Version classiqueVersion mobile

Microfinance challenges: empowerment or disempowerment of the poor?

Isabelle Guérin
Jane Palier

Part III - Assessing microfinance


Assessing the impact of microfinance in terms of empowerment: issues and challenges

Cyril Fouillet et Jane Palier

Texte intégral

  • 147 In the bottom half are those living below their country’s poverty line or below $ 1 a day (Microcre (...)
  • 148 See the data available from the 2001 Census on the website:
  • 149 See the World Development Indicators available on the World Bank website:

1According to the 2003 report of the Microcredit Summit Campaign, 2,572 microfinance institutions (MFIs) recorded, on a worldwide scale, more than 67 million clients with an outstanding loan. Among these clients, more than 41 million were regarded as poor people147 at the time of their arrival and 79 per cent, that means 33 million, were women. Many microfinance experiments carried out all over the world show that it is possible to set up financial services for the people “excluded” from traditional banking and financial institutions, and furthermore, to create profits, an important factor in the eyes of private as well as public investors. However, it is difficult to speak about “exclusion” when the absence of banking services concerns more than half the households. For instance in India, only 35.5% of the households use banking services148. The potential demand for microfinance in India is thus estimated at between 290 million (those below the poverty line149) and 650 million people (those who do not have access to bank accounts).

2Microfinance, as an alternative to the informal and the banking sector, engenders a keen interest among the development actors. Nevertheless, we must remain cautious. Up to what point, for example, is the support from public funds for microfinance programmes justified and under what conditions? Does microfinance induce an empowerment process for its beneficiaries? What is its effect on the living conditions of the men and women directly or indirectly affected by these programs? To increase the debt of millions of people is not automatically good news. Microfinance does not need preachers blinded by their beliefs, or their excessive goodwill. Microfinancial practices must be evaluated by the yardstick of their financial and social performances. To evaluate and measure all the changes in the life and environment of the people should be a step taken simultaneously with microfinance.

3Created as an assessment tool of a project, its design and its piloting, impact evaluation appears all the more useful and relevant in today's context of bitter competition. Even if impact evaluation expectations differ according to the actors (funding agencies, international donors, paid and voluntary staff of microfinance institutions), these evaluations prove to be essential in a sector of expanding activity.

  • 150 On that topic, see Guérin and Servet (2003).

4Today, the microfinance sector has established very good tools to measure and evaluate financial performance and to a lesser extent economic empowerment. On the other hand, there is still a long way to go to build comparable indicators in terms of social empowerment or social performance. Microfinance is considered by some funding agencies as a profit-making sector; the returns must be high to attract investment. Such an approach keeps microfinance far removed from one of these prerogatives: to be a financial tool serving to develop the capabilities of the most destitute people, while allowing them to become actors of their own life and to take part in public life. Credit is a factor of development only if it becomes a source of accumulation – or at least of income stabilization – and not of debt. While following a collective and solidarity-based logic, credit grounds the group in the collective and can become a factor of emancipation and power redistribution. But credit can also become an instrument of impoverishment and even bondage, and certain texts presented in this book demonstrate this150. The in-depth analysis of recent bankruptcies of microfinance institutions which, until recently, were believed to be firmly established, shows to what extent the negligence of social, cultural and political factors can lead in the worst case (Gentil 2002). Evaluation is thus necessary as much for securing the client's well-being and their loyalty in the long run as for the survival of the organization itself. To think of impact evaluation in terms of empowerment goes in this direction.

  • 151 We refer to the texts published in this Part 3 throughout this introduction. When we mention a cont (...)

5To describe all of the impact evaluation studies in terms of empowerment carried out in the microfinance sector is not very interesting. Indeed, the embeddedness of institutions in complex and diverse sociocultural contexts affects the various stages constituting an impact evaluation (1. Understanding the context; 2. Determining objectives; 3. Elaborating a global approach; 4. Developing indicators; 5. Selecting tools and methods; 6. Using impact data) (Simanovitz 2003). There are almost as many methods as programs. Nevertheless, in order to clarify this great confusion, we can establish some bench marks in order to understand better the main features which underlie the microfinance evaluation process, and in particular the impact evaluation in terms of empowerment. This introduction tries to go in this direction (1). The contributions brought together in this Part 3 reflect the diversity of empowerment evaluation methods151. Those obviously depend not only on the objectives pursued (What is one trying to measure? What is the subject evaluated) but also on the motivation behind the approach (Why carry out an evaluation? What is one really trying to prove?) and the human and financial means available. This last point involves asking who are the potential actors (Who is the sponsor? The evaluator? What is their relationship?). We will then try to understand what are the relevant indicators chosen to evaluate empowerment and what do they imply (2). This will lead us to illustrate the diversity and originality of the various approaches adopted by the authors of this Part 3 and to point out some of their most significant findings. The assumption mentioned in the general introduction of this book, namely that the link between microfinance and empowerment is far from being spontaneous and automatic, is largely confirmed. The results of the various studies nevertheless make it possible to go beyond this simple observation since they inform us about the obstacles. Once more, these texts remind us that access to financial services is only one tool of local development among others and that it is advisable to relativize and organize the many supposed virtues of microfinance.

1. Questions raised by the impact evaluation of a microfinance programme

1.1. Preliminary vocabulary questions: how to define an impact evaluation?

6The evaluation, whatever the sphere of activity, covers multiple meanings. The Development Co-operation Directorate (DAC) of the Organisation for Economic Co-operation and Development (OECD), worked out a definition of development aid programme evaluations adopted by the principal international aid agencies. Evaluation is defined as:

“The systematic and objective assessment of an on-going or completed project, programme or policy, its design, implementation and results. The aim is to determine the relevance and fulfillment of objectives, development efficiency, effectiveness, impact and sustainability. An evaluation should provide information that is credible and useful, enabling the incorporation of lessons learned into the decision-making process of both recipients and donors” (OECD 2002: 22-23).

  • 152 Fonds pour la promotion des Eudes préalables, des Études transversales (F3E) is a tool created and (...)

7This first definition can be supplemented by those proposed by F3E152 (1999) between results, effects and impact of a development project:

Box 1. Results, effects and impact of a development project

- Results: They are the quantitative and qualitative changes produced directly by the action. The results are in direct relation to the objectives of the project. It consists of comparing the results reached and the results defined by the initial objectives of the project.
Effects: They are the repercussions of the project on the physical environment and human surroundings. The effects combine the results of the project and other dynamics or constraints coming from the environment in which the project takes place. One understands by evaluation of the effects, a study related to the dynamics of change within the population concerned with the project.
Impact: The impact of a development project is the situation resulting from all of the significant and lasting changes, positive or negative, envisaged or unforeseen, in the life and environment of the people and the groups and for whom a bond of causality, direct or indirect, can be established with the development project. The impact evaluation is thus a judgment related to the dynamics of change within the population concerned with the project.

8Let us adopt a general definition of a microfinance impact evaluation: all of the processes whose goal is to determine which changes, significant and lasting, positive and negative, envisaged or unforeseen, occurred in a given space following the intervention of a microfinance institution (Fouillet 2003).

1.2. The evolution of microfinance impact evaluation practices

9As fundamental as it is, microfinance evaluation is, however, only very recent. One had to wait until microfinance suffered its first reverses in order to consider the need for impact evaluation and to question its advantages and its limits (Gentil 2002). At the end of the 1980s, the first microfinance impact evaluations were conducted according to methods, tools and with a precision responding initially and above all to academic requirements. These studies proved to be time-consuming and expensive, and, especially, were not very useful for the field operators. In the 1990s, a new approach to the impact described as minimalist (CERISE 2003:53) developed among the big networks. In this period, let us recall, microfinance is fashionable, but one is primarily concerned with the quantitative growth of the sector. The impact is then evaluated only on the basis of the institution's financial performance. The key words become: autonomy, financial viability and sustainability, overdue, outstanding loans, repayment rates, etc. One considers that a positive financial result means, therefore, the satisfaction and emancipation of the clients. This connection is obviously misused. The financial performance of a microfinance institution, whether it is positive or negative, is only one effect among others of a particular program. That an institution can survive autonomously is regarded as the best proof of its impact. This presupposition seems naive, to say the least. The number of clients does not tell us anything about the quality of services proposed, − all the more so when there is no competition in the sector.

10The first financial difficulties encountered by some microfinance institutions at the end of the 1990s started a new phase: one encounters a renewed interest in more operational analysis tools. The relationship between the evaluators and the subjects to be evaluated is reconsidered. The objectives of these studies and the indicators used become wider. There follows a proliferation of various studies seeking to show through analysis the effects in terms of income-generating activities, poverty reduction, the creation of employment, female emancipation, local democracy, etc. (Guerin 2003). The evaluation of empowerment becomes one of the top priorities of the studies.

  • 153 As an anecdote, on the Nabard Internet site (, one can read: “Did (...)

11The contributions of this Part 3 plainly fit in this perspective of going beyond the simple measurement of the number of people served by a microfinance program, the amount of disbursed credit or the rate of loan repayment. B. Thampi shows for example that a high repayment rate does not necessarily mean additional income from a productive activity: it can disguise the use of other sources of income (the husband) and credit (the money lenders) to repay the loans. In the same way, the high savings rate observed in the first years of the Community Development Society programme (Alleppey, Kerala) does not imply new savings habits (being able to reduce vulnerability), but rather a pragmatic use of the microcredit program. After 6 years, 62.5% of the women interviewed reduced their amount of savings after obtaining a bank loan via CDS. In the same perspective, L. Prasanna shows the limits of the minimalist approach (to focus only on financial performance) by analyzing the bad performances of SHGs and the many dropouts. In the context of a large increase in the number of SHGs created153 – growth which one can describe as unspontaneous (U. Oommen and A. Meenakshisundararajan), more especially as any group tends to be called a SHG (U. Kumar) – several authors of this Part 3 justify the interest in their study by the fact that not enough attention was given to social empowerment compared to the economic performances of SHGs. M. Indira uses more precisely the concept of social externalities, V. Rao speaks of social development, U. Kumar of socio-political empowerment. L. Devi and P. P. Pillai, insist on the insufficient acknowledgement of the role of institutions supporting SHGs in understanding the impact in terms of empowerment. Finally, L. Prakash is focused on the role and place of SHGs in Civil Society, in other words, on the impact that women SHGs have made on “local politics”.

1.3. The aims of microfinance impact evaluation

12When one tries to understand the aims of an impact evaluation, the analytical framework of David Hulme (2000) of a continuum of objectives between proving impacts and improving practice is very useful. In other words, does the impact evaluation aim at showing that a positive change took place in order to justify the continuation of funding? Or does it rather aim at redefining the practices of the microfinance institution? After having been regarded for a long time as paradoxical, these two extremes are now combined more and more. New methodologies seek to develop client analysis tools making it possible, at the same time, to prove the impact, but also to improve the practice, i.e. the capacity of the institutions to offer financial services adapted to the needs of the client.

13Most of the texts of this Part 3 tend rather to favor proving impact, even if their conclusions are obviously useful for the improvement of microfinance practices (via a better knowledge of the functioning of SHGs). However, two texts explicitly take the position in order to improve practices. H. Noponen shows the importance of a participatory monitoring and an impact evaluation system, the Internal Learning System (ILS) based on the experiment of PRADAN (Professional Assistance for Development Action), an organization promoting SHGs. As quoted in her paper, the medium is a multi-year pictorial diary suited to the illiteracy and poverty conditions of participants and to longitudinal perspectives of the process of development change. In the case of PRADAN, the tool was directed towards the analysis by the SHG women members of their living conditions in order to make more strategic decisions about the uses of scarce resources including credit. This type of tool (ILS) is clearly designed to improve programme operations. However, the data captured through the workbooks can yield quantitative data to be analyzed on a longitudinal or cross-sectional basis to assess the impact.

14M. Patole and F. Sinha present the approach taken by the EDA Rural System to evaluate the impact of 20 MFIs representing various models (group and individual banking) in several areas of India. The authors show that the approach of EDA rests on a triangulation of methods in order to benefit from the advantages of three different tools: the direction of change (Focus Group Discussions), the process of change in different contexts (case studies) and the assessment of that change (quantitative data). Moreover, it is a longitudinal study aiming at the same time “to document impact and to improve practices”. For this reason, they suggest strategies for microfinance institutions to be “more empowering”. They scored 16 MFIs on 10 empowering components (10 strategic features which support women as microfinance clients and other development support). Only 3 MFIs scored more than 0.5, they use the SHG model and are located in the South of India. The approach of EDA is unique: it does not aim at building an index of beneficiaries' empowerment as may be the case in other studies (cf. 2.) but an index of empowerment strategies carried out by MFIs.

1.4. When to carry out an impact evaluation?

  • 154 In this regard, the work of GRET is remarkable. See in particular Bousso, Daubert, Gauthier, Parent (...)

15Is there a rule about the moment when an impact evaluation must take place? The answer could a priori exclude the evaluations known as ex ante. How, indeed, to carry out an evaluation when the project hasn't begun? However, in the case of a longitudinal approach, an analysis ex ante can prove altogether useful, to establish, for example, the various poverty levels154. B. Thampi shows that one of the interesting features of the Community Development Society (CDS) program, Alleppey (Kerala) resides in its mode of identifying poor people using nine risk factors, like living in a kutcha house, no access to safe drinking water or a family eating less than two meals a day, etc.

16Programme impact evaluations are frequent when the operators wish to obtain a preliminary return on their work or the funding agencies on their financial assistance; thus, the strategies can be reconsidered, the intervention reorganized. The evaluative process can be considered half way when the duration of the project is determined in advance. The text of P. Labh is in accordance with this point of view and presents the results of a Mid-Term Assessment study of CASHE – a poverty-focused programme started by CARE in 1999 – carried out in 2003. It is also possible to establish a schedule or to carry out the evaluation continuously: one is then talking about a follow-up evaluation. Of course, an evaluation ex ante or half-way does not exclude an evaluation ex post.

17Evaluation studies are also distinguished by the opposition “instantaneous”/longitudinal approach. Whereas the “instantaneous” method obtains all measurements at the same time, the longitudinal approach takes measurements at various moments.

18The majority of contributions to this Part 3 are the fruit of studies undertaken by consultants or researchers in partnership with MFI or Self-Help Group Promotion Institutions but they do not belong to it. Their evaluations are by nature ad hoc and seek to point out dysfunctions, to show how dynamics of change can emerge or to compare the performances of various programmes and do not necessarily respond to an expectation initially expressed by the MFI or the funding agency.

1.5. The actors of impact evaluation

19Let us now question the role and the place of the three key actors of the evaluation process: the sponsor, the evaluator and the subject to be evaluated (Neu 2001).

  • 155 For example, EBRD: the European Bank for Reconstruction and Development.
  • 156 AFD; PRODEM; USAID; European Development Fund (EDF).

20First of all, the sponsor. Sometimes, a microfinance institution, aiming to improve the understanding of its own practice, can itself be the sponsor of the evaluation. In other cases, the sponsor is external, actual funding agencies (development banks155, development agencies156) but also potential funding agencies. It is, for example, the case of certain traditional commercial banks which wonder about the potential of the sector, either because they are in search of new niches, or because they want to upgrade their image. In the Indian context where the promotion and financing of SHGs are an integral part of government schemes, the sponsor can be a government agency. For example, the study undertaken by Binitha Thampi in Kerala was approved and funded by the Kerala State Planning Board. The strong involvement of the Indian government in the microfinance sector largely justifies their interest in impact studies. V. Rao considers, nevertheless, that though the state government of Andhra Pradesh has been spending lots of time, money, personnel and other resources on the group formation, there are not enough studies on the social development levels achieved by the women's groups in Andhra Pradesh. Let us note that this illustrate once again the lack of accountability of the majority of Indian government schemes. Lastly, the funding agencies of impact studies can also be public or private research institutions which finance an independent study, in the sense that there is no direct link between the sponsor/evaluator and the subject to be evaluated.

21Let us now introduce the second actor: the evaluator. His/her function is to assess the quality of the subject, to provide data regarding its state. The evaluator can be totally external to the subject and the authorities, but can also merge with them. Certain financial consultants work on databases sent directly by the field operator. In this configuration, the different actors never meet; the evaluator receives information and data on which it works, produces a report that it sends to the funding agency (who will take the decision whether or not to publish this information). On the other hand, there are evaluation mechanisms completely integrated into the subject and regularly interacting with it. Ultimately, the evaluators can be the operator him/herself (self-evaluation), a rating agency, a consultant, a research institution, as well as the clients, the local authorities, the local academics, social workers who belong to the third sector, etc.

22Lastly, the third actor in the evaluation process is the subject to be evaluated. The subject can be evaluated at several levels: micro, meso, macro. The last level is very rarely evaluated. The subject of the evaluation includes both the microfinance institution and its clients. One can speak about evaluation directed towards the client and evaluation directed towards the institution; clients being a person, a household, a group of clients, or a micro-enterprise. In view of the texts presented here and more generally in view of the Indian microfinance context, a distinction needs to be made: almost all the studies presented in this Part 3 take as their subject of analysis members of SHGs. They aim at analyzing the situation and the changes experienced by the members of SHGs considered as Community Based Organizations and not as microfinance institutions. Their studies relate only indirectly to Self-Help Group Promotion Institutions.

1.6. The relation between the different impact evaluation actors

23The actors immersed in an evaluation process thus do not always occupy the same position; the funding agencies can have a position completely external to the evaluation process and have trouble making a fair assessment of the complaints voiced by the field operators. Another configuration can see the subject to be evaluated and the evaluator merge; reality is not fixed and the situations are numerous and complex. The box below allows us to describe different types of evaluations:

Box 2. Types of evaluation

Internal evaluation: the sponsor and the evaluator belong to the same institution
External evaluation: the sponsor and the evaluator belong to two different institutions.
Exogenous evaluation: the sponsor and the subject to be evaluated are far apart.
Endogenous evaluation: the sponsor and the subject to be evaluated are close.
Self-evaluation: an institution evaluates itself without calling upon outside contributors
Assisted self-evaluation: the outside contributor is present during the evaluation, and is not a principal actor of this approach but an instructor, coordinator or adviser.
Participatory evaluation: throughout the approach, participatory methods are used.

  • 157 In the case where it is not a self-evaluation situation.

24D. Neu establishes a distinction between what he calls round evaluations and long evaluations. This distinction illustrates the degree of separation between the various actors of the evaluation configuration. In an extreme case, we have a wide separation between the funding agency, the evaluator and the evaluated subject. Decision-making is then established on a hierarchical basis, very codified: one speaks about a “bureaucratic model”. On the other hand, the degree of separation of the round evaluation is relatively low. The decisions more often proceed from a dialogue between the various parties; one tries to find a consensus, the approach is more comprehensive. This last distinction is relatively important, as the type of relations which the various actors maintain (round or long, bureaucratic or comprehensive) to a large extent conditions the evaluative process. On one side, the impact evaluation will be rigid and encounter problems in adapting its operational mechanism to the intervening changes during the process. On the other, unexpected elements will be debated and in the long run will be integrated into the approach. To illustrate the relations between the evaluator, the clients and the institutions, one can distinguish the evaluation according to whether it takes place with or without the institutions and with or without the clients/beneficiaries (cf. Box 3.). The closeness which is established between the evaluators, the field operators157 and the clients can support the dialogue, the detection and the resolution of problems with the help of innovations. H. Noponen emphasizes, for example, that the “bad scene/good scene pictures [part of the PRADAN's ILS workbook] have proved to be very popular with participants who use them for reflection and amusement, and with field staff who use them as effective ice breakers and discussion starters”.

Box 3. What relationships exist between the evaluator, the clients and the MFI?

With the clients but without the institutions: the objective of the evaluators is to collaborate only with microfinance beneficiaries. This approach recalls that of Laurence Touré (2003), when, in order to establish a neutral relationship with the clients, one does not inform the microfinance institution concerned about his/her field research.
With the institutions but without the clients: the development of a partnership between evaluators and institutions is then top priority. One can think of the integrated tools of data collection (continuous assessment) whose application allows a regular impact evaluation of the program's actions. Some organizations such as Micro Save Africa try to train the operators so that they can carry out an evaluation themselves, the goal being to increase self-evaluations.
With the institutions and the clients: both take part in the process of impact evaluation. Participative methods or PRA tools (Participatory Rapid Appraisal) are integrated into this approach (see the Internal Learning System presented by H. Noponen).
Without the institutions and without the clients: the evaluators remain completely isolated from the field. No participation is requested from the clients or the institution. This willingness to exclude the clients and the institution from the main part of the process may be compared to the financial analysis methods of consultants or rating agencies.

2. How to evaluate the empowerment induced by microfinance? (with special reference to SHGs in India)

2.1. The contributions of pioneer studies conducted in Bangladesh

  • 158 We think more particularly about the following studies: B. Ackerly (1995); A. M. Goetz, R. Sen Gupt (...)
  • 159 For this reason, reviews of literature on the extent of empowerment prepared, for example, by S. Ch (...)
  • 160 BRAC: Bangladesh Rural Advancement Committee; SCF: Save the Children Fund.

25We can now consider more precisely the indicators used to evaluate the impact of microfinance in terms of empowerment. Field studies seeking to establish proof of empowerment, but also showing the negative effects of microfinance, were mainly undertaken in South Asia, especially in Bangladesh158. Like the research studies presented in this part, the techniques, investigative methods and indicators used in Bangladesh were very diverse159. For example, Ackerly (1995), in her statistical study of 613 borrowers at BRAC, Grameen Bank and SCF160, built an indicator, Accounting Knowledge, to measure the probability that the changes associated with empowerment intervene. Goetz and Sen Gupta (1996) built an index of Managerial Control in order to classify the borrowers into five categories ranging from no control (no knowledge of the use of the loan or no contribution in terms of labor to the financed activity) to full control of the use of the loans (full control over the entire productive process, including marketing). Hashemi et al. (1996) carried out a quantitative analysis of 1,300 women clients of two microfinance institutions (BRAC and Grameen Bank) coupled with an ethnographic study of six villages and in-depth interviews. Their composite empowerment indicator is built from eight criteria: mobility, economic security, ability to make small purchases, large purchases, involvement in major household decisions, relative freedom from domination by the family, political and legal awareness, participation in public protests and political campaigns. Lastly, Rahman (1999) adopted an anthropological approach in a village covered by Grameen Bank using indepth interviews, participant observations, case studies and a household survey.

26These various studies, sometimes carried out with the same institutions (Grameen Bank or BRAC), have contradictory findings. The analysis of Goetz and Sen Gupta (1996) as well as that of Rahman (1999) are particularly critical. They show that the loans disbursed to the women often benefit the men and that the women have little control over the use of these new financial resources. The Rahman study (1999) shows the contradiction between what is officially said by the Grameen Bank (public transcript) and what is said by the field informants (hidden transcript). It shows that between 40% and 70% of the loans disbursed to the women are used by their spouse and that the tensions within the household increase (domestic violence).

27The study conducted by Hashemi et al. (1996) offers some more positive conclusions. They compare the impacts in terms of empowerment between the “minimalist” approach of the Grameen Bank and the “integrated” approach of BRAC. The first is associated with the paradigm of financial viability (as conceptualised by L. Mayoux 2000, 2003) according to which the access to credit is the determining element and trigger of empowerment. The second approach, also called the credit-plus approach, provides a whole range of services and fits into a more long term perspective, with the ultimate objective of acting on the structural causes of poverty (Poverty alleviation paradigm, Mayoux ibid.). According to the survey of Hashemi et al., BRAC has a more pronounced effect on women's mobility as well as on their participation in political campaigns and public protests, whereas the Grameen Bank has a stronger impact on the women's economic contribution to the household.

  • 161 N. Kabeer (2001: 79-81). S. Mahmud (2003) applies an analytical framework that recognizes the conce (...)

28N. Kabeer (2001) suggests that these conflicting conclusions are due neither to the time the survey was carried out, nor to the features of the microcredit programs, but to the questions asked by the researcher and the interpretations of the answers given; interpretation, in addition, being strongly influenced by the different understandings of intra-household power relations which these studies draw on. Those for whom the household is primarily a place of conflict and who are more interested in the processes of loan use develop negative conclusions, whereas those who consider the household as a place of co-operation and focus on the outcomes associated with, and attributed to, access to loans, have more positive conclusions. Starting from the testimonies of women borrowers, she proposes to see the household as a place of “unequal interdependence” and defines empowerment as the extension of the capacity for making choices, for taking actions, which express the women’s own values and priorities161.

  • 162 S.M. Hashemi et al.(1996) take great care to specify that the eight criteria making up their compos (...)

29Thus, there is no unique mechanism or one way to empowerment, especially one that is perceived differently by those who are the principal actors. It is also particularly difficult to evaluate: how to decide whether a person is empowered or not? Which criteria can be accepted, if not universally at least locally162, as relevant for measuring empowerment? N. Kabeer (2001) invites prudence in the judgments: autonomous individual behavior (strong decision-making, possession of goods in her name) of certain women can reflect, in fact, an increase in the tensions and violence within the household – “a form of divorce within marriage” (Kabeer 2001: 80) – rather than a real empowerment. Nevertheless, she specifies that it is very much the independent access of women to loans which brings about the emergence of new forms of cooperation and conflicts.

2.2. To evaluate the different dimensions of empowerment

30To conduct an impact evaluation, whatever the project, is to adopt a certain viewpoint. The direction taken will influence all the rest of the steps: conceptual methods and approaches, choice of tools. Nevertheless, the majority of the impact evaluations are the fruit of a compromise between several objectives, and this is particularly the case for an impact evaluation in terms of empowerment, insofar as this one matches up with necessarily different dimensions. However, these various dimensions of empowerment are not independent and are disconnected only to facilitate both the conceptualization and the assessment of empowerment. In the majority of studies carried out in South Asia and more specifically in India, economic empowerment, participation in the decision-making process, self-empowerment and socio-political empowerment are thus evaluated (Palier 2003).

2.2.1 Economic empowerment

31Economic empowerment includes the access to, use and control of economic resources (loans, income, budget) which implies the financing, creation and management of income-generating activities (and thus a widening of economic opportunities), the increase and diversification of income sources, the increase in consumption expenditure, savings capacity, access to property, etc.

32What are the results of the studies presented here in terms of economic empowerment? The study undertaken by EDA Rural Systems shows that women feel that saving in small amounts is convenient and safe and that 96% of women clients compared to 21% of women in non-client households contribute to household savings. L. Devi and P. P. Pillai analyze the economic impact of two microcredit programmes in terms of increased income (92% of the recipients reported an increase in income after participating in the program) and reduction of poverty (the proportion of households classified as poor declined from 92% to 48% in the case of the programme initiated by the Society for Rural Improvement and from 69% to 34% in the case of the “Kudumbasree” program). Then they evaluate the extent of women's control over income earned by them. 67% of the women reported that they had complete control, 20% partial control and 13% no control. L. Prakash reported in his study that before being in a SHG, 60% of the women had no control over the money earned by them, but after joining a group, this share declined to 15%. The number of those who had an active say in the use of the money earned increased from 11.4% before to 31% after being in a SHG.

33The text of B. V. Thampi, based on a study of a confederation of SHGs, the Community Development Society (CDS) of Alleppey in Kerala, more particularly illustrates the economic dimension of empowerment a. Let us look at the impact analysis of CDS in view of its main objective – to reduce poverty and provide economic security to poor women – after six years of operation. Out of the women who have taken loans, 58% reported they are not getting a steady income from the economic activities they set up. Among the 42% who reported a steady income, only 25% of the beneficiaries have seen a real improvement in their income. A further inquiry has been made to see the expenditure pattern of the enhanced income in the case of the 25%: the major share of women's income is spent on household needs like food, medical care for the children and their education. According to the author, these findings indicated the culturally constructed roles of mothers and the reproduction of patriarchal values, i.e. the phenomenon of contextual empowerment. She also argues that the orientation of programmes towards women failed to see the social context in which women take up these economic activities. The income-generating activities set up with microcredit (for instance, poultry farming or garment making) are highly gendered. This finding, combined with a high proportion of enhanced income spent for the children, is found to be a disempowering factor.

34U. Oommen and A. Meenakshisundararajan propose a descriptive analysis of the functioning of a dozen SHGs in the district of Kanyakumari (Tamil Nadu). They focus on the economic dimension of the SHGs by analyzing their financial performance. They establish a link between the diversification of economic activities undertaken within the group and the yearly per capita income generated by the SHGs. The study makes it possible to point out some of the advantages and dysfunctions characterizing a good number of SHGs. For example, the fact that the interest collected on internal loans is a source of common funds limits the dependence on moneylenders and the pledging of land or jewels as security. On the other hand, the principal activity of the SHG remains money lending for consumption, so that few jobs are created. In the same way, the fact that educated women, even those who are unemployed, do not participate in SHGs reduces the capacities of the groups in terms of self-management and therefore in terms of sustainability. Access to bank loans also remains very difficult for a majority of groups, some of whom are also not supported very much by the DRDA. Finally the authors suggest (by taking into account the wishes of women they met), a more active Government role in the development of marketing opportunities for SHG products. This question, already tackled in part 2, is indeed at the heart of the current concerns of SHG supporters (Guerin and Palier 2004).

2.2.2. Participation in decision-making within the household

35This indicator, favored in many studies, is close to economic empowerment: it is a question of evaluating decision-making in terms of household resource allocation in order to know who is responsible for the choices regarding productive investments (assets and land), but also daily expenditures. Decision-making is associated with the idea of being able to negotiate (bargaining power) within the household, whether it is considered a place of conflict or co-operation (Kabeer 2001). The analysis of decision-making power is often coupled with that of the improvement of the family's wellbeing by the increase in food, educational and health expenditure or the reassignment of certain resources (for example, equal meals taken by the children of both sexes and staying in school for the girls).

  • 163 This observation is in agreement with the results of a study carried out in Bangladesh, where the i (...)

36P. Labh presents the results of an impact study of the CASHE project in Orissa (started by CARE India in 1999). The methodology involved the use of control and experimental groups as well as “before/after” comparisons covering 1,300 respondents. Taking into account the difficulties encountered by the evaluation of the total income of poor households in rural areas, the study was focused on the evolution of expenses. The absolute amount spent on food expenses by the experimental group remained higher than that of the control group, but the percentage of expenditure on food is smaller (38% compared to 49%). Besides, the experimental group spent more than double the control group on education expenses. The average annual healthcare expenditure for the experimental group at Rs. 1,770 is significantly higher than the control group average of Rs. 1,238. It is also interesting to note that households in the experimental group spend a larger proportion of their money on marriages (14% compared to 4% by the control group) and a lower percentage on addictions (3% compared to 8%)163.

37In terms of reduced vulnerability, the access to loans through SHGs has allowed women to meet their emergency needs (food expenditure in times of crisis and medical expenses, notably in cases of seasonal illness like water-borne diseases or diarrhoea). The contribution of L. Prakash confirms this idea: before joining a SHG, only 9% of the women stated that they could have access to money in a crisis situation whereas later 38% reported they found it easy. In terms of decision-making power, the CASHE study shows that positive changes apply mostly to smaller amounts of money: decisions about the purchase of large assets and investments in agriculture are still largely taken by men. Likewise, B. Thampi shows that only 30% of the respondents reported that they have a major say in the household decisionmaking process.

2.2.3. Self-empowerment

38After the women's role in their household (in terms of financial independence and decision-making power), let's consider a dimension of empowerment which is connected to a critical assessment that the women focus on themselves.

  • 164 On this question, see for example T.K.S. Ravindram (1999) ou S.J. Jejeebhoy (2000). See also F. Jam (...)

39It is first of all about the control which women exert over their own body (mainly linked to sexual freedom and motherhood) and on their movements (i.e. their mobility)164. It also conditions other dimensions: how to develop one's own productive activity if one cannot sell one's products on the local market?

40Many microfinance programs, in addition to providing financial services, develop non-financial services like health, education, and awareness programmes (information about contraception or encouragement to denounce acts of domestic violence). L. Prakash approaches this dimension by evaluating the impact of SHGs on the status of women in the family with a focus on the following: running the house and managing expenses; deciding about the number of children; deciding about their daughter's future in terms of education, age for marriage, etc; and keeping control over income earned. Results of the study, which relate to four North Indian States (the sample size for the study was 600 SHGs, 3,000 women members and 250 facilitators) point to the urgent need for training and awareness programmes in SHGs, not only at the pre-formation phase, but during the formation and sustaining phases. The different trainings must be better articulated to constitute a true complementary and lasting device. He also observes some positive signs regarding the role played by SHGs in the area of health. Due to the awareness programs, some of the SHGs concentrate on preventive and promotional aspects of health and not just on curative ones.

41Organizing in groups can allow the setting up of spaces for discussion and collective action limiting isolation, self-absorption while promoting mutual support and aid. For instance, M. Patole and F. Sinha reported that unlike men, the feedback gathered from Focus Group Discussions and case studies shows that women clients like to be part of a group. They draw moral support from their peers and involvement in the group helps them in sharing their difficulties and relieving stress. L. Prakash notes that SHGs have provided a platform for dalit and tribal groups, not only in terms of savings and credit, but also to raise their voice effectively to highlight their issues.

42Self-confidence and self-esteem are particularly difficult to measure since they touch on the subjective dimension of empowerment (the power within). To estimate, one considers on the one hand the “objective” reinforcement of skills (leadership qualities, literacy rate, health care, management, etc). Many studies point out that very often only the group leaders developed management skills (for example, see Oommen and A. Meenakshisundararajan; V. Rao). M. Patole and F. Sinha notes with regret that only eight women (in a sub-sample of 92 women) knew all the answers to questions related to awareness about MFI procedures. On the other hand, one refers to the perception of one's own capacities (subjective dimension), to the feeling of having the right (often linked to becoming more aware) but also to be able to act and decide. The indicators used for evaluation are often connected with communication skills. For example, as quoted in the study undertaken by V. Puhazhendi, and K. C. Badatya and presented in 2002 at the seminar on the SHG-Bank Linkage Program:

“while most of the members (23%) did not talk freely in the pre-SHG situation, about 65 per cent of them expressed their desire to talk freely to others during post-SHG situations. In the pre-SHG situation, about 40 per cent of the members were speaking out only when asked, while it came down considerably to 9 per cent in the post-SHG situation” (Puhazhendi and Badatya 2002: 44).

2.2.4. Socio-political empowerment

43Contributing to the family’s financial resources can confer a greater legitimacy on women. Women's point of view (and more generally of all those excluded from the community) can become valuable and make it possible to pass over certain cultural and social barriers regarding equality among members of the same village or neighborhood. Empowerment known as “political” more specifically relates to involvement in local political institutions, participation in public demonstrations, the claim to rights, lobbying the authorities, etc.

  • 165 The same problem has been observed in other contexts, in particular in Africa (Guérin 1999).
  • 166 This lack of understanding of insurance mechanisms can lead to a certain amount of distrust among t (...)

44P.A.L. Prasanna evaluated this aspect of empowerment through a study undertaken in the districts of Medak and East Godavari in Andhra Pradesh involving 1,107 women members of 88 SHGs. She analyzes how far the constitution of SHG federations (thanks to the legislation APMACS, 1995) make it possible to reinforce the role and involvement of SHGs in local development. Besides having a considerable impact in terms of marketing the products sold by the group members, the SHGs, through federations, can question the power structure and establish their claims. She gave the example of lobbying the authorities who permitted the recognition of the SHGs formed and nurtured by NGOs in rural areas as recipients under the Deepam scheme (in which domestic LPG connections are provided to women below the poverty line at subsidized rate). She analyzes the advantages and risks of the internal group dynamics like those linked to inter-group co-operation. Some collective practices show an adaptation to the different needs of SHG members (internal loans for vulnerable sections and bank loans shared by members with regular income), whereas others show, on the contrary, the negative effects of the inequalities of intra-group income like the phenomenon of “debt rotation” (i.e. repaying the SHG loan by borrowing from outside). In another case, it is observed that the lack of agreement among members on the issue of which political party to support led to the group's failure. Inter-group co-operation also has perverse effects, like the “domino effect”. For example, due to lack of transparency in the management of financial resources, one SHG federation at the village level completely defaulted165. Similarly, an improper understanding about an insurance linked savings scheme led to default by several SHGs166.

45L. Prakash also mentions various risks associated with the development of SHGs. One of the challenges is to keep the SHGs away from the influences of political parties who would reduce the programme to a vote bank. In their enthusiasm to promote and support a large number of SHGs and their anxiety to meet targets, policy makers and resource providers run the risk of turning these groups into “state helped groups” or “target depicting” groups.

46Finally, M. Indira, while choosing to evaluate the impact of microfinance in terms of positive social externalities, uses an approach which recalls that of V. Rao and U. Kumar (cf. 2.3.) without building an empowerment index. Like L. Devi and P. P. Pillai in Kerala, she carries out a comparative analysis of two microfinance programmes in the district of Mysore (Karnataka) which differ according to their strategy and style of functioning. Both are state-led initiatives with the objective of women's empowerment: Mahila Samakhya (MS) is a quasi-central government programme and Shri Shakthi (SS) has been initiated by the DWCDRA (Department of Women and Child Development in Rural Areas). But whereas MS sees the goal of social SHGs being to create awareness and space for women and empowerment through transformation, SS addresses only the economic needs of the women by creating savings and credit groups. Her study shows that the participation of women in local government organizations helps to engender governance at the grassroots level (which is understood as a positive social externality). The results are systematically more positive in the case of the Mahila Samakhya program. She also points out several negative externalities: tension between men and women at the household and community level, limited reduction of the dependence on moneylenders, interference by political parties to influence the Anganwadi teachers in the Shri Shakthi program. Lastly, she recommends the combination of awareness raising and microcredit because “enlightened self-interest" makes members come together and actively participate if microcredit is part of the empowerment program”.

2.3. Building empowerment indexes

47Finally, there exist only a few studies trying to build a composite indicator that synthesizes various dimensions of empowerment, as in the case of the Hashemi et al. study (1996) in Bangladesh. In the first part of this introduction, we presented the empowerment index applied to MFIs by M. Patole and F. Sinha; we can now present three contributions of this part which attempt to measure the empowerment of microfinance clients/beneficiaries with the help of an index.

48V. Rao proposes to study the social development of women using a Social Emancipation Index inspired by that of Puhazhendi and Satyasai (2001). Seven inter-related and mutually influencing parameters are used to calculate it at the district level. The aim of this index is to give the direction rather than the precise and exact magnitude of empowerment, so that the Government can take active measures to improve the deficient parameter. It is thus conceived as a tool to aid in decision-making. The author makes several recommendations based on the results of a quantitative study of nearly 1,000 women from three different districts of Andhra Pradesh. For example, the women need further exposure to aspects such as sending girls to school and small families. The participation in local self-government is meager and membership in CBOs is negligible; only a few groups could achieve village development activities. Other groups need exposure to this aspect. Concerning the functioning of SHGs, V. Rao insists on the need for a periodic change of leaders and adequate training to generate leadership qualities in each member; still, it is necessary that the banks are not opposed to it (as quoted in U. Oommen and A. Meenakshisundararajan's paper, “the popular notion that the banks are ever willing to give loans to SHGs is a myth”). Like L. Prasanna, he also questions the value of an equitable sharing of the loans in the group. Lastly, he warns against the lack of Government and banking transparency in the choice of groups to finance and recommends that frequent interface meeting between group members and concerned officials be conducted.

49L. Devi and P.P. Pillai chose to conduct a quantitative study of nearly 200 women beneficiaries of two microcredit programmes in order to compare the effectiveness of NGO-led (Society for Rural Improvement – SRI) and State-led (“Kudumbasree” program) initiatives for women's empowerment in Kerala. The major difference is the style of functioning: the first one is a replica of the Grameen Bank of Bangladesh while the second is a State Government funded programme aiming at the empowerment of women through SHGs. After evaluating the economic impact of the two programmes (cf. 2.2.1) they finally built an Empowerment Index based on the respondent's answer to ten attitudinal questions relating to their self-esteem, self-confidence, autonomy and decision-making power. A majority of women beneficiaries reported a “high degree” of empowerment (a score above 70/100): 68% in the case of Kudumbasree and 42% in the case of SRI. Only for 27% (SRI) and 11% (Kudumbasree) of the women, did the microcredit programme not result in significant empowerment (a score up to 30/100). The main conclusion of their comparative study goes against some commonly held beliefs: the results are more convincing in terms of empowerment in the case of the governmental programme (Kudumbasree) than in that of the project carried out by the SRI NGO

50U. Kumar chose to develop an index to measure the socio-economic empowerment level of the SHGs. It has been applied to four selected SHGs supported by different institutions (classified as NGO, cooperative sector, civil society organizations) in order to test the role of intervening agencies in the empowerment process. Each of the 26 indicators chosen is ranked on a five point scale. The major result of these case studies is that the economic empowerment score is not very different from one group to another. On the other hand, the principal difference is due to different levels in terms of social empowerment: the SHG supported by NLA (Neo-Literates Association) records a much higher score than the others. The following explanation is proposed: the fact that it is a people’s organization committed well before microfinance to other socio-economic activities (literacy, fight against corruption, child labor, etc.) and one which operates on democratic principles. The members who have sufficient freedom and autonomy, along with the volunteers, seem to be at the forefront of the project. It thus supports the conclusions of M. Indira according to whom the strength of Mahila Samakhya in terms of empowerment lies in the fact that it is a “people’s program”. As quoted in U. Kumar’s paper, “consequently, one finds that it is the intervening institutions that utilize idle resources and not the people who utilize the financial resources for their benefit”.

51This contribution, as well as the others in this part, enables us to affirm that SHGs (and more generally microfinance) are a potentially effective empowerment tool, but that their virtues should not be over-estimated. The positive evaluated effects are encouraging, but remain limited and scattered all the same. It seems difficult to speak of an SHG movement: this movement lacks cohesion and integration in the wider social fabric to constitute a true social movement, capable of instigating social change on a large scale. Besides, one observes that certain SHGs borrow the traditional channels based on hierarchy, following the example of some initiatives from the so-called third sector (Georgeais 2004). The general conclusion reconsiders the question of the SHGs’ potential to constitute a social movement.


147 In the bottom half are those living below their country’s poverty line or below $ 1 a day (Microcredit Summit Campaign 2003).

148 See the data available from the 2001 Census on the website:

149 See the World Development Indicators available on the World Bank website:

150 On that topic, see Guérin and Servet (2003).

151 We refer to the texts published in this Part 3 throughout this introduction. When we mention a contribution on a particular point, it in no way implies that the other studies are not relevant, but simply that we have made a choice of presentation.

152 Fonds pour la promotion des Eudes préalables, des Études transversales (F3E) is a tool created and managed by international solidarity associations.

153 As an anecdote, on the Nabard Internet site (, one can read: “Did you know… more than 400 women join the SHG movement in India every hour. One NGO joins our mF programme every day”.

154 In this regard, the work of GRET is remarkable. See in particular Bousso, Daubert, Gauthier, Parent and Zieglé (1997).

155 For example, EBRD: the European Bank for Reconstruction and Development.

156 AFD; PRODEM; USAID; European Development Fund (EDF).

157 In the case where it is not a self-evaluation situation.

158 We think more particularly about the following studies: B. Ackerly (1995); A. M. Goetz, R. Sen Gupta (1996); S. M. Hashemi et al. (1996); A. Rahman (1999); N. Kabeer (2001).

159 For this reason, reviews of literature on the extent of empowerment prepared, for example, by S. Cheston et al. (2002); N. Kabeer (2001); Z. Oxaal et al. (1997) or G. Sabharwal (undated) are particularly enlightening.

160 BRAC: Bangladesh Rural Advancement Committee; SCF: Save the Children Fund.

161 N. Kabeer (2001: 79-81). S. Mahmud (2003) applies an analytical framework that recognizes the conceptual shift in emphasis in the empowerment definition, from notions of women’s greater well-being to notions of women’s choice and active agency in the attainment of greater well-being.

162 S.M. Hashemi et al.(1996) take great care to specify that the eight criteria making up their composite empowerment indicator cannot be applied in other contexts and has been built on the basis of their prior ethnographic studies.

163 This observation is in agreement with the results of a study carried out in Bangladesh, where the increase in the dowry amount appears to be one of the results of a microfinance project (Guérin 2004).

164 On this question, see for example T.K.S. Ravindram (1999) ou S.J. Jejeebhoy (2000). See also F. Jamet (2004).

165 The same problem has been observed in other contexts, in particular in Africa (Guérin 1999).

166 This lack of understanding of insurance mechanisms can lead to a certain amount of distrust among the women towards the NGO which supports these SHGs (Study conducted in 2003 in the Tiruchipalli district (Tamil Nadu) by J. Palier, IFP, report to be published in 2005).


PhD Research Scholar in Economics, Université Lumière Lyon 2 (France), French Institute of Pondicherry (India)

Economist, PhD student at the University Lumière Lyon 2 (France). She is also associated with the Institut de Recherche pour le Développement, Laboratoire Population-Environnement-Développement (IRD-Université de Provence) and with the French Institute of Pondicherry, where she spent one year conducting her field work.

© Institut Français de Pondichéry, 2005

Conditions d’utilisation :


Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search