A credit-plus approach : the way to increase women’s empowerment in microcredit?
- Evidence from Colombia.
Master’s thesis 30 credits Sara Eriksson
Department of Government Supervisor: Olle Folke
A credit-plus approach : the way to increase women’s empowerment in microcredit?
- Evidence from Colombia.
Abstract
Does participation in microcredit programs lead to the empowerment of women? The answer is not clear-cut, and earlier studies report ambivalent results. The United Nations Sustainable Development Goals number 5 argues for the importance of finding paths to empowerment. In order to facilitate women’s empowerment in microcredit scholars suggest the use of a credit-plus approach. The purpose of this paper is to reassess the impact of microcredit by making use of the credit-plus approach of Fundación WWB. The study is based on new primary data collected in a Minor Field Study conducted in Colombia. Through developing and testing three hypotheses I assess the impact on three dimensions of women’s empowerment: economic, social and cultural, and psychological. A mixed method approach of both OLS regressions and semi-structured interviews are applied. The results indicate that a credit-plus approach does increase the positive impact of credit on women’s financial control within the household. Yet, no significant result could be found in the case of decision-making power or self-efficiency. The paper also concludes that non-financial services per se may constitute a way to increase women empowerment.
Keywords: Colombia, microcredit, credit-plus, empowerment, decision-making, control over finance, self-efficiency
Acknowledgements
First and foremost I like to thank the Swedish International Development Cooperation Agency (Sida) and the Department of Government at Uppsala University for granting me the scholarship through which I was given the opportunity to conduct my minor field study in Colombia.
In Colombia I would like to thank everyone at Fundación WWB for being so welcoming, and for all of the help I have been given during my time at the organization. I also like to give a special thanks to Delia, Carolina, Sandra and Eximena. Furthermore I like to give a warm thanks to all of the women who participated in this study, particularly those who have opened their hearts and homes to me.
Thank you all for the help with my research, for making me feel welcome and giving me an unforgettable time in Cali.
Last but not least I like to thank my supervisor Olle Folke for all his help. His support and advice, during the whole process have been invaluable, especially in the many times of encountering problems.
Table of content
Introduction ... 5
Microcredit and microcredit plus ... 7
Characteristics of Fundación WWB ... 8
Theoretical discussion ... 10
Empowerment and microcredit ... 10
Earlier research and findings ... 11
Empowerment: the concept and its definition ... 13
Conceptual framework and development of the hypotheses ... 14
Methodological discussion ... 18
Mixed method approach ... 18
Data collection and sample ... 19
Operationalization and survey questions ... 20
Treatment and control groups ... 22
Model specification ... 25
Empirical result ... 28
Impact on the different empowerment indexes ... 28
Control over finance ... 29
Decision-making ... 31
Self-efficiency ... 33
Qualitative results ... 35
Limitations and problems of the study ... 40
Summary and conclusion ... 41
References ... 44
Appendix ... 1
Appendix 1. Table7: Index for control over finance as dependent variable ... 1
Appendix 2. Table 10 independent indicators for control over finance ... 4
Appendix 3. Table 14. independent indicators for Decision-making ... 8
Appendix 5. Table 16: Index for self-efficiency as dependent variable ... 10
Appendix 6. Table 19: Independent indicators for self-efficiency as dependent variable .... 13
Introduction
”End all forms of discrimination against all women and girls everywhere”
This quote is part of Sustainable Development Goal 5 concerning gender equality, set in 2015 as part of Agenda 2030 (UN). It highlights the importance of women’s empowerment as not only as a road to other development goals, but as the end goal itself. The setting of the goal also put weight on finding paths toward reaching female empowerment. Microcredit, as the provision of small loans to poor female entrepreneurs, has come to be described as the silver bullet for women’s empowerment (Pitt et al 2006). Scholars argue that microcredit has the ability to generate empowerment among its female beneficiaries. For example, Hashemi, Schiler and Riley (1996) found an increase in empowerment among the women taking part in a rural credit program in Bangladesh.
Yet not every scholar does agree and not every study has found a positive effect. Scholars like Karlan & Thuysbaert (2014) found no effect at all on women’s empowerment in a study with females participating in a microcredit program based in Mali. Furthermore, Mayoux (1999) argues that taking microcredit can, in some settings, instead have a negative impact on the women. According to her, some women taking credit are facing the risk of not becoming empowered, but rather more vulnerable, due to reinforcing patriarchal gender norms.
As the ambivalent results from various studies indicate, there is no common conclusion on the ability of microcredit to empower women. Consequently, scholars increasingly question the notion of microcredit per se being a silver bullet for womens’ empowerment. Kabeer (2001c) questions the notion that the sole access to credit generates empowerment. She argues that empowerment is not something that can be imparted by a third party. Empowerment includes a desire to make changes in one’s own life. Changes that can only come from the woman herself, and a woman’s empowerment are therefor fully dependent on if she herself wants to become empowered. Consequently, microcredit organizations cannot create empowerment per se. They can only serve a means to facilitate this process of empowerment. Therefore, one needs to make a difference between women as beneficiaries of credit and women who are empowered. However, this is not the reality and the microcredit market is growing into a number of millions of women that are today participating in the many programs around the world. Now, if microcredit is to be used as a path to women’s empowerment, and achieving the sustainable development goal number 5 for gender equality, there is a need to reassess its effect on the women.
women (Mayoux 2000). Microcredit combined with other services like voluntary saving, insurance, development of business, marketing support and gender awareness training is all part of the credit plus approach. Yet, there are only a few studies that formally test the impact of incorporating these non-financial services and training programs on women’s empowerment in microcredit. A review of the empirical literature of women’s empowerment and microcredit done by Sefa Awawaryo (2014), display that there is a need for further research using a wider range of metrics for microcredit. Rigorous evidence on the combined impact is still lacking, especially when looking at non-financial services with the explicit focus of women’s empowerment as the end goal.
The aim of this paper is to respond to the current discourse. By using the credit-plus approach of Fundación Woman World Banking Colombia (Fundación WWB) as an example I assess the impact of microcredit on women’s empowerment. The study is based on new primary data collected through a Minor Field Study conducted in Colombia between October and December 2016. With the use of a mixed method approach the paper address three specific dimensions of women’s empowerment: economic, social and cultural, and psychological empowerment at a household level. By developing and testing three hypotheses I answer the research question; how does participation in microcredit-plus programs impact women’s empowerment in Colombia? The purpose of the paper is to contribute the understanding of microcredit and women’s empowerment within the field itself, and serves as base for decision makers when taking action towards reaching Sustainable Development Goal for gender equality.
Furthermore, a majority of the previous evaluations of the impact of microcredit still come from Asia, particularly Bangladesh, where the movement first originated. Yet, there is an increasing amount of women reached by microcredit outside of Asia and earlier findings suggests that the impact differ by regions in the world and even countries. Differences in country characteristics, structure of the microfinance institutions and prevailing social norms and traditions, all play a part in a woman’s empowerment according to Mayoux (1999). This geographical limitation in earlier studies could create problems of external validity, and make it difficult to generalize results obtained from one particular geographic or socioeconomic setting to countries such as Colombia (Lieberman 2005:441) Latin American and Caribbean (LAC) countries are particularly under researched and evidence from these countries is lacking. (Gutierrez and Soares 2011).
Considering this, the contribution of this paper lies within three aspects. First, I provide new primary data from Colombia. Second, I contribute to a better understanding of the impact of microcredit on women’s empowerment in an LAC country, more specifically Colombia.
Third, I reassess the impact of microcredit on women’s empowerment from a credit plus approach at a household level.
The rest of the paper is structured as follows; below I continue with a brief background of the minimal microcredit and the credit plus approach, followed by a presentation of Fundación
of the relationship between microcredit and women’s empowerment. Here I discuss empowerment as concept and its definition. I conclude this chapter by the construction of my conceptual framework and development of the hypotheses I will test in order to answer my research question. A methodological discussion where I describe the two methods used and the construction of my sample as well as the specification of the regression models are provided in chapter three. In the last chapter I present the empirical results found in this study and provide an analysis. I then finish with a summary and conclusions of the result.
Microcredit and microcredit plus
Muhammad Yunus is the developer of the concept of microcredit and the first Microfinance Institution (MFI) called The Grameen Bank. The bank gave small credit to very poor female entrepreneurs without security. The credit was meant to serve as working capital in order to create self-employment and an independent income (Grameen).
In 2005 the United Nations held an event to highlight microcredit. It was made as an attempt to make microfinance and financial services more available to people, even at the very lowest economic levels of society. In 2006 Muhammad Yunus received the Nobel Peace Prize for his efforts. Because of this Microcredit gained a high reputation and spread across the world. In 1997 about 8 million people were reached by a MFI, and by 2014 there were as many as 204 million using MFI (Microcredit Summit Campaign 2014).
As the people borrowing the credit did not hold any security, a special lending model was created. Today this lending model is called The Grameen Bank Group Methodology and is based on the model of group lending and joint liability. When taking credit the beneficiaries are asked to organize themselves into solidarity groups. If one of the participants in these groups does not repay his/her loan, then the rest of the group is not able to get a new credit before this person has made the repayment. In this way the MFIs get around the problem of insecurity and increases its repayments (Grameen). Furthermore, credit is almost exclusively given to women. In 2008 the World Bank stated that the focus on women as microfinance clients can empower them (World Bank 2008) Yet, originally the focus on women was closely related to the fact that women are considered less of a risk for the MFIs, as women generally are better at paying back the loan (Morduch et al 2005:139).
When first introduced, the MFIs focused on the minimalistic version of microcredit, i.e. the sole provision of small loans. Yet, some MFIs began to include alternatives for saving and insurance in their microcredit programs. Later came the incorporation of business training, health care and social services. This incorporation of non-financial services came to be called the “credit plus” approach. By definition the services are provided by the MFI themselves alongside the standard financial services (Sievers and Vandenberg 2007). The credit-plus
assistance, product design, legal services or b) social services such as health, education to raise health consciousness etc. (Sievers and Vandenberg 2007).
Some MFIs that have been linking credit to education and other non-financial services are ProMujer, FINCA (Foundation for International Community Assistance) and BRAC (Bangladesh Rural Advancement Committee). They all use education and training as part of their group-lending methodologies for credit. They have either incorporated social services, such as health education of child healthcare, legal advice or personal training or services aimed at developing one’s micro-business, which involves financial and technical assistance (Aghion and Morduch 2005). Yet, none of these services provided have the explicitly focused on empowerment of women, which can be seen as the outcome per se.
Characteristics of Fundación WWB
This study is based on the credit-plus approach of the organization Fundación WWB in Colombia. In 1980 they opened in the city Cali, located in the south west of Colombia, with the explicit goal of the organization of empowering women and aid their families, through microcredit, education, training and complementary non-financial services. Fundación WWB was one of the first microfinance institutions in Colombia and the first one to enter the Women World Banking association in 1982 (womensworldbanking). In 2008 the organization were divided in two parts, the foundation Fundación WWB and the bank Banco WWB. Yet, they still work as a united organization.
Fundación WWB has some general characteristics that are more prevalent in Colombia, and distinguish it´s microfinance from that in Asia, which is based on the Grameen bank model (Grameen). First, these MFIs are based in more rural areas and have a larger focus on the poorest in society. Fundación WWB, on the other hand, does not target the rural population.
Their focus is on urban areas of all sizes. Nor do they focus specifically on the poorest part in society but provide for anyone that does not have access to banking services. Yet, the majority of the women who participate in their programs are from lower or middle socio- economic spheres of the society. Second, another important difference from the Grameen bank structure is that Foundation WWB does not follow the specific group based lending model. They only offer individual credit. This means that they do not have the social pressure of paying back the loans as in the group-based method, nor are the participants able to gain other social contributions from within a group.
What makes Fundación WWB a credit-plus organization is their provision of informal education in the form of specialized courses. The courses aim to help the social, political and legal development of women as leaders, and through the knowledge of her rights as a woman, she gets empowered. The courses cover; female entrepreneurship and empowerment, economic management, virtual and digital design of their business, information about saving and leadership (with the goal of helping the women to administer and manage her business better), managing personal finance within her business as well as in her family. The courses
and the rest of the courses have different levels to complete. The women are encouraged to complete the total package. Some of the training is for free but others require a small fee.
Fundacion WWB argues that providing women with this knowledge gives them the ability to empower themselves and transform the gender relations in their own lives.
Additionally, the organizations uniqueness also lies in the fact that they are providing these education opportunities for the female entrepreneurs without the need for the women to hold a loan at the bank. Yet, the women taking part in the courses gain a close relation to the bank, and some have the opportunity to apply for credit, while some are offered a credit from the bank (FundacionWWB).
Theoretical discussion
This theoretical discussion is constructed as follows; I first display the connection between women’s empowerment and microcredit. Then I continue with a review of earlier studies and findings, followed by a discussion on the concept and definition of empowerment. Finally, I conclude this chapter with the construction of my conceptual framework, and the development of the hypotheses.
Empowerment and microcredit
Before looking at the impact of microcredit on woman’s empowerment it is important to understand the theoretical background. Microcredit leads to the inclusion of women in the economy. It includes and integrates a marginalized group within the financial economy, as women are often excluded (Bayulgen 2008). The aim of microcredit is therefore to include the women in a system were they were earlier excluded, by placing financial recourses that were not previously available to them, in the hands of the women themselves.
Furthermore, a woman would benefit more from the services of the microcredit then a man as the credit and financial constraints facing economically deprived people are generally greater for women, than for men. Women often lack the opportunity to gain formal employment and regularly own nothing in their own names (World Bank 2008). Furthermore, enabling them to earn an independent income is a priority. When earning an independent income it is argued that the women improve their position within the household as it is assumed to increase the relative value of a woman, as she becomes an income provider to the household. Therefore, by getting assess to credit a woman can earn her income independently and gain respect within the household through increased economic power and control over her own income (see for example Hashemi et al 1996, Moyle et al 2006).
Providing microcredit to women further enables them to free themselves from patriarchal gender norms of the society. A women who has credit has greater opportunities of contributing to the family support and is therefore arguably able to gain a higher value and decision-making power over her own income. For example, microcredit assists the women to gain a higher bargaining power - permitting them to exercise power within and outside the household (see for example Amarta Sen 2000:201, Hashemi et al 1996). Women’s empowerment manifests itself in many different areas and dimensions. For example: greater bargaining power, increased decision-making power, increased self-efficiency and self- esteem, political awareness, legal awareness and a gain in higher collective identity.
Furthermore, giving credit to a woman, rather than a man, would not only benefit the woman but also her family. Women generally have a greater concern for the children, and she is therefore more likely to use her increased decision-making power because of her contributed income and to invest her credit and profit in education and healthcare for her children, or the general welfare of the family. Microcredit can therefore provide the opportunity for girls to go to school. Hence, giving microcredit to a woman is expected to lead to improved level of
health in the household, education and the use of family planning (se Hashemi et al 1996 and Kumar 2009).
Earlier research and findings
There is a wide range of literature assessing the impact of different microcredit projects. Yet, the findings are ambivalent and results are inconclusive. I therefore limit the discussion to some of the most cited studies of the effects of microcredit and women’s empowerment.
Furthermore, as mentioned earlier, most of earlier studies on microcredit and women’s empowerment are made in an Asian context, specifically in Bangladesh. Therefore, I will first discus these findings, but then discuss the few earlier findings that have been made outside of Bangladesh, with a focus on countries in the LAC countries. I then conclude with a discussion of the findings from the credit-plus approach, even though studies with this focus are a bit scarcer.
One of the most cited studies within microcredit and women´s empowerment is Goetx &
Gupta (1996), who look specifically at women’s control over their own income and credit amongst the women participating on rural credit programs in Bangladesh. They find that it was the women’s partners that controlled the credit allocated to the women. Taking credit therefore resulted in a further subordination of woman within the household rather than empowering her. Another study by Kato and Kratzer (2013) who, similarly to Goetx and Gupta (1996), also examined how participation in a microfinance program affects women’s control over savings and income generated from their business. However, they compared women who participated to women who did not participate, in the context of Tanzania. They did not find a negative impact, but a positive impact.
Other important studies from Bangladesh are Hashemi et al (1996) and Amin et al (1998).
Both of these studies observed many different dimension of empowerment amongst women participating in rural credit programs, and they found that credit had a positive and significant effect on all these different dimensions. Hashemi et al (1996) found that program participation and credit was closely associated with the size of the women’s economic contribution, as well as an increase in purchasing power and legal awareness. They also argue that as women interact more outside the house, they gain self-efficiency and thereby become further empowered. Amin et al (1998) made use of both quantitative and qualitative data, and used three empowerment indexes of inter-spouse consultation, individual autonomy and authority. They find a significant difference between members and non-members of credit programs. Yet, a study in Mali does not find any impact at all (Karlan & Thuysbaert 2014).
The authors make use of a randomized evaluation, but do not find any impact on any of the decision-making, health, education, or social capital indicators.
measuring the impact on choice-enhancing resources. Yet, he did find a strong positive effect on enhancing women’s exercise of agency in the household. The study reassesses the impact in the Bangladeshi context by drawing on data collected in rural program- and non-program villages. The study of Wrigley Asante (2011) is based in the context of Ghana and argues for a positive impact on the women’s socio-economic status and being recognized as income earners and contributor to the household. Yet, the change sometimes created a “power conflict” within the household. Mayoux (2002) looked at women’s empowerment in India and argued that the economic, as well as social empowerment impacts, are rather marginal and some are even becoming disempowered. For example, she argues that targeting women without supporting empowerment strategies lead to a double burden for the women.
Supporting this notion, a study made by Berge et al (2011) in Tanzania confirms that female entrepreneurs did not only spend less hours in their own business compared to male entrepreneurs, but they were also limited in their flexibility by their responsibility of the household. Most female entrepreneurs therefore operated within, or close to, their home.
Furthermore, in a review of empirical results on microcredit and empowerment outside of the Asian context found a few studies made in countries like China, Indonesia, Zambia, Ghana and South Africa as well as the LAC countries Bolivia, Honduras and Ecuador. Results from these studies are diverse and suggest that the impact of microcredit on women’s empowerment is highly dependent on the specific contexts (see Bolivia (Mosley 2001), China (Park and Ren 2001), Ghana and South Africa (Afrane 2002), Honduras and Ecuador (Smith 2002), Indonesia (Bolnick and Nelson 1990).
Furthermore, all of the studies discussed above are made from within the minimalistic credit approach. Scholars argue that credit alone might not improve the empowerment level of women within the household (Rahman 1999). Studies suggest that the integration of non- financial services in the credit-plus approach may be important for its effect. Most studies made with a credit plus approach contain different focus areas that are not directly expected to create women’s empowerment, but rather a focus on improving the repayment rate at the MFIs or in reaching a bigger profit for the women. Yet, there are a few studies made with the credit-plus approach and women’s empowerment specifically.
For example, a study within the area is conducted by Nathalie Holvoet (2005) in the context of south India. She looks specifically at decision-making through the credit-plus approach.
Her findings suggest that a combining credit with financial social group intermediation changes the overall decision-making pattern, from male dominated decisions and norm following, to joint or female decision-making. Furthermore, and Greuble (2012) finds a significant decrease in intimate partner violence and improvement in empowerment, if the husbands are included in the plus courses given to the women.
Yet, some recent Randomized Control Trial demonstrates mixed result. For example, Kim et al (2007) combined the impact of microfinance and health training on HIV infection and gender norms as well as domestic violence in South Africa. They argue that there are
made by Gené and Mansuri (2011) and Karlan & Valdiva (2011) reveal no significant changes of the incorporation of business training on attitudes toward domestic violence. Gené and Mansuri (2011) look at women in rural Pakistan and did not find a positive impact for the women. Finally, the study by Ashe & Parrott (2002) evaluated a women’s Empowerment Project in Nepal that combined credit with business and literacy programs. They found that having more business knowledge increased the self-esteem of the women participating. The increase in self- was the most significant change, and a change that they themselves appreciated the most.
Empowerment: the concept and its definition
The concept of empowerment is multidimensional, complex and not easily defined. It therefore still lacks a clear definition. Self-direction, agency, decision-making, control, autonomy or living a life of dignity, are some terms that have been related to the concept (Ibrahim & Allen 2007). Yet, all the different definitions of empowerment contain the idea about making a change in perception at community or individual level. This change is described as a path to personal transformation, which leads to the improvement of individual capabilities and a person’s ability to make and exercise decision about their own life (Malhotra et al 2002). Below I discuss the most influential definitions of the concept of empowerment, that all have been commonly used in the literature about microcredit and empowerment.
The first definition of empowerment has a close relation to the concept of power. The definition was first introduced by Rowlands, J (1997) and is built on four different categories of “power”. It is argue that a woman’s empowerment is based on these difference powers.
These are described as “power to”, “power with”, “power over” and “power within”. “Power to” refers to creating new economic and human possibilities. “Power with” refers to acting in a group. “Power over” refers to once possibility to resist manipulation, and “Power within”
refers to enhancing self-respect and self-acceptance.
The second definition refers to what is called Agency and Opportunity Structures. Agency is the person’s ability to make purposeful choices, and influenced by the individual’s material, social and psychological, as well as collective assets and capabilities. Opportunity Structure refers to one’s institutional, social and political context. The institutions and politics interact with a woman’s Agency, and might possibly constrain it. The definition states that empowerment is a process of increased individual capacity. Opportunity structures should therefore be seen as preconditions for one’s agency, but it is the combination that decides a woman´s empowerment (Narayan 2005:5-10)
The World Bank commonly uses this definition. The main difference in general
and TV etc. Yet, the definition also includes capabilities like education, health, capacity and information as well as participation and self-confidence (World Bank 2012).
The final two definitions build on each other, and are related to the central features of Process and Agency. Process is defined as the change towards a greater freedom of choice and action.
In this definition, empowerment is not an outcome created by the process, but rather the very process per se. Agency is the very essence, the heart of empowerment. It implies that the woman herself is the actor within the process of change that is measured. The two most influential definitions are the ones by Amartya Sen (1985b) and Kabeer (2001c).
Sen (1985b) made groundbreaking and important contributions to the definition of empowerment. He argues for the central role of agency in what he calls Human Development and Capability Approach. Sen defines empowerment as “what a person is free to do and achieve in whatever goals or values he or she regards as important” (Sen 1985b:206). In his definition, Agency is the process of freedom, and refers to a person acting in a way to create change.
Kabeer (2001) draws from the definition of Amartya Sen (1985b). She defines empowerment as “expansion in people´s ability to make strategic life choices in a context where this ability was previously denied to them” (Kabeer 2001c:19). Her definition stresses the notion of choice, and that the ability to make choices necessarily implies that there are alternatives. It is described as the ability to choose between different options on where to live, whether to marry, how many children to have and the freedom from the state of disempowerment.
Conceptual framework and development of the hypotheses
Many different proposed frameworks for measuring empowerment exist within the literature.
In this study I draw from the framework of Kabeer (2001c), and make use the framework of dimensions and indicators from Malhorta and Schuler (2002, 2005). Below I first discuss my conceptual framework of microcredit and women’s empowerment, and then I develop the hypotheses.
As discussed above, Kabeer (2001c, 2005:13) concept of empowerment refers to “the ability to make choices”. According to her, empowerment implies actively exercising choices that are challenging power relations. Getting out of disempowerment, and becoming empowered, is defined as making choices that you were denied before. It is the gaining of the ability to exercise choice. She argues that empowerment should be thought of in terms of resources, agency and achievements. Recourses include material, social and human resources of society, community and family. One’s agency is exercised through, and within, these resources. It refers to the institutions and relationships in society were certain actors have a privileged position over another. Agency is the process by which the choices are made. It is the ability to make choices in their life and act upon them, even when faced with opposition. Achievements
are the outcome of these choices, the outcome of one’s capabilities based on their Resources and Agency (Kabeer 2005).
Furthermore, Kabeer (2005:15) make a distinction between a passive form of agency and an active form of agency: that of a greater effectiveness of agency and a transformative agency.
A greater effectiveness of agency relates to an increased effectiveness within their given roles and responsibilities. A transformative agency is an agency defined as the women’s ability to challenge their roles and responsibilities.
The framework of Malhorta and Schuler (2002:13) refers to women’s empowerment as the ability to make strategic life choices. Their framework is based on different dimensions and indicators that affect the women’s wellbeing. The dimensions are: economic, social, cultural, legal, political and psychological empowerment. All of these dimensions described contains sup-dimensions or indicators of control over finance, control, decision-making, legal awareness etc. to mention some. A woman’s ability to make choices is, according to Kabeer (2001:c), limited because of their position in the society and they remain in the state of disempowerment. Every one of these indicators described is part of a woman’s empowerment process. Women are denied a choice within these dimensions due to oppressiveness structures and being able to make choices within these areas is the process of empowerment.
Some of these dimensions and indicators of Malhorta and Schuler (2002:13) are universally considered important in measuring women’s empowerment, such as control over resources and, mobility, involvement in household decision-making and self-efficiency (see for example Goetz & Gupta 1996, Pitt et al 2006, Steele, Amin, and Naved 1998). Other dimensions such as freedom form household domination or legal awareness are indicators that have been used, but a much less frequent. Furthermore, a woman´s agency and inequalities can vary across these dimensions, and she may be empowered in one dimension of their life, while not on others. This is does not imply that empowerment in one dimension cannot translate to empowerment in other dimensions. However, it does not necessarily follow. Furthermore, the independent interpretation of the different dimensions serves the understanding of their agency (Malhorta & Mather 1997, Hashemi, Schuler, and Railey 1996).
In this study I look at the economic, social and cultural as well as psychological empowerment of the women within the household. These domains are provided by Malhorta and Schuler (2005), and universally accepted as important when assessing women’s empowerment. The limitation of the different domains is done due the limited time and space of the paper, and in relation to the focus of the organization. The courses provided at the organization are focused on economic, social and cultural as well as psychological empowerment.
Figure 1. The framework for understanding women’s empowerment. Constructed from the framework of Kabeer (2001:c) and Malhorta and Schuler (2005)
Through the credit-plus programme women are provided with resources that remove them from their state of disempowerment. Yet, empowerment cannot be imparted by a third party.
A woman’s empowerment is fully dependent on if she herself desires to become empowered according to Kabeer (2001c). Microcredit organizations can therefore only facilitate this process and do not create empowerment itself. As empowerment includes a desire to make change in one’s life, that can only come from the woman herself. So you have to make a difference between women as beneficiaries, and women who are empowered. Agency is closely related to self-efficiency, the woman realizing that see can be an agent of change (Kabeer 2005, Malhorta & Schuler 2005:72-73).
As presented in Figure 1 above, it is hypothesized that the women move from the state of disempowerment to a state of empowerment through; their resources, which is the assess to microcredit and non-financial services, and their Agency of participating in these microcredit services. This enhances their ability to exercises control and strategic choices that positively affects their life (Kabeer 1999). A woman is hypothesized to gain the achievements in Figure 1 as an outcome. The achievement as empowerment in the indicators of: economic-, social- and cultural- as well as psychological empowerment. Three hypotheses and expected outcomes are developed below, from within these different dimensions of women empowerment.
The first and second hypotheses are related to the women’s economic empowerment as well as social and cultural empowerment. It is expected as an achievement from within the constructed framework. It is argued that microcredit, through the creation of opportunities and the integration of women in the economic system, is empowering women. However, who is making the decisions about the money, and keeping the income and/or profit is crucial for the understanding of the impact of microfinance on women’s empowerment according to Hunt & Kasynathan (2001).
Additionally, participating in the microcredit plus services enable the women not only to gain an independent income, but also to gain greater economic control, as the complementary courses is expected create a environment for challenging norms and opportunity for greater control over her own independent income and savings. This leads to a greater economic independence and higher bargaining power within the household. With credit, a woman would be more likely to challenge the norms that are restricting her decision-making (Swain
Resources Provision of Microcredit
Provision non-financial services
Agency Participation in the
services
Achievements Economic empowerment Social- and cultural empowerment
Psychological empowerment
& Wallentin 2008 and Ashraf, Karlan & Yin 2008). Based on my conceptual framework my first two hypotheses follow as:
H1; Participation in microcredit-plus services leads to an increase in the women’s economic empowerment
H2; Participation in the microcredit-plus services leads to an increase in the women’s social and cultural empowerment
Third, one of the most crucial areas for empowerment is a woman’s self-efficiency. It is argued that taking part in microcredit enhances the socioeconomic level of the woman, and an improved living standard can contribute to the process of challenging the social norms and cultures within the home. Women holding credit are believed to feel more effective as this unlocks their capabilities and opportunities (Bayulgen 2008:533). Yet, self-efficiency is not an easy concept as it relates both the woman’s perception about her capabilities and her actual capabilities.
Microcredit services are equipping women with both information and new ideas of herself and her capabilities. Through gaining alternative ways of thinking and doing, and to perceive the role of women, it leads to changes in the perception of themselves (Mahmud & Shah 2012). The women start believing in themselves and what they are capable of doing. In both within their business as well as within the home, and make her feel confident in making decisions for their business and their families. Hunt & Kasynathan (2001) found that most women participating in the microcredit plus program valued their confidence, knowledge and training they have gained the highest. This indicates that women participating in the program increase their self-esteem and self-efficiency, and further so if the credit is combined with complementary courses. Therefore, my final hypothesis follows:
H3: Participation in microcredit-plus services leads to an increase in the women’s psychological empowerment
Methodological discussion
In this chapter I provide a methodological discussion. I first lay out and discuss the methods used, and then provide a description of the collection of data and the construction of my sample. This follows by the speciation of the regression models, and a discussion of the variables. I continue with the operationalization and measurement of the three different empowerment dimensions, as well as the construction of the treatment and control groups.
Mixed method approach
Drawing from Dunning (2016) I use both a quantitative and a qualitative method in order to answer my research question and test the hypotheses developed in the conceptual framework.
The quantitative method serves as the base, and the weight of this study lies with this part.
Yet, by using both methods I bring together their different strengths and overlapping some of the weaknesses in quantitative method.
Exercising both a quantitative and a qualitative approach are according to Dunning (2016) central. The combinations give a strong research design and contribute to the understanding of the context in the process. Quantitative indicators are according to useful for collecting a bigger about of data. It lends itself to the construction of a research design that gives opportunity for statistically evaluations. It can also be used in order to check the balance of pre-treated characteristics in a treatment and control group (Dunning 2016). Qualitative data is way to gain detailed case experience and validate the research model. It provides a better understanding and interpretation of the outcomes from the quantitative estimations. It also offers a way to contextualize the evidence in the way that gives you knowledge of what something means in that specific context. (Dunning 2016:26, 326-327)
In the quantitative part I statistically estimate the impact of a microcredit plus approach on women’s empowerment using Ordinary Least Square (OLS) regressions. To assess the impact of microcredit I also construct a treatment and control group. The use of a treatment and a control group is one of the most common approaches when assessing the impact of microcredit. As my study analyzes the impact of both credit and participation in the courses, I construct two treatment and control groups. The first treatment group contains women who have credit, and the control group is women who do not have credit. The second treatment group includes women who take part in courses at the organization. Its control groups are women that have not finished any of the courses at the organization yet. That is to say, the women in the control group have been accepted to a course, but have only just started. Due to the short time participating in the course I therefore do not expect it to having made a difference in outcome yet. I further use the quantitative method to check the balance of the characteristics in the treatment and control groups.
The qualitative method is used in order to triangulate my result and provide support for my
study is important, as empowerment is strongly associated with the social and cultural norms of the society, which shape the agency of the women. Qualitative data would therefore enrich the quantitative analysis trough further understanding and help in interpretation of the result of the quantitative part.
Data collection and sample
The primary data is collected through a Minor Field Study conducted in Cali, Colombia, between October to December 2016. To collect the quantitative data I used a standardized survey. Using a standardized survey gave me the possibility to collect new primary data and statistically assessing the result. For the qualitative part I conducted semi-structured interviews. The use of semi-structured interviews is common in the social science. The choice of semi-structured rather than structured interviews was based on the methodological use of the interviews in this paper. I wanted to be able to change the formulations and questions dependent on the direction that the respondent wanted to go. Semi-structured interviews are often are used as they guide the interview according to what the respondent thinks is important, rather than contain them within a structure (Lieberman 2005: 436,441).
The surveys were written in Spanish, and all of the interviews were done in Spanish with a digital voice recorder. Before the interviews the study was described to them and the women were informed about their anonymity, and were given further information about informed consent. The interviews and the surveys were either done in a room provided at the Fundación WWB or at the home of the respondent.
The population is women taking credit and/or participating in the courses provided by Fundación WWB. The sample is not based on any random sampling, but on strictly voluntarily participation. In order to construct my sample the original plan was to use the detailed register of the women having credit and taking part in the non-financial educational program, kept by the organization. The register would have been utilized in order to build a random sample for my treatment and control group. However, arriving in Colombia I was faced with problems due to a new law about secrecy of personal information that prevented me from getting access to this resister. This was a setback for the study. I therefore had to change my plans on how to create to my sample. Instead I had to rely on a complete voluntary participation of the women taking part in the courses at the organization at that current moment (or women having any other direct connection with the organization at that time).
The problems I faced resulted in a small sample. A bigger sample would have been preferable as this increases the chances of finding a statistical significance, and a stronger reliable representation of the population. Around 150-200 women were actively taking part at the organization on a day-to-day basis. 98 of the women volunteered to take part in my questionnaire. The sample is therefore entirely based on voluntary participation. 6 women
within which I had to work it was not possible to collect a bigger sample. I was informed by the organization that the best way to get as many women as possible to participate in my study was to take part myself in the courses. This would give the women a better opportunity to understand who I was, and why I wanted to talk to them. This procedure is naturally very time-consuming but necessary and this was how I proceeded. Furthermore, while participating in the courses, and asking the women if they wanted to participate, I highlighted my independence from the organization, and the anonymity of their participation. Even so, many of the women remained very skeptical and did not want to take part.
Operationalization and survey questions
The survey consists of two parts. The first part asks questions about the woman’s personal characteristics and socioeconomic level. Question like; the age of the woman, amount of children and her level of education. The second part asks the questions related to the empowerment indicators within the different dimension described earlier in the conceptual framework.
Table 2. Empowerment indicator and their measurement
Indicator name Indicator description Survey question Economic dimension: Control over finance
CINC Control over own income Who make most decisions about how to use your salary?
CSAV Control over own savings Who make most decisions about your savings?
CFE Control over family economy Who make most decisions about your family’s economy?
Social and cultural dimension: decision-making DMP Participating in decisions about
minor purchases Who make most decisions about minor purchases?
(Groceries or cloths made) DMAP Participating in decisions about
major purchases Who makes most decisions about major purchases, like durable items? (TV, fridge furnisher)
DIH Participating in decisions about expenses for the home
Who make most decisions about expenses for the house?
(Improvement or repair) DTTA Participating in decisions about
time and task allocation
Who make most decisions about your allocation of time?
(Between business and home shores) DED Participating in decisions about
education for child
Who make most decisions about education for the children? (Expenses etc.)
DAMC Participating in decisions about how many children to have
Who make the decision about how many children to have?
Psychological dimension: Self-efficiency
SENH In general, feeling confident in
negotiation with partner I do in general feel confident negotiation my opinion with my partner
SEMB Feeling confident in managing own business
I do feel confident managing my own business
SEMFE Feeling confident in managing family economy
I do feel confident managing the family economy
The economic dimension consists of indicators of financial control. The social and cultural dimension consists of decision-making indicators, and the psychological dimension has indicators of self-efficiency. The entire set of specific indicator used in this study is presented in Table 2 above, along with its measurement in the form of the survey question. All of the questions in the survey were written in Spanish, but I present the translated questions in the table.
There exists many ways of measuring women’s empowerment, and there are other indicators that I have excluded. Furthermore, the selected indicators can be discussed and problematized. Yet I have chosen these indicators based on the notion of having internationally comparable indicators, and being able to compare my results to earlier studies made within the area. The indicators in this study are based on the framework and indicators by Malhotra et al. (2002). Below I describe how and why I have chosen these specific indicators in Table 2 and how they are measured.
When choosing the indicators it is important to take into consideration the cultural aspects of Columbia. Agency and empowerment are highly cultural concepts that relate to the norms and values within the society. An attribute might signify empowerment in one context, but have different meaning elsewhere. For example, the shift in a woman’s ability to visit a health center without the permission from her husband is a sign of empowerment in Bangladesh, but not in Peru (Malhotra et al. 2002). I therefore made the choice to exclude some that are not valid in the context of Colombia.
Financial control is part of the Economic empowerment dimension. It refers to a woman’s control over her own finance. I include the indicators; control over own salary, control over own saving and control over family economy. The indicators are suggested by Malhorta et al (2002) and are as mentioned earlier and internationally acknowledge. The economic empowerment indicator is measured through the question; who make most decisions about (...). This question has been use in studies like Hong Nhung (2014) and Ibrahim & Alkire (2007).
Decision-making power displays a woman’s level of bargaining power within the household, and is part of the Social and cultural empowerment dimension. It is based on if she has a say in decision making about major and minor purchases, expenses for the house, time and task allocation, educational expenses, and the decision about how many children to have. These indicators are extensively used in earlier studies. Indicators for decisions about children’s health and mobility have often been included in earlier studies. Yet, I choose not include them with reference to the contextual factors.
When the survey first was developed I included the internationally accepted indicators for
indicator in the final survey as it might bias my result or not really indicate empowerment.
Malhorta et al (2005:77) argue that in areas where there is a relatively strong public health system the ability of women to manage the children’s health is not that important.
Furthermore, there are no cultural restrictions on female mobility and the indicator is therefore not included. The decision-making indicator is measured through the question “Who make most decisions about (…)”. These specific survey questions have been used by earlier studies to measure decision-making power by scholars such as Mosonera & Heshmati (2016) and Mahmud, Shah and Becker (2012).
Self-efficiency is part of the Psychological empowerment dimension. These indicators are generally considered to be the most difficult indicators to measure as they refer to both women’s perception of their capabilities and their actual capabilities (Cheston and Kuhn 2002). The indicators in this study are self-efficiency in; managing family economy, managing own business as well as general confidence in negotiation with partner. The question I use to measure this indicator is “I do feel confident….” and is inspired from researcher like Kato & Kratzer (2013).
For the questions about financial control and decision-making, the women can answer either a) that she makes most decision alone b) that her partner makes the most decision c) mostly joint the decision with her partner or d) mostly joint decision with other in the family or e) mostly someone else in the family make the decision. Answers a, c and d are then computed as 1, which means that she is involved in the decisions. If the answer were b or e then this is computed as 0, non-involvement in decision. For the questions related to her self-esteem the women could answer 1 for feeling confident and 0 for not feeling confident.
The different indexes are then constructed from the indicators within each of the three dimensions. The index for financial control ranges from 0-3 and indicates the level of control over economy. The index for decision-making ranges from 0-6 and indicates the level of involvement in decision-making within the household. Finally, the index for self-efficiency ranges from 0-3, and indicate the level of self-efficiency. In order to answer my research question I make use of the indexes. However, separate regressions with the independent indicators have also been made in order see if anything significantly differs from what is expected. These regressions can be found in the appendix.
Treatment and control groups
Having both a treatment and control group serves two goals, as a statistical description of the characteristics of the women that are taking part in the courses and hold a credit at the organization, as well as a control for the similarity between the treatment and control groups.
Having a randomization of treatment would be an ideal case. That is to say, access to credit and participation in the courses would be randomly given to the women. This is not the case.
randomization of treatment implies that I cannot test my hypotheses through a simple comparison of the empowerment levels of the women. The level of empowerment or differences in outcome might be due to some confounding factors (Dunning 2016:5).
Table 3. Characteristics of the women in percent, divided in treatment and control group
Variables Credit No
credit t-test Course No
course t-test
AGE 0.1175 -1.5346
18-25 7.69 10.17 4.92 16.22
26-35 15.38 10.17 14.75 8.11
36-45 28.21 28.81 24.59 35.14
46 or more 48.72 50.85 55.74 40.54
EDU 0.3672 1.7270*
Primary 20.51 16.95 22.95 10.81
Secondary 35.90 37.29 37.70 35.14
Higher 43.59 45.76 39.34 54.05
CHILD -0.7448 -1.8021*
0 23.08 27.12 22.95 29.73
1 7.69 25.42 13.11 27.03
2 41.03 23.73 32.79 27.03
3 23.08 11.86 19.67 10.81
4 2.56 6.78 6.56 2.70
<4 2.56 5.08 4.92 2.70
HH -0.2015 -0.7985
Yes 64.10 66.10 62.30 70.27
No 35.90 33.90 37.70 29.73
ESTRATO 0.4674 -1.3286
1 12.82 11.86 8.20 18.92
2 30.77 27.12 29.51 27.03
3 41.03 42.37 42.62 40.54
4 15.38 18.64 19.67 13.51
Nr observ: 39 59 61 37
*** p<0.01, ** p<0.05, * p<0.1
Yet, having a control group with the same characteristics as the treatment group except the difference of having credit/and or taking part in the courses I can examine the impact that the course makes on women’s empowerment. If the treatment and control groups were similar in pre-treatment characteristics, this would most likely not bias the treatment result. In the Table 3 below I present the variables from the structure of the treatment and control groups. The different in characteristics of the respondents are portrayed by the percentage. The percentages are divided according to treatment and control groups. The treatment group of credit (CREDIT) consists of 39 women, and the control group 59 women. The treatment group for (COURSE) consists of 61 women, and the control group contains 37 women. In order for the treatment and control group to be valid there should be no significant difference
