new wave’ microfinance models are emerging from other parts of the world such as the Mondragon model in Spain, Self Help Groups and Neighborhood Help Groups as well as farmers’ cooperatives in the Indian state of Kerala (Bateman, 2011). The key difference between these models and the microfinance model lies in their collectivist approach as opposed to the individual entrepreneur focus of microfinance models. Both models use social solidarity of poor communities to promote empowerment, the difference being that collective approaches use solidarity as a means for collective production whereas microfinance relies on social pressure from the community to compel individuals to repay loans (Bateman, 2011). More research is needed to understand the conditions that can enable social solidarities to reduce vulnerabilities and achieve different outcomes.

Lastly, more research is needed to understand how microfinance influences the role of women in poor rural regions in developing countries. While it is true that giving loans to women is changing long-held cultural and social practices as more women involve themselves in farming activities to support their families, it would be unwise to infer that patriarchal structures are declining as a result. As we saw earlier, the loans taken out by women invariably were given to their husbands or male family members. Increase in dowry payments as a result of the availability of microfinance, which is what our study and others have found, certainly does not reflect any level of ‘empowerment’ among women.

Our study has some limitations. We do not claim that microfinance does not provide any benefits to poor populations or that our findings are generalizable to all poor communities that are recipients of microfinance. Rather, our aim is to generalize our findings into theory – in this case, theories of vulnerability and social capital in the context of poverty. It is reasonable to ask what would the situation have been if no microfinance was available? Would individuals have been pushed further into poverty? Randomized trials and field experiments with economic, social and environmental indicators are needed to answer this question but such studies are rare given the challenges of conducting these types of studies in rural areas of the developing world (Kent and Dacin, 2013).

In his analysis of development, Crush (1995: 5) raised some key questions:

The texts of development have always been avowedly strategic and tactical – promoting, licensing, and justifying certain interventions and practices, delegitimizing others . . . What do the texts of development not say? What do they suppress? Who do they silence – and why?

Perhaps these questions can serve as a starting point for asking the big questions about inclusive capitalism and development. If impoverished communities are to be empowered, we need to provide opportunities for chronically poor communities to tell their own stories about their real situations and discuss their real needs. We can start to empower them by listening.