Introduction

Microfinance institutions (MFIs) provide loans, savings, and other basic financial services to the poor in an effort to help alleviate poverty (CGAP 2011; Morduch 1999). Many of these organizations operate at the base of the economic pyramid (BoP)—an economic segment of the population that represents approximately four billion people worldwide who survive on less than US $1.25 per day. The microfinance literature has highlighted the benefits of microfinance, such as the capacity of microfinance to help develop the BoP economically and socially by benefitting both MFIs and borrowers (Goldberg 2005; Odell 2010). The borrowers at the BoP, for instance, can use the loans provided by MFIs to become entrepreneurs by creating and growing viable microenterprises that are characterized by “few employees, few assets, and informal operations” (Gudz 1999, p. 1). The idea behind microfinance is that the loans may help the borrowers get out of poverty.

The microfinance literature has also highlighted the financial challenges faced by both MFIs and borrowers at the BoP. A major challenge faced by MFIs in providing loans to borrowers is unrecoverable loans (Rosenberg 2009). As a corollary, a major challenge faced by the borrowers is the risk that they might be unable to repay the loans to the MFIs (Godquin 2004). While financial challenges have received the most attention in the literature, other important social and ethical challenges have received less attention.

Businesses and people at the BoP often suffer because of the negative contexts under which they operate. The negative contexts are often due to “institutional voids” at the country level (Chakrabarty 2009; Chakrabarty and Bass 2013; Khanna et al. 2005; Mair and Marti 2009). That is, a lack of functioning intermediaries in a country (e.g., ineffective government institutions, ineffective regulatory agencies, regressive laws and regulations, weak policing, etc.) creates market inefficiencies and difficulties for both businesses and people. Thus, institutional voids in a country can make the lending relationship between MFIs and borrowers more difficult.

Social norms and conventions can contribute to the existence of institutional voids at the BoP. Consider the role of gender inequality. Gender inequality is a major problem at the BoP (Cheston and Kuhn 2002; Karnani 2008). Social conventions that implicitly or explicitly promote gender inequality exclude women from participating in market-based activities (Karnani 2008; Mair et al. 2012; Rice 2010). Moreover, gender inequality at the BoP can also be manifested as aggressive/unethical/abusive acts against women (García-Moreno et al. 2005; Mair et al. 2012). In the microfinance industry, unethical or abusive acts against women may arise from the MFI loan agents tasked with recovering loans from women borrowers.

Despite the numerous challenges faced by women at the BoP, many MFIs prefer women borrowers because women are considered to be more compliant in following policies and procedures and more responsible in repaying loans (Brau and Woller 2004; Cheston and Kuhn 2002; D Espallier et al. 2011; Milgram 2001). How then, can MFIs serve women borrowers while operating in negative contexts? We suggest the answer might lie in Positive Organizational Ethics (POE). An organization that employs POE would view “the right thing to do is the only thing to do” (Verbos et al. 2007, p. 17). By knowing the right thing and doing the right thing, the organization can stand out as a beacon of positive ethics despite operating in negative contexts (Comer and Vega 2011). In addition to providing guidelines to avoid litigation or prevent unethical behavior, positive ethics codes allow organizations to build ethical virtues proactively in the organization (Caza et al. 2004; Sekerka et al. 2011). Thus, a formal POE code can be a catalyst of ethical strength and aspiration in the face of negative contexts that are characterized by deprivation, disorder, and dysfunction (Linley et al. 2006).

Our main argument is that MFIs that serve more women borrowers are more likely to adopt a POE code, especially in contexts where poverty is high and women’s empowerment is lacking. We define POE code as the MFI having a written code of business ethics that spells out organizational values and the standards of professional conduct expected of all staff (MIX Market 2010). We argue that the presence of a POE code is an indicator that the MFI is demonstrating the process of institutionalization of POE (Jose and Thibodeaux 1999; Sims 1991). In sum, the purpose of our paper is to examine how MFIs that serve women borrowers can adopt a POE code to build positive ethical strength in negative contexts. We extend the literature to examine what influences MFIs to adopt a POE code and what contextual factors affect the likelihood that MFIs that serve women borrowers will adopt a POE code. In the following paragraphs, we expand on the tenets of institutional theory, specifically (i) country-level institutional voids (Chakrabarty 2009; Chakrabarty and Bass 2013; Khanna et al. 2005; Mair and Marti 2009; Mair et al. 2012) and (ii) institutionalization of ethics inside organizations (Jose and Thibodeaux 1999; Sims 1991; Weber 1993).

Background Literature

We believe that ethical strength can be built in the face of institutional weakness and that the promise of microfinance can benefit both MFIs and borrowers. However, we contend that many existing models ignore the darker elements of the microfinance business at the BoP, and as a result, may be incomplete. We believe that the threat of unethical or aggressive behavior against borrowers, especially women, in contexts of desperate poverty and gender inequality, is a relevant and important issue faced by both MFIs and their borrowers. We seek to develop a better understanding of lending and borrowing at the BoP and look for mechanisms that may be employed by MFIs to build ethical strength from institutional weakness at the BoP.

Women as Borrowers of Microfinance: Gender Inequality in Impoverished Countries

BoP markets tend to suffer from gender inequality issues (Karnani 2008; Rice 2010) and extreme poverty (UNDP 2007). Yet, many MFIs that operate in these negative contexts choose to serve women borrowers (Brau and Woller 2004; Cheston and Kuhn 2002; D Espallier et al. 2011; Milgram 2001). Though women often represent a marginalized segment of society for the BoP, women borrowers are ideal customers for MFIs for two reasons. First, women are more likely to repay loans in full, and as a result are less risky borrowers than men (Cheston and Kuhn 2002; D Espallier et al. 2011; Milgram 2001). Second, women borrowers are perceived to be more compliant than men, creating smoother relationships between MFIs and women borrowers (Cheston and Kuhn 2002).

Though women may be the preferred clients of MFIs, gender inequality persists in many underdeveloped countries. The persistence of social conventions arising from traditional legal, political, and religious institutions, as well as customs (North 1990), prevents women in these countries from participating in market transactions with organizations, such as MFIs. This excludes women from partaking in market-based activities (Mair et al. 2012), such as becoming entrepreneurs by creating microenterprises. Thus, the exclusion of women from participating in market-based activities is a characteristic of the negative context of the BoP.

Additionally, gender inequality issues create an increased risk of the victimization of women due to unethical or aggressive behaviors by the more dominant members of society (Pickup et al. 2001; Raphael and Tolman 1997; Swain and Wallentin 2009). The likelihood that women at the BoP will be victims of unethical or aggressive behaviors is aggravated in countries where women lack political representation in governmental and policy-making authorities (Molyneux and Razavi 2002). In contexts where women are politically underrepresented, women suffer due to incapacity to voice concerns, lack of policies that aid women in market-based activities, and lack of protection from unethical or aggressive behaviors against women (Iyer et al. 2011). We highlight some of the literature on unethical/aggressive behaviors against women at the BoP in Table 1.

Table 1 Women at the base of the economic pyramid
Full size table

MFIs that choose to serve women do so under institutional voids that make serving an excluded, underrepresented, or victimized segment of society more difficult. Social conventions, for instance, may exclude women from participating in market-based activities (Mair et al. 2012). Further, MFIs that serve women borrowers in difficult contexts—where women are underrepresented politically or victimized by unethical or aggressive behaviors—face dilemmas that arise from targeting a customer segment that lacks both voice and power. To overcome such difficulties, an MFI can choose to institutionalize—within its organization—the ethical treatment of women borrowers. By institutionalizing organizational ethics, the MFI can demonstrate the highest ethical values of the organization to both internal and external stakeholders and impede further gender inequality in the contexts in which they operate.

Institutionalization of Positive Organizational Ethics

Institutionalization of ethics is an important topic in business ethics research (Chakrabarty 2013; Jose and Thibodeaux 1999; Sims 1991; Weber 1993). Ethics become institutionalized by the “persistence of specific behaviors associated with an organization and its employees behaving ethically” (Sims 1991, p. 494). That is, institutionalization of ethics is a process cultivated through continual ethical operation of the organization. It is important to note that institutionalization of ethics is not an all-or-nothing concept (Sims 1991). Institutionalization of ethics involves many implicit and explicit actions by an organization and its members to varying degrees. For example, institutionalization of ethics may occur explicitly through codes of ethics (Adams et al. 2001; Murphy 1995) and ethics training, as well as implicitly through organizational culture and leadership (Chakrabarty and Whitten 2011; Jose and Thibodeaux 1999; Sims 1991). Thus, one or many of these implicit and/or explicit facets could indicate the process of institutionalization of ethics within organizations.

“Positive” ethics shift focus from rules and discipline to virtues and emphasis on what organizations have potential to become (Handelsman et al. 2009). Whereas “organizational ethics” focus on avoiding litigation and prescribing organizations and organizational members what not to do, “positive organizational ethics” go a step further to direct attention toward principles that will move the organization toward its highest ethical potential (Handelsman et al. 2002). The institutionalization of positive ethics, then, moves organizations from focusing on rules of how not to act, or how to avoid litigation, toward principles that build ethical strength in the organization. Positive ethics make the organization’s values explicit—to guide ethical action by the organization and its members.

We define positive organizational ethics as ethical principles and statements of value that guide an organization toward its highest ethical potential. We believe that organizations, such as MFIs, can institutionalize POE to guide ethical action in the organization. Adoption of a POE code is a key indicator of the process of institutionalization of POE in MFIs. A POE code in MFIs will not merely be rule-based or focus on behavior to avoid litigation, but rather will indicate that the MFI is aspiring to its highest ethical potential by focusing on values and aspirations of the organization and its members.

Theory and Hypothesis Development

A formal POE code can be a catalyst of ethical strengths, potentialities, and aspirations that can help overcome ethical weakness, disorder, and dysfunction (Linley et al. 2006). The literature on POE has focused on aspects such as “loyalty, productivity, organizational citizenship, desire to comply with organizational rules, reactions to organizational change, and willingness to communicate” (Verbos et al. 2007, p. 21). We extend this literature by arguing that MFIs that serve women borrowers would be more likely to adopt a POE code—to overcome threats arising from the negative contexts in which MFIs operate, specifically the threat of unethical or aggressive behaviors against women borrowers. We operationalize the extent to which an MFI serves women borrowers in terms of the ratio of women borrowers to total borrowers (MIX Market 2010). Though incidents of unethical and aggressive behavior against women borrowers are not necessarily associated with all MFIs, acknowledging the potential threats and ethical dilemmas associated with choosing to serve women borrowers may create a more accurate and realistic picture of the threat of the dark side of microfinance.

Further, we consider the negative contexts that are more likely to enable these threats of microfinance to surface, specifically in contexts of high poverty and lack of women’s empowerment. We define the poverty of microfinance borrowers as the borrowers’ attempts to survive on inadequate finances, reflected as miniscule borrowings, which is very common among the vast majority of people at the BoP who live on less than $1.25US per day (World Bank 2011). We define the lack of women’s empowerment as the exclusion of women from processes of political power, reflected as a lack of representation of women in important political power structures (Inter-American Development Bank 2005). We suggest that MFIs that serve women in these negative contexts are more likely to adopt a POE code.

Serving Women Borrowers: Institutionalizing Positive Organizational Ethics in MFIs

The role of MFIs in providing financial services to women customers is an important dimension of microfinance research (Cheston and Kuhn 2002; Goldberg 2005; Odell 2010). Many MFIs prefer women as borrowers because women are more compliant in following policies and procedures and more responsible in repaying loans. They are easier to conduct lending with—loans become easier to collect, resulting in lower write-offs for MFIs (Brau and Woller 2004; Cheston and Kuhn 2002; D Espallier et al. 2011; Milgram 2001). As a result, many MFIs prefer to have more women customers and provide services that more closely fit their interests and needs (Cheston and Kuhn 2002; Green et al. 2007; Whitten et al. 2010; Zardkoohi et al. 2011). Though MFIs may prefer women borrowers, gender inequality at the BoP can create difficulties for women to secure loans, work outside the home, and gain independence (Cheston and Kuhn 2002; Mayoux 2007; Swain 2007). Gender inequality in society may exclude women from market-based activities or make it more likely that women will be victims of unethical and aggressive behaviors.

In order to serve women who are excluded from market-based activities but still comply with social conventions arising from the traditional legal, political, and religious institutions, an MFI can institutionalize POE. One way to institutionalize POE is to make the organization’s guiding principles explicit to internal and external stakeholders (Jose and Thibodeaux 1999; Sims 1991). Institutionalizing POE by adopting a formal POE code would indicate to internal and external stakeholders that the MFI places a high level of emphasis on ethics. It can help legitimize the MFI’s role in serving women borrowers at the BoP, without offending the traditional legal, political, or religious institutions in the country (Mair et al. 2012). It can allow the MFI to hold itself to its highest ethical potential and serve an excluded segment of the population while operating alongside traditional institutions.

MFIs can also institutionalize POE to prevent women borrowers from becoming victims of unethical or aggressive behaviors. The potential for victimization of women borrowers may arise from various members of society at the BoP, including MFI employees. Consider, for example, the conduits of the relationship between MFIs and borrowers—the loan agents of MFIs. When tasked with maximizing the number of borrowers and then recovering loans from the borrowers, the loan agents might turn to unethical and aggressive behaviors. These unethical and aggressive behaviors might include not only stealing, collecting more than is owed, and taking unnecessary collateral (Duggan 2009) but also verbal, physical, psychological, and emotional abuse (Cons 2010; Swain 2007). Abusive acts are those unethical or aggressive behaviors that can involve psychological or physical maltreatment (García-Moreno et al. 2005). This might be especially problematic for women borrowers at the BoP, because the BoP is a context of desperate poverty where gender inequalities are pervasive (Karnani 2008; Rice 2010). For instance, a study of women borrowers in Bangladesh finds that women who receive and have difficulty in repaying microfinance loans may be vulnerable to abuse by the agents collecting the loans (Karim 2011). Other studies find that this threat of abuse is especially valid when women borrowers are unaware of their rights (Mayoux 2007). Anecdotes in the practitioner literature suggest that abuse and aggressive behavior by MFI loan agents can sometimes result in tragic consequences for women (Economist 2010):

“There have been abuses.…Some ‘recovery’ methods have involved intimidation.…Laxmi Peta is mourning her 16-year-old daughter Lalitha. The family ran up 66,000 rupees in debts from five MFIs to pay for the wedding of their elder daughter. Unable to meet a payment, they went away to seek help from the new in-laws, leaving Lalitha alone. An MFI officer arrived, along with the village head and the four other members of her mother’s ‘joint liability group’—fellow villagers who had taken collective responsibility for the debt. After their harangues, Lalitha drank pesticide. Her mother treasures a tattered suicide note.”

In seeking to protect both themselves and women borrowers from unethical and aggressive behavior on the part of loan agents, MFIs can benefit from adopting a POE code as a mechanism to reduce the threat of these behaviors from occurring. On a pessimistic note, the adoption of a POE code might be merely window dressing, designed to signal intended behaviors without actual implementation of the same—the utility of which in comforting women borrowers would obviously be short-lived. On a more optimistic note, however, the adoption of a POE code can be a reflection of a genuine effort that exemplifies an organization-wide realization for the need to build ethical strength. Institutionalization of a POE code can influence those in the organization to act more ethically (Jose and Thibodeaux 1999; Sims 1991). This genuine effort of MFIs would be to convert a richer understanding of institutional weaknesses (where women borrower suffer in negative contexts) to inform a more durable form of ethical strength (formalized by a positive ethics code). Further, akin to any firm that adopts formal codes to “signal the firm’s social responsibility to external stakeholders” (Bansal and Hunter 2003, p. 292), an MFI might believe that having a POE code would act as a positive signaling mechanism that would be a source of comfort and assurance to its women borrowers.

Overall, in the face of institutional voids manifested as gender inequality issues, MFIs can adopt a POE code to reduce the threat that organizational members will behave unethically or aggressively against women borrowers. The combination of exclusion of women from market-based activities with the possibility of unethical or aggressive actions against women borrowers may influence MFIs to adopt a POE code to build positive ethical strength in negative contexts. In sum, we suggest that

Hypothesis 1

MFIs with a greater percentage of women borrowers are more likely to have a POE code.

Contextual Factors

Poverty

Contextual factors are vital to understand the social, economic, and developmental complexities at the BoP (Rasche et al. 2008). Contextual factors, such as the degree of poverty in the region or country, can influence how MFIs cope with and operate in such contexts. Poverty can contribute to economic aspects such as workforces characterized by low earnings and productivity, as well as social aspects such as increased crime, decreased health and education, and increased gender inequality (Fafchamps and Minten 2006). Inextricably, gender inequality in countries with higher poverty levels is often ubiquitous, jeopardizing the socio-economic freedom of women in these countries.

The aggravation of gender inequality issues in impoverished countries can result in a greater exclusion of women from market-based activities (Mair et al. 2012). This requires organizations, such as MFIs, to fill institutional voids and grant women access to markets. However, MFIs must do so without operating in opposition to political, legal, and religious institutions that traditionally contributed to the social conventions that perpetuated these inequalities. We suggest that in impoverished contexts, MFIs that serve women borrowers may be even more likely to adopt a POE code—to gain legitimacy, explicitly state the organization’s ethical purpose, and guide the organization to its highest ethical potential.

Additionally, countries with higher degrees of poverty might be more vulnerable to unethical and illegal behaviors. This is because the legal institutions, ethics codes, and ethics training that guide “moral decision-making and action” (Sekerka and Godwin 2010, p. 64) are usually not widespread in such countries (Pickup et al. 2001). When serving women borrowers, MFIs that operate in contexts of higher poverty may be more prone to instances of unethical or aggressive behaviors by MFI employees, especially loan agents. MFIs can reduce the threat of unethical or aggressive behaviors on the part MFI loan agents by employing a POE code to guide these agents toward ethical behavior. Thus, in contexts in which the degree of poverty is higher, it is more imperative for MFIs that serve women borrowers to adopt a POE code—to reduce the exclusion of women borrowers from market-based activities and reduce the threat of unethical or aggressive behavior on the part of MFI loan agents. As such, we propose the following interaction:

Hypothesis 2

Poverty of borrowers moderates the influence of the percentage of women borrowers on the likelihood of the MFI having a POE code. The influence is stronger when the poverty of borrowers is high.

Lack of Women’s Empowerment

Women, relative to men, tend to lack representation in political power structures in many underdeveloped countries (Kenworthy and Malami 1999). The lack of women’s empowerment in political power structures in a nation silences women’s voices, concerns, and aspirations both during lawmaking by national legislative bodies and in law enforcement by their national executive bodies (Inter-American Development Bank 2005; UNDP 2007; World Bank 2011). Accordingly, in countries in which women are less empowered in political power structures, the women at the BoP are more likely to be excluded from market-based activities and be victims of unethical or aggressive behavior. We suggest that for MFIs that serve women borrowers, adopting and enforcing a POE code to guide ethical behavior of the MFI and its loan agents can help overcome these gender inequality issues.

As women are most often the poorest of the poor at the BoP (Cheston and Kuhn 2002; Mayoux 2007; Swain 2007), financial services offered by MFIs can help women turn into entrepreneurs who operate microenterprises. However, women at the BoP often find it difficult to operate entrepreneurial ventures because the society and markets are often plagued by unethical discrimination against women (Mair et al. 2012). Additionally, financially independent women borrowers might become targets of violence (Rahman 1999; Swain 2007) due to their new, and sometimes objectionable, position in society. These gender inequalities may make it more difficult for both MFIs and women borrowers to interact and lend/borrow from one another. In these contexts, women may be more vulnerable to aggressive or unethical behavior on the part of MFI loan agents—many of whom are male. From a moderate perspective, male loan agents may view the women borrowers as unequal. From an extreme perspective, male loan agents may view the actions of these women borrowers in securing loans, working outside the home, and gaining independence as offensive (Rahman 1999). As such, MFIs can adopt a POE code to help alleviate tension between male loan agents and women borrowers as these women become included in market-based activities. Further, POE codes can provide guidelines to prevent unethical or aggressive behavior on the part of the MFI’s male loan agents.

Thus, in contexts in which women’s empowerment is lacking, it is even more critical for MFIs that serve women borrowers to adopt a POE code that can help provide assurance that the women borrowers would be treated ethically. Therefore, we suggest that

Hypothesis 3

Lack of women’s empowerment moderates the influence of the percentage of women borrowers on the likelihood of the MFI having a POE code. The influence is stronger when lack of women’s empowerment in the country is high.

Methods

Sample and Procedure

Our primary sample consists of MFIs from regions across the world including Eastern Europe and Central Asia, East Asia and the Pacific, Africa, South Asia, Latin America and The Caribbean, Middle East and North Africa. Data on the selected MFIs are collected by the Microfinance Information Exchange (MIX). MIX was incorporated in 2002 as an independent, non-profit organization that promotes transparency and dissemination of accurate information on the microfinance industry (MIX Market 2010). Using its publicly available data collection tools, MIX collects annual reports, audits, financial statements, ratings, and other relevant documents to provide comprehensive and accurate information on the microfinance industry.

Financial indicator data are directly submitted to MIX by each MFI, by the affiliated network that files on the MFI’s behalf, or gathered from public documents published by the MFI, such as annual reports. MIX supplements these data with archival documents, such as ratings, annual reports, donor/investor reports, and audits to capture market dynamics as well as more integrated performance data of individual MFIs. These data are supplemented by organizational data voluntarily provided to MIX by the institution or affiliated network. Data are validated by MIX employees using a three-step process. First, MIX employees utilize external data sources made available through local microfinance networks to verify data supplied by MFIs. Second, MIX uses information supplied by partners such as MFTransparency and the Grameen Foundation to insure certification of MFI data through independent assessment. Third, MIX collaborates with third-party funders, auditors, and raters to perform on-site data assessments.

Because the study focuses on whether MFIs that serve women borrowers are more likely to adopt a POE code in negative contexts, only MFIs that report data on POE code and percentage of women borrowers are utilized for this study. For the purpose of this study, a dataset was created by merging databases from MIX (MIX organizational survey database and MIX financial indicators database) and the World Bank Development Indicators data. The sample size was dictated by the extent of overlap among the merged databases and the availability of non-missing data for the variables of interest. During the data collection stage of this study, the entire 2009 MIX organizational survey database had data for 348 firms, out of which 336 firms had non-missing data for the POE code variable. Of these, 275 firms had non-missing data for variables from the other two sources of data (2008 MIX financial indicators database and World Bank database). Thus, the merged dataset allowed a sample size of 275 MFIs.

Table 2 provides the sample characteristics. The MFIs included in this sample are distributed across 58 countries, with MFIs from the Latin American region having the largest representation. The World Bank (2011) defines high-income countries as those with GNP per capita greater than $12,275. None of the MFIs in our sample operate in high-income countries. Furthermore, we verified that the MFIs in our sample function primarily in the poorer regions within their respective countries. Sixty-three percent of the MFIs in our sample are non-profit organizations and 41 % are non-governmental organizations (NGOs). The sample means of financial and operational data suggest that an average MFI is relatively small in size (in terms of total assets and number of employees) with a very strong focus on the microfinance business (more than 90 % of operations is in microfinance).

Table 2 Characteristics of sample
Full size table

Measures of Variables in Hypotheses

Percentage of Women Borrowers

This is measured as a percentage, calculated as the ratio of the number of active women borrowers to the total number of active borrowers. The number of active borrowers refers to the number of individuals or entities that currently have an outstanding loan balance with the MFI or are primarily responsible for repaying any portion of the gross loan portfolio. Individuals that have multiple loans with an MFI are counted as a single borrower. Data are obtained from the MIX financial indicators database.

Poverty of Borrowers

This moderator variable, poverty of borrowers, is one indicator of the negative contexts in which MFIs and borrowers operate. A low average loan balance per borrower is a good indicator of the poverty of borrowers because poorer people tend to borrow smaller amounts and relatively wealthier borrowers tend to borrow larger amounts (Rosenberg 2009). The average loan balance per borrower is measured as a ratio of the MFI’s gross loan portfolio to the number of active borrowers. The numerator, gross loan portfolio of the MFI, is defined as all outstanding principal in dollars for all outstanding client loans, including current, delinquent, and restructured loans, but not loans that have been written off, not including interest receivable or employee loans (MIX Market 2010). As noted, MFIs whose borrowers are poorer tend to have a lower average loan balance per borrower, whereas MFIs whose borrowers are relatively wealthier tend to have a higher average loan balance per borrower. Accordingly, the ratio is reverse-scaled by multiplying by −1, so that the higher values (negative values closer to zero) indicate higher poverty and the lower values (negative values farther away from zero) indicate lower poverty. Data are obtained from MIX financial indicators database.

Lack of Women’s Empowerment in Country

This moderator variable, lack of women’s empowerment in a country, is another indicator of the negative contexts in which MFIs and borrowers operate. This country-level variable, indicative of the exclusion of women in a country’s power structures, is measured using the proportion of seats held by women in national parliaments of the country (i.e., the percentage of parliamentary seats in a single or lower chamber held by women) (World Bank 2011). This percentage is reverse-scaled by multiplying by −1 to indicate the lack of women’s empowerment, so that the high values (negative values closer to zero) indicate that lack of women’s empowerment is high in the country, whereas low values (negative values farther away from zero) indicate that lack of women’s empowerment is low in the country. Data are obtained from the MIX financial indicators database.

Positive Organizational Ethics (POE) Code

The dependent variable, POE code, is measured as a binary variable. It has a value of 1 if the MFI confirms in its disclosure to MIX that it “has a written code of business ethics that spells out organizational values and the standards of professional conduct,” and has a value of 0 if it confirms otherwise. In this phrasing used by MIX, the “has a written code of business ethics …” portion indicates the existence of an organizational ethics code. Further, the “…that spells out organizational values and the standards of professional conduct” portion indicates that the ethics code has a positive framing. Hence, in sum, the phrase “has a written code of business ethics that spells out organizational values and the standards of professional conduct” attempts to capture the existence of a positive organization ethics code. To confirm the accuracy of data provided regarding the existence of a POE code, MIX follows a three-step validation process outlined in the Sample and Procedure section above. MIX works with one of its external partners, The Smart Campaign, to address the validity of ethics codes policies. MIX and The Smart Campaign have developed a certification process that will certify MFI ethics codes among other institutional facets based on a standard set by The Smart Campaign. Data are obtained from MIX organizational survey database.

Control Variables

MFI Size

Firm size is included as a control because larger MFIs are likely to have a greater influence among the community and other stakeholders and perhaps more inclined to adopt a POE code. Firm size is measured as the log of total assets, where total assets are reported in dollars. Data are obtained from the MIX financial indicators database.

MFI Reach

Reach is measured by the number of staffed points of service and administrative sites used to deliver or support the delivery of services to microfinance clients. It is included as a control because MFIs with wider reach, in terms of access and connection points, might be more inclined to adopt a POE code. Data are obtained from the MIX financial indicators database.

MFI Social Mission

MFIs often have a social mission to complement their lending business among the poor. Accordingly, this control variable is measured as the aggregate number of social priorities that comprise the MFI’s mission. The social priorities that are checked (i.e., whether they are part of an MFI’s mission or not) are increased access to financial services, poverty reduction, employment generation, development of start-up enterprises, growth of existing business, adult education improvement, youth opportunities, children schooling, health improvement, gender issues, water and sanitation, and housing. It is included as a control because socially responsible organizations might be more inclined to adopt a POE code. Data are obtained from MIX organizational survey database.

Country Female Population

This is measured as the percentage of the country’s population that is female. The population of the MFI’s country counts all residents regardless of legal status or citizenship (except for refugees not permanently settled in the country of asylum) (World Bank 2011). It is included as a control because much of this study is relevant to women’s issues. Data are obtained from the World Bank Development Indicators database.

Country Prosperity

Country prosperity is an indicator of economic wealth and quality of life, and is negatively related to poverty. Country prosperity is calculated as gross national income (GNI per capita) in thousands of U.S. dollars and converted using the World Bank Atlas method, divided by the midyear population for the country. GNI is measured as the sum of value added by all resident producers plus any product taxes (less subsidies) not included in the valuation of output plus net receipts of primary income (compensation of employees and property income) from abroad (World Bank 2011). The World Bank Atlas method applies a conversion factor that averages the exchange rate for a given year and the two preceding years, adjusted for differences in rates of inflation between the country and countries in the Euro area, Japan, the United Kingdom, and the United States (World Bank 2011). Data are obtained from the World Bank Development Indicators database.

Country Mortality

The country mortality rate is measured as the crude death rate for the country or the number of deaths occurring during the year, per 1,000 population estimated at midyear (World Bank 2011). This human factor measure, in contrast to the financial measure of country prosperity, is an indicator of human suffering that arises from poverty and poor health infrastructure (Cabigon 2005). This is included as a control because women and children tend to suffer the most under conditions of high mortality. Data are obtained from the World Bank Development Indicators database.

Results

Table 3 provides the descriptive statistics and correlations for our study. Logistic regressions are used to test the hypotheses, the results of which are included in Table 4. For obtaining the regression results presented in Table 4, all the independent variables were centered and standardized. Centering (setting mean = 0) helps reduce the possibility of multicollinearity problems, especially when interactions are involved. Standardization (setting mean = 0 and standard deviation = 1) helps in the comparative interpretation of effect sizes (parameter estimates) when the independent variables are measured on widely different scales/units (which is common in financial/economic/operational data). Standardization makes the “standard deviation” the unit of measurement for all variables, which allows comparison. Further, we took advantage of the availability of longitudinal data, to lag the independent variables behind the binary dependent variable by 1 year (this is does not indicate causality; it simply indicates the longitudinal direction of the influence being tested).

Table 3 Correlations: operational and economic data of sample of MFIs
Full size table
Table 4 Logistic regression: positive organizational ethics code as a consequence of serving women microfinance borrowers in negative contexts
Full size table

Consistent with Hypothesis 1, the percentage of women borrowers has a significantly positive influence on the probability of the MFI having a POE code (β = 0.16 with p < 0.05 in model A2). Consistent with Hypothesis 2, the poverty of microfinance borrowers is a marginally significant moderator of the influence of the percentage of women borrowers on the probability of having a POE code (β = 0.27 with p < 0.10 in model A3). Figure 1 provides the logistic interaction plots (the moderator variables are continuous, but only lines representing high and low values of the moderators are plotted for ease of visualization). As shown in the interaction plot in Fig. 1, when the poverty of borrowers is high, the percentage of women borrowers has a strong positive influence on the probability of a POE code. Consistent with Hypothesis 3, the lack of women’s empowerment in the MFI’s country is a significant moderator of the influence of the percentage of women borrowers on the probability of having a POE code (β = 0.17 with p < 0.05 in model A3). As shown in the logistic interaction plot in Fig. 1, when the lack of women’s empowerment in the country is high, the percentage of women borrowers has a strong positive influence on the probability of a POE code.

Fig. 1
figure 1

Interaction plots: are microfinance institutions that serve women borrowers in negative contexts likely to adopt a code of positive organizational Ethics?

Full size image

In sum, the results of our econometric analysis suggest that MFIs that serve women in negative contexts are more likely to adopt a POE code, where negative contexts are those in which poverty is high and women’s empowerment is lacking.

Discussion

We use tenets of institutional theory including (i) country-level institutional voids and (ii) institutionalization of ethics in organizations to guide our discussion on the relationship between MFIs serving women borrowers and adopting POE codes in negative contexts. We use operational and economic data collected by MIX on an extensive sample of MFIs from across the world for our study. Employing this dataset allows us to garner a greater understanding of the association between MFIs serving women borrowers and MFIs adopting a POE code, especially in countries in which poverty is high and women’s empowerment is lacking.

Implications for Theory

Our study contributes to the literature and offers implications for theory. First, our study extends the literature on institutional voids by exploring the difficulties MFIs and women borrowers face at the BoP. We believe that microfinance can produce many positive outcomes such as poverty alleviation (Morduch 1999; Odell 2010), economic and social development (Goldberg 2005), and women’s empowerment (Cheston and Kuhn 2002; Mayoux 2007; Swain 2007). However, we caution future research against adopting an overly optimistic approach. By acknowledging the presence of institutional voids, we hope to not only provide an accurate view of microfinance but also examine a mechanism, positive organizational ethics, that can potentially help MFIs serve borrowers in the presence of institutional voids.

Second, our study incorporates negative contextual factors. We support the idea that MFIs that serve more women borrowers might be more vulnerable to threats of aggressive or unethical behavior by MFI loan agents, which can negatively impact both the women borrowers and the MFI. By highlighting the negative contexts, we believe that we provide a more realistic and complete picture of microfinance at the BoP. Though microfinance does serve a noble purpose and can help to alleviate poverty, an overly optimistic and incomplete view of microfinance at the BoP can stifle the future development of microfinance models. By incorporating both positive and negative dimensions of microfinance, we hope to provide a more accurate and complete view of microfinance.

Third, we choose to focus on POE to understand how MFIs can mitigate threats associated with serving more women borrowers, especially in contexts of desperate poverty where women’s empowerment is lacking. We find that the adoption of a POE code can be important for MFIs that serve more BoP women borrowers, especially in negative contexts represented by poverty and lack of women’s empowerment. Thus, MFIs that incorporate a written or formal POE code can provide ethical guidelines to loan agents that can reduce the instances of unethical or aggressive behaviors against women borrowers. Consequently, we believe that POE can be a powerful mechanism that brings light to the dark side of microfinance. We choose to focus on POE to understand how MFIs serve women borrowers in negative contexts. We argue that a written or formal POE code helps build positive ethical strength in negative contexts. That is, a POE code helps the MFI to gain legitimacy by explicitly demonstrating that the MFI is creating a social space for women in market-based activities. Additionally, the adoption of a POE code provides MFIs with ethical guidelines for loan agents that can reduce instances of unethical or aggressive behavior against women borrowers. Though POE might not be the silver bullet in overcoming gender inequality, it is an important mechanism that an MFI can implement for the betterment of both the MFI and its women borrowers.

Overall, though microfinance is a viable tool to alleviate poverty at the BoP, it must be examined from a neutral, rather than an overly optimistic, lens. MFIs operate in very difficult contexts in order to provide financial services to the world’s poor (Feigenberg et al. 2011; Schreiner 2002). Thus, all of the threats associated with microfinance from both the MFIs’ and borrowers’ perspectives must be acknowledged in order to propel this industry and future research forward.

Implications for Practice

Our study also provides practical implications. First, we believe that the BoP is a rich source of business and entrepreneurial activity that should not be ignored (Prahalad and Hammond 2002). In particular, “businesses can gain three important advantages by serving the poor—a new source of revenue growth, greater efficiency, and access to innovation” (Prahalad and Hammond 2002, p. 51). The BoP presents viable and profitable opportunities for business and entrepreneurship.

Second, a relationship exists between MFIs serving more women borrowers and MFIs adopting a POE code. By adopting a POE code, MFIs can provide guidelines to loan agents on ethical treatment of women borrowers, especially in contexts of poverty and lack of empowerment. Further, a POE code can aid in establishing legitimacy for the MFI, particularly in contexts plagued by gender inequality. Thus, an MFI, by adopting a POE code, can build positive ethical strength in the negative contexts of the BoP.

Third, institutional voids must be taken into consideration in advancing opportunities at the BoP. Specifically, the BoP is beset with institutional voids. As such, organizations interested in entering and operating at the BoP must be aware of issues related to poverty, gender inequality, health problems, and illiteracy. Awareness of these challenges is important for organizations that wish to operate successfully at the BoP.

Limitations and Future Research

We acknowledge some limitations of our study, which can be addressed by future research. First, we argue that a POE code is one indicator of institutionalization of positive organizational ethics. At the same time, we recognize that institutionalization of POE is a process that might include other complementary aspects such as ethics training, leadership, and an organizational culture that promotes POE. Hence, future research should examine these other indicators of institutionalization of POE. Second, it is possible that a POE code acts as a positive signaling mechanism (Bansal and Hunter 2003; Chakrabarty and Wang 2012, 2013). For instance, scholars have argued that firms may adopt formal codes to “signal the firm’s social responsibility to external stakeholders” (Bansal and Hunter 2003, p. 292). Though a POE code may be a reflection of a genuine effort that exemplifies an organization-wide realization for the need to build ethical strength, it might also merely be a window dressing designed to signal intended behaviors without actual implementation of the same. Our study does not test whether the adoption of a POE code is a genuine effort by MFIs to inspire the organization and its members to strive for a higher ethical potential (as opposed to being a mere window dressing). Hence, future studies should investigate the intention of MFIs in adopting POE codes.

Conclusions

Microfinance is an important mechanism to help alleviate poverty at the base of the economic pyramid. Though microfinance offers viable prospects for economic and social development at the BoP, challenges arise from operating in these markets. Our primary finding is that MFIs that serve more women borrowers are more likely to adopt a POE code, especially in contexts where poverty is high and women’s empowerment is lacking. We suggest that MFIs that adopt a POE code, in contexts of desperate poverty and disparity in gender equality, can build positive ethical strength in negative contexts. A POE code can guide the organization and its members toward a higher ethical potential and grant the MFI legitimacy in social contexts that exclude women from market-based activities. Our study lays the groundwork for future research to explore the relationship between gender inequality and positive organizational ethics to uncover the complexities of the challenges and opportunities inextricable to microfinance.