Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

NGOs are essential actors in delivering aid programs and their purpose in this context is primarily to address the needs of beneficiaries. Accordingly, NGO accountability to beneficiaries in the form of efficient and effective programs and operations is essential. This is particularly so, given that funds are entrusted by donors to NGOs on beneficiaries’ behalf (Agyemang, Awumbila, Unerman, & O’Dwyer, 2009; Najam, 1996). Various researchers have noted the need for strengthening NGO accountability to beneficiaries (O’Dwyer & Unerman, 2008) in order to better meet beneficiaries’ needs. Specifically, benefits identified from developing NGO accountability to beneficiaries include increasing the poor’s sense of ownership of poverty alleviation projects, enhancing their self-esteem and confidence, and eliminating the risk of program fraud and inefficiencies (Mango, 2010).

Social audit is identified as one mechanism to strengthen NGO accountability to beneficiaries. Specifically, this process involves identification of and dialogue with salient stakeholders, developing performance indicators or benchmarks, evaluation of performance enabling continuous improvement, and public disclosure of findings (Ebrahim, 2003a). However, its application in practice is very limited (Agyemang et al., 2009). The value of social audit lies in it representing both a participation and evaluation process. However, conducting social audits involves both time and cost, particularly through the participation of multiple stakeholders (Agyemang et al., 2009; Assad & Goddard, 2010). Further, given the lack of prevailing regulation, the conduct of social audits remains largely voluntary. This results in self-selection bias, limiting the effectiveness and usefulness of this mechanism (Ebrahim, 2003a, 2003b).

NGOs engaging in microenterprise development programs (MED NGOs) have a strong focus on poverty alleviation through social and economic development. Specifically, MED involves working with and training the poor to develop their own microenterprises as a means to progress out of poverty. However, the participation of various actors (e.g. beneficiaries, companies, local governments) is central to link microenterprises with public sector resources and support, and private sector customers and suppliers in order to develop sustainable social and economic outcomes. While participation is central within MED, a similar participatory approach in the context of social audit is also essential. In particular, this approach has the potential to increase program effectiveness, enhancing MED NGOs’ accountability to beneficiaries (Bhatt, 1997; LeRoux, 2009; Schmitz, Raggo, & Vijfeijken, 2011). However, the nature of social audit practices within these NGOs has received only limited attention in the literature.

Accordingly, the objective of this chapter is to identify the prevalence, scale, and scope of social audits in MED NGOs. From this process, the benefits and constraints of conducting social audits are also investigated. The next section reviews literature on social audits as one accountability mechanism for NGOs—its characteristics and limitations. The research method section follows. The findings and discussion sections are then presented, reflecting on social audit practices of MED NGOs, suggesting a better distinction between social audits and other mechanisms of NGO accountability is required, and outlining a pragmatic approach to social audit to expand the scope of this practice.

2 Social Audits as an Accountability Mechanism

Within the NGO sector, Ebrahim (2003a) identifies five mechanisms for accountability, being reports and disclosure statements, performance assessments and evaluations, participation, self-regulation, and social audit. While tools such as reports and disclosure statements, performance assessment and evaluation are relatively well-developed in practice (serving upward accountability to donors), participation, self-regulation, and social audit (addressing accountability to a broader range of stakeholders) are noticeably less developed (Agyemang et al., 2009; Ebrahim, 2003a). As noted by various researchers (Jäger & Rothe, 2013; Mason, Kirkbride, & Bryde, 2007; Sinclair, 1995), however, accountability is perhaps most appropriately viewed as a web involving multiple dimensions and multiple actors. Thus, accountability mechanisms which extend to and incorporate a range of stakeholders are essential for a more holistic recognition of accountability.

Of the three less developed accountability mechanisms identified by Ebrahim (2003a)—participation, self-regulation, and social audit—social audit is somewhat unique in that it represents both a tool and process, combining participation and transparency. Participation focuses on meaningful involvement of various stakeholders to evaluate NGOs’ operations, reinforcing collaboration and co-ordination between NGOs and their stakeholders (Jordan & Tuijl, 2006; O’Dwyer & Unerman, 2010). Specifically, this process takes into account stakeholders’ (including beneficiaries’) views of an organisation’s goals and operations, in particular the impacts of an NGO’s activities on beneficiaries’ lives (Agyemang et al., 2009). Transparency evaluates the level at which stakeholders can access information on organisational procedures, structures, and assessment processes on a timely basis (Hammer & Lloyd, 2011). Through the social audit process, an organisation is able to evaluate its performance according to the expectations of stakeholders including the communities in which they operate, learn from feedback and disclose findings publicly, promoting transparency (Deegan, 2002). This process helps NGO achieve effective long-term program outcomes, strengthening their social responsibility and ethical behavior (Dawson, 1998; Mason et al., 2007; Owen, Swift, & Hunt, 2001). It also assits in reinforcing an NGO’s organisational legitimacy.

While NGO legitimacy is in part derived from compliance with legal requirements, in the context of social audit, legal and regulatory frameworks in both developed and developing countries are largely silent. Thus, typically NGOs’ wider legitimacy is morally derived from their social mission and performance (Slim, 2002). Intangible sources of legitimacy such as trust, integrity and reputation (Slim, 2002) are invaluable. However, how NGOs develop trust and demonstrate integrity is somewhat discretionary. While social audits are one such mechanism, the literature suggests this mechanism is not widely adopted in practice, in part due to lack of formal regulation.

As noted by Owen, Swift, Humphrey, and Bowerman (2000), the term social audit relates more to ‘taking a pulse’ of the organisation’s operations and effectiveness based on feedback from various stakeholders, rather than attesting to verifiable standards. Indeed, one of the limitations of social audit (considered further below) is the absence of such standards. Hence, social audits, while not legally required, can be a valuable resource—not only for an NGO’s reputation and legitimacy, but also for the learning it generates, and the operational benefits which ensue. As such, social audit practice is typically dependent on self-regulation (Gugerty, 2008; Gugerty, Sidel, & Bies, 2010), whereby the adoption of social audit rests on individual NGOs or donors recognising or identifying the need to engage in this activity.

In theory, social audit represents a tool for strategic planning and organisational learning. In practice, however, social audit has not been widely used due to several organisational level challenges (Agyemang et al., 2009; Dawson, 1998). These challenges include time and cost constraints (particularly if the audit is externally verified), as well as lack of agreed processes (e.g. systematic approach, use of appropriate indicators for performance evaluation) and experienced staff to conduct these audits (Owen et al., 2000). However, given NGOs engage in a wide range of activities, development of universal performance indicators or benchmarks for comparison or evaluation within the NGO sector remains a significant challenge (Dawson, 1998). Performance assessments typically involving quantitative measures are provided mainly to donors or organisational managers for decision making. Incorporating qualitative or non-economic outcomes generated from NGOs’ projects is often considered more complex (Jäger & Rothe, 2013). Other constraints and challenges of social audit include privileging the voices of some (more powerful) stakeholders over others, and encouraging those less powerful (e.g. beneficiaries who are often vulnerable and dependent on NGO assistance) to voice concerns (Kang, Anderson, & Finnegan, 2012). While these challenges are not insurmountable, the need for a pragmatic approach underpinned by the legitimacy of the process is highlighted, such that the social audit process is useful in serving the function of stakeholder accountability, rather than simply stakeholder management (Owen et al., 2000).

3 Social Audit in an MED Context

Unlike other types of NGOs (e.g. emergency help, healthcare service or education), NGOs engaging in MED have a strong focus on the social and economic development of poor communities by helping the poor to engage in income-generating activities, as a way of progressing out of poverty (Strier, 2010). In this context, a participatory approach to accountability has been promoted, such that beneficiaries are actively involved in various aspects of NGOs’ poverty alleviation programs. Specifically training and networking processes allow the poor to develop the skills and experience required to establish micro or small businesses (under NGOs’ guidance) and continue operating these businesses once NGOs’ support ceases, rather than being passive recipients of aid (Brown & Moore, 2001; Choudhury, Hossain, & Solaiman, 2008).

Importantly, however, effective MED programs rely on the participation of various stakeholders (Islam & Morgan, 2011; Peredo & Chrisman, 2006), consistent with the notion of socialising accountability (Roberts, 2009). Engagement with multiple stakeholders recognises the limited resources of NGOs, and the importance of connecting microenterprises established by the poor with both the public and private sector (Janvry & Sadoulet, 2009; Jones, Kashlak & Jones, 2004; Karnani, 2007). Such connections enhance program effectiveness by developing long-term working relationships with these groups which can continue once NGOs’ support ceases (Jones et al., 2004). Specifically, public sector support—both at the national and local level—is important to ensure participation in MED programs is encouraged and promoted, available public sector resources are identified and utilised, and the proposed business activity is supported (Karnani, 2007). Similarly, private sector buy-in is important to ensure connections with suppliers and buyers are made, and microenterprises’ goods are tailored to market demands, as the poor learn to operate sustainable (long-term) businesses (Jones et al., 2004). Thus, a collective approach is important for effective and sustainable (long-term) poverty alleviation program outcomes.

Given effective MED relies on the participation of multiple stakeholders, it is perhaps not surprising that social audit has particular relevance to this form of aid program. From a financial perspective, social audit allows donors to hold NGOs responsible for effective use of funds received through social audit reports and reporting systems (Deegan, 2002). From an operational perspective, however, the value of social audit extends to participation of beneficiaries and those in the local community (both public and private sector actors) to provide feedback and input and voice and concerns through a participatory process. However, as noted by Keystone (2006), few (26 %) donors expressly request that NGOs involve beneficiaries in developing performance indicators, and fewer still (5 %) showed an interest in discussing beneficiary feedback with NGOs. While creating an institutional culture among MED NGOs that encourages and values beneficiaries’ participation in social audits remains challenging (Kang et al., 2012), conducting social audits within the NGO sector often rests on the individual interests and priorities of donors or NGOs themselves. This effectively results in self-selection bias, limiting the effectiveness and usefulness of social audits more broadly as a control and evaluation mechanism (Ebrahim, 2003b). As such, competent and effective NGOs have strong incentives to conduct social audits and participate in self-regulation, whereas struggling NGOs may have little to gain from revealing their underperformance and ineffectiveness (Burger, 2012; Ebrahim, 2003b).

While the notion of socialising accountability reflects a sense of shared responsibility, social audit represents an opportunity for shared communication and understanding on what is perceived as effective, and what could be improved. Ultimately, however, without regulation requiring social audit in some form, the use of this mechanism rests on NGOs or powerful stakeholders identifying the need and allocating the resources for this task. Yet, given the power imbalances within accountability relationships between NGOs and other stakeholders (Assad & Goddard, 2010; Ebrahim, 2003a), it is unlikely beneficiaries or poor communities would be in a position to require or request social audits. However, from the perspective of more powerful stakeholders (e.g. donors), and consistent with the notion that with power comes responsibility (Keystone, 2006; Kilby, 2006), arguably, donors could require this process be undertaken on a regular (e.g. annual) basis, if NGOs do not initiate it voluntarily.

The focus on a participatory approach within MED NGOs has strong potential to increase the effectiveness of social audit in this context, yet the nature of this mechanism has received only limited attention in the literature. As such, it is necessary to understand what forms of social audits exist in practice, and their scale and scope. Before examining this issue, the next sections present an overview of the context and research method employed in this study.

4 Contextual Background: Bangladesh and Indonesia

Bangladesh and Indonesia were selected for this study as they are well known for their poverty reduction needs and activities. In 2012, total development aid provided by the Organisation for Economic Co-operation and Development [OECD] to Bangladesh and Indonesia was US$2,252 million and US$7,076 million respectively (OECD, 2014a, 2014b). Economic development initiatives (such as MED) are considered a central approach to poverty alleviation in both countries. However, despite the large number of poverty alleviation programs being established in both countries, success has been limited (Deen, 2010; Islam & Morgan, 2011). Bangladesh in particular remains one of the world’s poorest countries. In 2012, nearly 43 % of its population were classified as extremely poor (living on less than $1.25 a day) (AusAID, 2013a). Further, the unstable social and political environment of Bangladesh presents additional challenges for both the poor and NGOs trying to assist them (AusAID, 2013a; Islam & Morgan, 2011).

Unlike Bangladesh, Indonesia has a much higher Gross Domestic Product (GDP) per capita ($3,557 in 2012 compared to $1,679 for Bangladesh). However, economic growth has not benefited Indonesia’s population consistently, with poverty remaining a challenge. In 2012, more than 120 million (approximately 48 %) of Indonesia’s population lived on less than $2 a day, 44 million of which (18.7 %) were living on less than $1.25 a day (AusAID, 2013b).

Notably, public sector corruption in both countries has been identified as an issue. While Indonesia was ranked 114 among 177 countries, Bangladesh was considered to be one of the most corrupt countries (ranked 136)Footnote 1 (Transparency International, 2014). These high levels of inequality and corruption often result in low levels of trust among local communities when engaging with the public sector. This effectively weakens their voices, affecting the quality of dialogue with local communities (Ghuman & Singh, 2013; Warhurst, 2005). In particular, it likely hinders the quality of social audit processes aiming to consider local communities’ views, feedback, or complaints (Ahmad, 2008; O’Dwyer, 2005).

A high level of poverty and a large number of economic development programs implemented in these two countries presents a valuable context for the examination of MED programs and the use of accountability mechanisms such as social audits. Essentially, different social and economic contexts influence NGOs’ operations and approaches to achieving project outcomes and developing successful accountability mechanisms (Gibelman & Gelman, 2004; Islam & Morgan, 2011; Jordan & Tuijl, 2006). Accordingly, by examining NGO activity with respect to MED in these countries (which have made modest progress in alleviating poverty, with at times limited success), this study will provide valuable insights and understandings into social audit as an accountability mechanism and its barriers within NGOs’ practice.

5 Research Method

Given little information exists regarding social audit practices within the NGO sector, MED NGOs in particular, this study adopted an exploratory process and explanatory approach (Cavana, Delahaye, & Sekaran, 2001). Within this approach, this study was undertaken on the basis that “reality exists only in the context of mental framework” (Guba, 1990, p. 25). Therefore, realities are multiple, they exist in people’s minds and are constructed based on individuals’ social experiences (Creswell, 2009; Guba, 1990). Interaction with participants allows the researcher to adopt an interpretivist approach uncovering the realities constructed and held by participants, within the local and specific contexts that have given them meaning (Guba, 1990; Liamputtong, 2009). As such, this study adopted a qualitative approach, involving semi-structured in-depth interviews with NGO senior executives and beneficiaries in two developing countries where MED programs operate: Bangladesh and Indonesia. In addition, documentary analysis of publicly available data relating to the participating NGOs was conducted.

The latest directory of NGOs operating in each country is listed on the website of Directory of Development Organisations (2011)Footnote 2. Based on the list of local and international NGOs mainly focusing on MED detailed in this directory, there were 57 such organisations operating in Bangladesh and 31 in Indonesia as at 2011. Interview invitations were sent to all MED NGOs detailed in the list and senior executives of 20 NGOs (12 in Bangladesh and 8 in Indonesia, including local and international NGOs) accepted the invitations. The most recent annual report and other publicly available documents of the participating MED NGOs were reviewed to gain an understanding of their operations. Interviews of approximately one and a half hours each were then conducted with one to five senior executives of each NGO. In addition, one NGO operating as a donor (‘donor NGO’) providing funds to the Indonesian Government (which then funded NGO projects) was interviewed. With the support of the participating NGOs, ten interviews (six in Bangladesh and four in Indonesia) were also conducted with individuals and groups of beneficiaries from four NGOs (two in Bangladesh, two in Indonesia), in order to gain an understanding of beneficiaries’ perceptions of the NGOs’ projects and accountability mechanisms. Interviews with beneficiaries lasted approximately 30 minutes. A summary of the interview participants is detailed in Table 1.

Table 1 Summary of interview participants
Full size table

As noted in Table 1, of the 20 interviews conducted with NGOs, seven interviews were conducted with more than one executive (as a group interview) at NGOs’ request. In total, the interviews with NGOs involved 34 NGO staff (23 in Bangladesh, 11 in Indonesia), and were conducted in the cities Dhaka, Bangladesh and Jakarta, Indonesia, where the NGOs’ offices were located. Interviews with beneficiaries were conducted in rural areas (up to 100 km from the main cities of Dhaka and Jakarta), where the NGOs’ projects were based. Interview sites, both metropolitan and rural, provided the researcher with the opportunity to observe differences in the social and economic conditions of both countries (e.g. infrastructure, living conditions).

All interviews were conducted by the lead researcher in English. Interviews with beneficiaries were arranged with the support of the relevant NGO staff and a local interpreter (independent of the NGOs). Interviews were audio-recorded (with permission) and transcribed. NVivo was then used to assist with data analysis, allowing data to be deconstructed but also reviewed as a whole.

Thematic analysis was conducted to analyse the interview data. The coding process involved four stages, as suggested by Boyatzis (1998). As a first step, the structure of the interview protocol was used to identify general themes or nodes in Nvivo. These themes included forms of social audits, frequency of conducting social audits, who instigates them and how they are conducted, and the benefits and constraints of the process. Next, these themes were coded systematically for each transcript to maintain consistency. Codes were then refined to capture the essence of the data. Through deductive and inductive analysis, themes identified in the literature were refined and new themes emerged with respect to social audits employed in practice (e.g. variations in their form, scope and purpose). In the fourth stage, the themes were interpreted to identify the underlying meaning of the data. The findings were compared with publicly available data (e.g. annual reports, and other publicly available documents) and observation during the interview process, enabling triangulation. Comparison of the data sources provided clarification on the findings and helped to avoid bias (Guba, 1990), ultimately increasing reliability (Berg, 2009). The next section presents the findings, incorporating excerpts from interviews which were supported by secondary data, but provided significantly richer detail.

6 Findings

6.1 Prevalence and Forms of Social Audit

While social audit is a relatively well established term within the private sector (Bauer & Fenn, 1972), within the NGOs investigated, only two NGOs (both operating in Indonesia) acknowledged and understood the term ‘social audit’. The large majority (executives from 16 of the NGOs interviewed) expressed an understanding of this practice in more general terms such as monitoring, evaluation, or review processes that could be conducted by NGO staff or a third party. These three terms were neither completely distinct nor used consistently by interviewees. ‘Monitoring’ was commonly referred to as a process often conducted during the implementation of a project, and its results were intended to help NGOs learn from their current practices and improve their project design for better outcomes. Both evaluation and review processes were generally conducted at the end of projects. While evaluation was mainly used to collect data regarding the outcomes or effectiveness of the particular projects, review processes often involved a more comprehensive approach, encompassing multiple projects or the NGO's whole operations. For example, an executive of one NGO used the term 360-degree review process referring to reviewing its whole operations involving various projects, operational areas and stakeholders.

Of the 16 NGOs which adopted monitoring, evaluation or review processes (rather than social audit specifically), typically, they considered these processes as a form of social audit, but interpreted and applied them in different ways, with an emphasis on impact and final evaluation.

That is [an] impact evaluation obviously, final evaluation (Senior Executive NGO 4, Bangladesh, 2013).

…it’s the monitoring part, we call it monitoring and evaluation. We never call it social audit (Senior Executive NGO 2, Bangladesh, 2013).

Yeah, like a social impact assessment (Senior Executive NGO 6, Indonesia, 2013).

The remaining two NGOs (both operating in Bangladesh) did not conduct any form of review and considered their social audit practice to be ‘under-developed’ or not their key focus. Executives from both NGOs acknowledged it as a systematic process, distinct from financial audit, indicating some level of awareness regarding what a social audit represents.

No, we have no systematic social audit…but we have a very strong process in financial auditing. But we are thinking now about the social audit (Senior Executive NGO 11, Bangladesh, 2013).

Not yet, not yet…there is a system. The system is an initiative for social review. But that is not so strong (Senior Executive NGO 5, Bangladesh, 2013).

Further, the two NGOs that acknowledged using the term social audit were international and well-known with established operations. Thus, very few of the NGOs conducted social audits, however monitoring, evaluation and review process were more common. The NGO executives’ responses in relation to whether they adopt social audits (or variations thereof) are summarised in Table 2.

Table 2 Forms of social audit adopted
Full size table

6.2 Frequency of Conducting Social Audit

Findings indicate that NGOs had different timeframes for conducting social audits, monitoring, evaluation, or reviews, depending on a project’s design or requirements, or the NGO’s own purposes. With respect to project design or project requirements (typically required by and agreed with donors), executives from eight of the 20 NGOs indicated that their NGOs conduct social audits either throughout the project’s duration for projects which extend over several (e.g. 3–5) years, or when the projects finish for short-term projects (e.g. 1–3 years).Footnote 3 Thus, from a practitioner perspective, emphasis was placed on impact, often at the end of a project.

If it’s a 5 year project, we’d normally have a mid-term evaluation, then end-project evaluation. And if it’s a short 3 year project, we have an end-of-year evaluation (Senior Executive NGO 3, Bangladesh, 2013).

Yeah, if it is a 1 year project, they come only once at the end of the project. If it is a 3 year project, maybe they come in the middle of the project and also at the end of the project. It’s called Mid-term Evaluation and the Final Evaluation (Senior Executive NGO 12, Bangladesh, 2013).

It depends on the project design or budget that is given (Senior Executive NGO 1, Indonesia, 2013).

Further, executives from 11 of the 20 NGOs (seven in Bangladesh and four in Indonesia) noted that their social audit process (or variations thereof) was conducted periodically. Typically, this was every 6 months, annually, or longer intervals (up to 5 years); with annually (eight NGOs) being the most common time period.

…we are now doing once every year in every community (Senior Executive NGO 3, Bangladesh, 2013).

Evaluation activities, we do have 3–5 years, including from external [evaluation]…We always do like that, every 3 years or 5 years (Senior Executive NGO 1, Indonesia, 2013).

6.3 Conduct of Social Audits

Social audits (or variations thereof) were conducted by NGOs themselves, third parties nominated either by donors or NGOs, or government bodies. A summary of NGOs conducting social audits (or variations thereof) is presented in Table 3.

Table 3 Actors conducting social audits
Full size table

For NGOs conducting social audits as an internal audit process, the process and timeframes were often flexible, based on NGOs’ internal policies or initiatives (rather than imposed by donors or government).

Internally, we have the audit sector, our audit people are frequently visiting the area, and they’re providing feedback to me (Senior Executive NGO 4, Bangladesh, 2013).

[Conducting social audits] is [pretty] much every year. I mean we do our own internal [review]…we have this annual review process (Senior Executive NGO 8, Indonesia, 2013).

NGOs also nominated third parties to conduct social audits at the end of MED programs or during long-term projects.

…every 5 years, we do evaluation using external consultant to look at the progress of our activities (Senior Executive NGO 7, Indonesia, 2013).

…we agreed [to] appoint this independent auditor to audit our work…actually [we have] two [social audit] mechanisms, internal mechanism and external mechanism…For external audit we always do it at the end [of] the programs (Senior Executive NGO 5, Indonesia, 2013).

Typically, both internal social audits and external (third party) social audits were required by donors who wanted an independent view of the projects they funded in terms of progress achieved, and ongoing work required.

…evaluation by a third party, not the project team, third party, they’ll do [the social audit] on behalf of donor (Senior Executive NGO 2, Bangladesh, 2013).

If they want to go in-depth, then they hire one evaluator and they evaluate the program and they submit a report. As an example, like [one particular donor], after 5 years, they are doing some evaluations to find out what benefit we did in the area and what more requirements [are] needed (Senior Executive NGO 1, Bangladesh, 2013).

One NGO operating in Bangladesh noted that social audits were also conducted by a government body, with an emphasis on the NGO’s transparency and accountability.

And MRA, Micro-credit Regulatory Authority people is also coming, visiting our office, and they’re going to the field, so they’re checking…transparency is maintained, accountability is maintained (Senior Executive NGO 4, Bangladesh, 2013).

Thus findings suggest that within the NGOs investigated, the conduct of social audits (or variations thereof) was commonly an internal process. Importantly, however, both internal and external social audits were typically undertaken due to donor requirements within the broader project requirements as a condition of donor funding.

6.4 Social Audit Processes

Among the NGOs interviewed, social audits (or variations thereof) were commonly conducted using different methods, including focus group discussions, surveys or questionnaires, and case studies. These methods were used individually or collectively, depending on the NGOs’ purposes. The number of NGOs using these methods is summarised in Table 4.

Table 4 Methods for conducting social audit
Full size table

As detailed in Table 4, NGOs mainly used focus group discussions as a method for conducting social audits. Under each of the methods, facilitating the participation of various actors involved in and benefiting from NGOs’ projects, particularly beneficiaries, was considered important.

We use a methodology called focus group discussion (FGD). That is one and a half to 2 hour discussion with selected people and with representatives [e.g. community leaders, local government], and through that we collect this kind of information [on what these people think about the NGO’s projects]. And that is reflected in our annual planning (Senior Executive NGO 6, Bangladesh, 2013).

…the consultants will visit our programs and they will meet the beneficiaries, visit the villages, conduct interviews directly [with various] people (Senior Executive NGO 5, Indonesia, 2013).

Participants in this process included local partners (e.g. civil and government organisations and local private sector businesses) within the communities.

We sit together, we facilitate [review activities] so that they [actors within communities] can review [our operations] by themselves…where they are, what they need more from their [co-operation]. Did the [co-operation] work properly or not? So we get some answers from them, sometimes good, sometimes bad (Senior Executive NGO 7, Bangladesh, 2013).

[At the] end of the project, around two and a half years back, when we did the monitoring we set up questions, we interviewed some of our producers, suppliers, raw material suppliers, company staff… (Senior Executive NGO 2, Indonesia, 2013).

We do FGD with the communities, we do FGD with the partners—local organisations—their staff, we talk with the government and other local governments to understand what went well, what are their suggestions, what they think will be better for their own communities. So, this kind of thing we always do for learning and for improvement (Senior Executive NGO 10, Bangladesh, 2013).

Facilitating the participation of the above actors, particularly beneficiaries, was identified as a central approach to social audits (or variations thereof), by all NGO executives interviewed. One of the main reasons for this was to ensure sustainable (long-term) outcomes were developed for poverty alleviation programs, such that poor communities (with the support of local partners) had the capacity to continue operating microenterprises once NGOs’ projects finished.

We want the continuation of the project [such] that the community people will be in [the] driving seat, take the decisions, continue the work of the project after [us] phasing out of the project. If it is [managed], then okay. Somehow, it is satisfactory level. All the time, there are unsatisfactory [issues identified], and from the dissatisfaction our innovation will increase, our innovative work will increase. If there is no dissatisfaction, there will not be any good work [changes and improvement implemented] (Senior Executive NGO 4, Bangladesh, 2013).

However, the benefits (e.g. sustainable project outcomes) of social audits depended on the way that NGOs utilised the results of this process, and the beneficiaries’ perceptions towards this process. These issues are discussed in the next sections.

6.5 Social Audits as Performance Measures

While donors imposed reporting requirements on NGOs regarding economic and other quantitative measures (e.g. number of beneficiaries reached, amount of money spent), in the context of social performance reports and evaluations, indicators or measures of social performance were voluntarily identified and reported by NGOs rather than required or specified by donors.

The one that we have to do as mandatory is more on the income and then in sanitation, sanitation coverage…so ours is more focused on that…With social issues [the] main approach of the projects should have [appropriate measures] but it’s not a mandatory [part] of what we report on (Senior Executive NGO 6, Indonesia, 2013).

Interestingly, the donor NGO operating in Indonesia viewed social audit as equivalent to monitoring and evaluation and noted this was tailored to individual projects and NGOs.

…we have what we call monitoring/evaluation, and usually each project will develop their own evaluation system (Donor NGO, Indonesia, 2013).

Economic and other quantitative measures in the social audit reports prepared internally by NGOs were compared with evaluation or review reports prepared by government bodies (e.g. economic, agricultural departments).

…the report produced by the district [local government] management office, they get by talking with the communities (not through the NGO)…we can check and recheck [compare performance measures used between] the government and NGO. So the [government] people will have their evaluation [report on] the NGO but the NGO will also have their own evaluation [report] to give (Donor NGO, Indonesia, 2013).

6.6 Benefits of Conducting Social Audits

NGOs identified various benefits of conducting social audits, including sharing information with beneficiaries and actors participating in NGOs’ projects, and learning from the feedback.

Usually we use the participatory approach to collect and share information regarding this project (Senior Executive NGO 12, Bangladesh, 2013).

The benefit of course we will get what we call learning. Learning and also critiques and also warning; we call it warning about [what] we have done for one phase, like 3 years or 5 years, so I think that’s the benefit. We will get a lot of information in regards to how we do our work…people say, we need to change, what we do in the last 5 years is not really targeting our goals, so we need to change the other projects, for example (Senior Executive NGO 7, Indonesia, 2013).

…to learn about how the projects make changes in term of the outcomes, the objectives, and addressing the problem faced by the communities (Senior Executive NGO 5, Indonesia, 2013).

One NGO also noted that actions needed to be taken based on the social audit reports. If not, the process would be an ineffective tool.

…[based] on the review, there is scope to [act], to rectify, or to reduce. If it is done only for the theoretical words then both [parties will] decide they don’t feel interested. After the review, if anybody makes some comments, [partners] should take action…[beneficiaries would] give up if they find that nothing is happening after that. So it doesn’t work (Senior Executive NGO 7, Bangladesh, 2013).

To effectively address issues identified from social audits, NGOs used the results from this process to compare with their initial objectives or benchmark performance indicators.

…We have to have some baseline [initial indicators], and after that we compare all the baseline things to final things (Senior Executive NGO 2, Bangladesh, 2013).

…to come in and look at what we were doing, how we were working, whether we were achieving our objectives. So that kind of external evaluation…we found that very useful (Senior Executive NGO 8, Indonesia, 2013).

The results from this process were also used to improve NGO staff capabilities.

…and then we have our [staff] meeting, where we look at what’s come from the villages, and we talk amongst ourselves about things that have worked, things that can be improved in our work (Senior Executive NGO 8, Indonesia, 2013).

NGOs also emphasised using social audit findings to modify the designs of current or future projects for more effective outcomes.

Through this user survey, we know where we are. That last [project] we did, we conducted four, five user surveys—last 5 years, five user surveys. From the first user survey we found that our design needed to be modified. We did [this] (Senior Executive NGO 8, Bangladesh, 2013).

…on the basis of [social audit findings] we put together a plan for the next year. So there’s an internal review, an internal annual review process, which is quite rigorous (Senior Executive NGO 8, Indonesia, 2013).

Through the process of social audit, NGOs identified opportunities to be closer to beneficiaries, listen to their needs, feedback, or complaints.

…you are more people-friendly and people-oriented, you are reaching your targets [the poor]. We are here for the benefit of the poor. So if you are more people-friendly, you are doing more good to the people (Senior Executive NGO 6, Bangladesh, 2013).

Once the social audit results were reviewed and verified by NGOs’ field staff, findings were then discussed with communities to identify required actions, and report back to NGO managers. This process helped NGOs learn from beneficiaries’ feedback and complaints, potentially improving current practices and achieving more effective program outcomes.

There are some audit reports [auditors] are submitting to me, we’re sending it to the field level, and field level people are asked to [review and reply to] it. With that, they’ll find out if it’s [correct], or has been written [incorrectly] by the audit people. They’re collecting [comments] from people, and they’re sending again to us, then we’re sitting with the people [who give complaints] and the audit report with the audit people. Then we’re minimising [problems] (Senior Executive NGO 4, Bangladesh, 2013).

For NGOs focusing on poverty alleviation through economic development, reporting in a way that reflected NGOs’ performance in quantitative and qualitative terms was perceived as both challenging and rewarding, improving NGOs’ reporting tools.

Actually benefit [of qualitative data] is more…it makes our reporting system [more useful] where we can write the proper stories of quantitative information and qualitative information. We can include all the [relevant] things, the impacts [with] our own eyes (Senior Executive NGO 2, Bangladesh, 2013).

Importantly, the independence of external auditors in the social audit process was valued by NGOs and helped them access unbiased information on NGOs’ operations. These reports were used to compare project data collected internally by NGO staff.

That’s why, sometimes we have to be more [connected] to the audit sector, because the audit sector is independent. So, in both ways, it helps us. [Internal] program people are providing some sort of information, audit people are providing some sort of information. So, being a manager, we’re matching both information (Senior Executive NGO 4, Bangladesh, 2013).

Hence external social audit findings provided not only independent views but also an important tool for comparability at a management level. However, despite of the benefits derived from social audits, NGOs also faced constraints illuminated by this process. These constraints related to their own operations, beneficiaries, or local contexts, and are considered below.

6.7 Constraints of Conducting Social Audits

6.7.1 NGO-related constraints

Despite NGOs acknowledging the benefits gained from conducting social audits, they also indicated some major constraints. From an internal perspective, conducting social audits was perceived as costly and time-consuming.

When it comes to the socio-economic data, yes, definitely that’s always a challenge, because it becomes very costly to try and do that for every single household. So, there’s no way we can get that level of detail (Senior Executive NGO 2, Indonesia, 2013).

Another thing is time-consuming. [We] sometimes hired the monitoring team, evaluation team, and also we’re implementing the project, so sometimes we have to do some implementing and also do the monitoring, side by side. So this is time-consuming, time is the main factor. Sometimes, [it is] difficult to complete within the time period (Senior Executive NGO 2, Bangladesh, 2013).

In some cases, the NGO staff implementing programs were also data collectors for review and evaluation processes, leading to potential conflict of interest and data distortion or bias.

The program facilitators also visit the field, they’re also taking some time from the field level. But sometimes program facilitators are also reluctant [to collect data] because they’re implementers (Senior Executive NGO 4, Bangladesh, 2013).

While concerns about lack of independence were identified regarding social audits conducted by internal auditors, limitations were also acknowledged regarding social audits conducted by external auditors. In particular, external auditors often did not fully understand NGOs’ activities, or were not able to provide the same depth or level of understanding of some internal auditors due to time and resource constraints. This situation limited the benefits of the social audit process, particularly in terms of identifying areas for improvement.

They [external auditors] need to clarify again and again due to [being] outsider evaluators. So, this is our very big constraint (Senior Executive NGO 12, Bangladesh, 2013).

I think it’s difficult to find [external auditors] who really understand this work (Senior Executive NGO 7, Indonesia, 2013).

…so we’re not entirely sure, you know, what is the impact at every level in the value chain of a particular activity or initiative that we have supported, because [external] monitoring does not go deeply enough into the [value] chain (Senior Executive NGO 2, Bangladesh, 2013).

Another challenge within NGOs was that when areas requiring improvement were identified, NGOs sometimes lacked the capacity to implement changes.

So in the design of projects, there is a need for improvements, but when looking back there [were] things that we couldn’t have known more thoroughly [due to lack of capacity] (Senior Executive NGO 1, Indonesia, 2013).

6.7.2 Beneficiary-related constraints

From a beneficiary-related perspective, one of the main challenges was their perception of and participation in the social audit process. Given beneficiaries typically perceived that they had a weak negotiating position or wanted to please NGOs, they were often not willing to express any criticism (constructive or otherwise) in their feedback.

There is some distorted data. Sometimes they try to please us. So [what] they would say are not really their real thoughts; to share it, rather they try to say things which they think would please us (Senior Executive NGO 6, Bangladesh, 2013).

In some cases, beneficiaries wanted to receive or retain benefits from the projects, and tailored their responses accordingly, at times dishonestly.

Sometimes people are getting benefits, but they are saying we are not getting any [benefit]. Because if they are saying they are getting any [benefit], they have to pay the installments. There is a non-paying culture in this subcontinent. Not all the customers, like 5-10 %, or 20 % [of beneficiaries] are like this, and when you ask people are you happy…they will say, well we are happy with [activities] but we have some problems [that they need more time to repay investment loans or need more financial assistance] (Senior Executive NGO 8, Bangladesh, 2013).

Further, beneficiaries were often uneducated and lacked confidence to share information and provide feedback during social audits.

Then also sometimes, maybe this has been due to years of exclusion, years of isolation, years of not being respected of their opinions…some are reluctant to give feedback (Senior Executive NGO 3, Bangladesh, 2013).

…when you are talking to people who are essentially rejected, discriminated against, stigmatised, judged by society at large…even if you try hard to not do that, it impacts their ability to trust anybody else, right; to listen, to come up with ideas; their self-esteem as well (Senior Executive NGO 4, Indonesia, 2013).

Given beneficiaries were poor, they were typically pre-occupied with basic livelihood activities (e.g. family commitments and earning extra income to feed their families). As such, sitting with NGOs for lengthy review processes was considered a burden and less of a priority.

Mostly I think because of the time. Because you know the community, they’re very busy with not just their livelihood activity but also because of the family. If we asked them to come and sit together for 3 days or 4 days, it’s really difficult. They don’t have a lot of time like that (Senior Executive NGO 7, Indonesia , 2013).

This constraint was observed by the researcher during the interviews conducted with beneficiaries. Specifically, when asked about social audits, none of the beneficiaries seemed to show enthusiasm for this process. For them, social audits seemed more of an administrative process done for the benefit of NGOs, rather than a process for improvement, ultimately benefiting beneficiaries.

Yes, some staff visited…they checked IGA [Income Generating Activities] every week and took photos (Beneficiary 4, Bangladesh, 2013).

6.7.3 Context-related constraints

Compared to Bangladesh, Indonesia’s social and economic setting appeared to be better developed.Footnote 4 Such development plays an important role in the success and constraints of poverty alleviation programs and social audits. In particular, differing levels of participation by beneficiaries in the social audit process were evident. For example, executives of several NGOs operating in Indonesia indicated that, within an economic environment where the poor have more opportunities to participate in economic mainstream activities, there were often conflicts of interest between individual beneficiaries with working groups. This resulted in people (beneficiaries) being less motivated to participate in groups. This situation however, was less prevalent in Bangladesh where opportunities for the poor were limited, and participation by the poor in MED programs was typically more positive, resulting in less conflict within groups.

Manage the conflicts in a group. Manage the self-interests in a group. Manage the self-motive. So these are the some [boundaries] in development process (Senior Executive NGO 5, Indonesia, 2013).

I found that in Indonesia to make them solid as a group is difficult. This is our challenge (Senior Executive NGO 7, Indonesia, 2013).

However, compared with Indonesia, Bangladesh’s persistent political instability and less developed regulatory environment (with a high level of corruption) (Islam & Morgan, 2011) often limited the effectiveness of the social audit process.

Sometimes here in Bangladesh, lots of bad political situations we are facing. In that case we are facing the problem [of people in public sector organisations asking NGOs for money for doing things (e.g. to participate in social audits)] where we cannot complete our monitoring or evaluation on time (Senior Executive NGO 2, Bangladesh, 2013).

But the government officers go to these offices [of microenterprises] and seek grants and bribes and [we] suffer many problems…and sometimes you know, [they are like] the burglars (Senior Executive NGO 5, Bangladesh, 2013).

…for any informal microfinance, [we] have some sort of different types of risk. Risk means corruption…(Senior Executive NGO 11, Bangladesh, 2013).

6.8 Beneficiaries’ Perspectives of Social Audits

Amongst the 10 interviews with beneficiaries, in seven interviews it was noted that beneficiaries either participated in monitoring, evaluation, or review process; or were aware of this process within their NGO. Further, beneficiaries primarily referred to it as an annual process in their NGO.

We had discussion about how to perform certain things, how much was the loan amount, and how much was the savings (Beneficiary 1, Bangladesh, 2013).

…yeah, it was just about yesterday that we had an evaluation (Beneficiary 2, Indonesia, 2013).

The remaining three beneficiaries interviewed (all based in Bangladesh, all from the same NGO), however, did not seem to be familiar with or have an understanding of the terms social audit, social review or evaluation process; reinforcing that social audit is not a common practice, or does not involve beneficiaries more broadly in terms of participation or communication. This raises concerns regarding how NGOs conduct such evaluations, monitoring or review processes, and the extent to which beneficiaries are involved in or informed about this process. Table 5 summarises the responses of beneficiaries compared to NGO executives regarding forms of social audit adopted by the participating NGOs.

Table 5 Forms of social audit: comparison of NGOs’ and beneficiaries’ accounts
Full size table

For NGOs, social audits (or variations thereof) provided opportunities to engage in dialogue with beneficiaries on programs’ progress and impact. However, as noted by one Indonesian beneficiary, the meetings they were required to attend were perceived to be time-consuming and of little benefit.

Because they have to congregate in meetings out there…in farmer groups, in women entrepreneur groups. It’s a problem for them to congregate with others [for so long] (Beneficiary 3, Indonesia, 2013).

Further, the depth and value of the dialogue was at time questionable, as this beneficiary indicated that NGOs came and gave groups questionnaires rather than having in-depth interviews with beneficiaries in order to gain a deeper understanding of beneficiaries’ situations, issues and feedback.

…no-one ever asked this [e.g. one on one interview], instead [auditors used] a questionnaire (Beneficiary 3, Indonesia, 2013).

Thus, findings indicate that formal social audits are not widely used within the MED NGOs investigated; rather NGO executives refer to monitoring, evaluation or review processes conducted by both internal and external parties. This practice remains unregulated and is based largely on donors’ requirements or NGOs’ own initiatives. Through these processes, NGOs focused strongly on the participation of different actors involved in and benefitting from NGOs’ projects. These actors often included beneficiaries, local private sector businesses, and local government organisations. The social audit practices examined reveal both benefits and limitations. Reflections on how social audit is adopted, and compares with the literature on social audit regulation, is discussed in the next section.

7 Discussion

7.1 Distinction of Social Audits and Other Mechanisms for NGO Accountability

Returning to Sects. 6.1 and 6.2, findings suggest that social audit is not a well-established term among MED NGOs investigated. More commonly, these NGOs refer to monitoring, evaluation, or reviews as tools and processes to engage in dialogue with their stakeholders, particularly beneficiaries. These forms of social audits are conducted throughout the MED projects, when projects finish, or across multiple projects or activities of NGOs’ operations in order to assess effectiveness and consider opportunities for learning and change. The finding is similar with Mason et al. (2007) who suggest that the routines of social audit practice provide a process for achieving ongoing accountability. However, the practices of social audits identified from the findings are far less rigorous than the expectations of a social audit detailed in the literature. Specifically, within the NGOs examined, the conduct of social audits mainly rested on donors’ requirements or to a lesser extent, NGOs’ own interests or initiatives. However, none of the NGOs investigated published social audit reports on their website, suggesting transparency is limited. Rather, these reports were mainly used by donors or NGOs themselves. Further, conducting social audits at the end of projects (particularly short-term 1–3 year projects) and learning from social audit findings and results once projects had been completed, reveals results were of limited benefit for the projects reviewed.

The lack of distinction between social audit and monitoring, review, or evaluation, made by NGO executives suggests a blurring of boundaries, which are assumed by others in the literature. Ebrahim (2003a) for example, details five accountability mechanisms with social audits distinct from performance assessments and evaluations and participation, with performance assessments and evaluations argued mainly as being used for NGOs' internal accountability or accountability to donors. In the context of this study, forms of social audit (while blurred in practice) represented important tools for broader accountability to a range of stakeholders. However, reporting of social audits remained mainly to NGOs and donors. Specifically, this process was used for monitoring, evaluation and assessment involving participation of stakeholders through communicating, sharing and learning from NGO stakeholders’ feedback; in particular beneficiaries.

7.2 Participatory Approach to Social Audits

As mentioned in Sect. 2, NGOs (characterised as self-governing, non-profit seeking, charitable organisations) typically rely on funding from donors (Martens, 2002; Vakil, 1997). Thus it is not surprising that the conduct of social audits is often based on donors’ requirements (agreed within the project design). This process provides the opportunity to review the outcomes and impacts of NGOs’ projects, such that NGOs’ responsibility and accountability to a range of stakeholders (e.g. beneficiaries, donors, local communities) is addressed (Ebrahim, 2003a). As such, the participatory approach embedded in social audits facilitates participation of not only beneficiaries, but also the private sector and local governments. Their involvement and feedback through social audits (or variations thereof) helps NGOs to achieve more effective program outcomes so that beneficiaries, local private sector business, and local public sector organisations can continue working together after NGO projects finish.

Whilst focus group discussions were identified from the findings as the most popular method for conducting social audits, this finding is slightly different to the existing literature, which suggests questionnaire is the most common method (Owen et al., 2001). However, as noted in the findings regarding beneficiary-related constraints, beneficiaries are often not interested in participating in evaluation or assessments as they perceive these tools to be time-consuming and primarily for NGOs’ interests. As such, the findings of this study suggest that when social audits involve engaging in dialogue with stakeholders (including beneficiaries), facilitators should emphasise the importance of open and honest communication, constructive criticism and suggestions, and the importance of the process for both NGOs and stakeholders, including beneficiaries. This approach of sharing and learning from stakeholders’ feedback and suggestions is also preferred by beneficiaries, as they are more interested in the social audit approach where NGOs engage in dialogue (i.e. two-way exchange), listen and acknowledge their needs and constraints, to help them progress out of poverty.

7.3 A Pragmatic Approach to Social Audits

Returning to the central characteristics of social audit as identified by Ebrahim (2003a), and based on our understanding of how social audit is currently being adopted in practice, we suggest a modified, pragmatic approach may be most appropriate to promote the increased use of this tool within MED NGOs. The five characteristics of social audit identified in the literature and the modified pragmatic approach to social audits based on our finding are presented in Table 6.

Table 6 A pragmatic approach to social audit
Full size table

First, identification of and dialogue with stakeholders is important based on both the literature and the findings. Further, this process should encompass a range of stakeholders from different sectors, and with different power bases, emphasising the importance of two-way exchange such that stakeholders (including beneficiaries) understand the importance of giving open and honest feedback. With respect to development of performance indicators or benchmarks, given the very early development stage of the social audit process in the context of the MED NGOs examined, arguably individual benchmarks and indicators relevant to individual programs is an important first step. While the concept of universal benchmarks has been raised in the literature (Dawson, 1998; Owen et al., 2000), we argue a tailored approach has more relevance to individual programs, given that NGOs’ operations cover a wide range of activities. Thus, universal benchmarks will have less relevance and risk decreased learning, by potentially overlooking what is most important to individual programs. Continuous improvement, while an important goal, is perhaps somewhat subjective. Thus, we propose a more immediate objective of ‘organisational learning’ through a shared or collective responsibility which shifts the emphasis from accountability as a more traditional concept of being held responsible for results, to collective or socialising accountability to develop more effective (long-term) outcomes. Importantly, this objective should be acknowledged and supported by both beneficiaries and donors, to encourage leaning from failure rather than focusing solely on successful outcomes and incidences. Last, while the literature promotes public disclosure of social audit reports, given that none of the NGOs investigated provided social audit reports on their websites, and that many seemed to focus on reporting internally or upwards to donors, we propose stakeholder disclosure rather than public disclosure is a more realistic development aim.

While the proposed steps are a more simplistic approach to social audit (and do not extend to consideration of external verification, for example) they represent an important and practical development, which would help NGOs adopt a more formalised and accessible process. This is particularly important given the early development stages of the sector, its limited adoption of social audit, and lack of an inclusive stakeholder focus (i.e. specifically including beneficiaries in both obtaining feedback and communicating results) (Kang et al., 2012). Further, once adoption of social audit becomes a more widespread, institutionalised practice (Owen et al., 2000) involving a range of stakeholder input and broader stakeholder feedback, attention can then turn to standardising this process, sharing indicators and developing benchmarks, external verification, and communicating findings more broadly to the public at large.

From an internal NGO perspective, an appreciation of the need for feedback and organisational learning should be promoted. This is consistent with Mills and Friesen’s (1992) notion of the learning organisation (transferring learning, commitment to knowledge development, and openness to the outside world) being important for commercial organisations to survive and progress. While organisational goals may differ between commercial and non-commercial organisations (such as NGOs), the value of learning remains. In 2000, Owen et al. argued it was too early to provide formal regulation for social audits, contending that to strengthen accountability and transparency within the NGO sector, social audits should first be considered as a norm required within NGOs through self-regulation or donor expectations. Our findings in 2014 continue to support this view, given that monitoring and evaluation have become accepted practice, but have not necessarily progressed to formalised social audits. Thus, the challenge (and opportunity) presented to NGOs (and their stakeholders) is to create a culture which encourages and values evaluation, and recognises mechanisms such as social audit as a valuable approach to achieve this.

Social audits offer opportunities for learning and improving, however the process needs to take into account obstacles that can affect their effectiveness. Along with elements identified in the literature (e.g. cost, time, lack of agreed processes), the findings of this study suggest additional issues related to NGOs, beneficiaries, and the local context. NGOs’ projects are often perceived as charitable by beneficiaries. As such, their feedback on NGOs’ projects may be compromised or presented in a way that aims to ensure benefits continue to be received. In addition, changing beneficiaries’ perception of social audits is also important, such that they understand and appreciate this process is essential not only for NGOs’ purposes, but also for beneficiaries to benefit from more effective NGO projects. Regarding the local context, both the social and political environment are important influences (Gibelman & Gelman, 2004; Islam & Morgan, 2011; Jordan & Tuijl, 2006). Unlike Indonesia, in Bangladesh—a country struggling with a high level of poverty—the poor have fewer opportunities to participate in economic mainstream activities. Thus, working in groups and supporting each other to gain benefits from NGOs’ projects and beneficiaries’ collective actions is common, potentially increasing beneficiaries’ participation and contributing to effective program outcomes. However, the effectiveness of social audits also depends upon the participation of different actors, particularly local government and the private sector. As such, an unstable political environment and high levels of corruption (e.g. Bangladesh) can adversely affect the effectiveness of social audits.

Conclusion

While the findings presented in this chapter are limited by the relatively small sample from two countries, engaging with both NGO executives and beneficiaries provides valuable insights into social audit within two developing countries working to alleviate poverty. Specifically the benefits and limitations of this mechanism for strengthening NGO accountability have been highlighted, particularly for beneficiaries. Through these findings, NGOs’ social audit practices highlight the need to develop more systematic approaches to social audits (e.g. reporting to stakeholders) to distinguish social audits from performance assessment and evaluation and participation. As social audits emphasise the important role of engaging in dialogue with stakeholders, a participatory approach involving meaningful dialogue (e.g. focus group discussions) is essential. In addition, examination of current practice also suggests a more pragmatic approach to social audit. This approach encourages dialogue with various actors benefiting from and involved in NGOs’ projects as a first step, and tailored performance indicators or benchmarks (before considering the development of universal benchmarks for the NGO sector). It also emphasises stakeholder disclosure, rather than public disclosure. Further, in developing social audit processes, challenges need to be considered such as honesty and openness of beneficiaries, and the local context where NGOs operate. As social audit regulations within the NGO sector remain limited, the findings provide stakeholders and regulators with valuable guidance for better understanding the value of social audit as a mechanism to strengthen accountability of the NGO sector, particularly accountability to beneficiaries.