Introduction

Non-governmental organizations (NGOs) are important policy actors that undertake activities such as running schools and local soup kitchens, protecting the environment, providing health services, and advocating for social justice. They bring the community together and facilitate collective efforts to pursue shared goals. When natural disasters strike—which is happening more often with climate change—NGOs are often the first to arrive for relief delivery and often continue to work toward disaster recovery and reconstruction.

Yet, like markets and governments, NGOs also suffer from organizational failures or dysfunctionalities. Take the case of Oxfam, a leading global NGO, which is well known for its humanitarian work. On February 9, 2018, the Times of London reported that Oxfam’s senior staff in Haiti paid earthquake survivors for sex on Oxfam premises.Footnote 1 Although Oxfam’s top managers in the UK commissioned an internal inquiry, they covered up its report, indicating governance failure at individual and organizational levels. Soon, there were multiple exposes about misconduct in other prominent NGOs such as Save the Children, Mercy Corps, World Vision, the Norwegian Refugee Council, and Medecins Sans Frontieres (Scurlock et al., 2020).

Failure can be observed at both the organizational and individual levels. Theoretically, organizations fail (or show dysfunctionalities) when conditions necessary for their efficient and effective functioning do not hold, or they do not function as per their missions. For example, markets work well when there are a large number of buyers and sellers with full information about products. However, when these conditions do not hold, markets suffer some level of failure.

What are the conditions for NGOs to work efficiently, effectively, and per their mission? The literature suggests that NGOs’ actions are guided by their social missions and not the organization’s or its managers’ instrumental considerations (Keck & Sikkink, 1998). Because the social mission is rooted in the community, NGOs represent and are accountable to the communities they serve. Accountability often requires that they mobilize resources from the local community; it is difficult to be accountable to stakeholder A when the organization derives most of its resources from stakeholder B. Finally, given the shared mission of serving humanity, NGOs are expected to work cooperatively (Cooley & Ron, 2002; Clifford, 2005;  Prakash & Gugerty, 2010b), differentiating them from atomistic, profit-seeking firms.

Following the “virtue narrative,” scholars argue that NGOs are critical in the policy process because of their close connections with the local community. NGOs’ selfless actions create social capital, which has spillover effects on democracy, good governance, and economic development. Thus, NGOs are a third way of governance (Giddens, 1998), reflecting a new type of politics beyond the state and the instrumental motivations guiding market exchanges. At the end of the Cold War, Salamon (1994) declared the onset of the global associational revolution, which he suggested was at par with the Westphalian nation-state system’s emergence in the seventeenth century. In 1998, United Nations Secretary-General Kofi Annan described civil society as a “new superpower.”

NGOs show some level of failure when these conditions do not hold. Drawing on the literature (Balboa, 2018; Edwards & Hulme, 1996; Johnson & Prakash, 2007; Martens, 2002; Morfit, 2011; Watkins et al., 2012), we identify four types of failures, which are inter-related. Further, some failures apply to any organization type, including governments and for-profit, while others are specific to NGOs. These failures are (1) principal-agency issues (agency failure); (2) managerialism and professionalization, which transforms NGOs from community actors to service delivery subcontractors (NGOization failure); (3) failure to work as per the preferences of the community they serve (representation failure); (4) lack of cooperative behaviors (cooperation failure).

Organizations consist of individuals, and this might require differentiating between organizational and managerial failures. The reason is that while most organizations probably have “bad apples,” their presence does not necessarily mean that the organization has failed. Take the case of governments and firms which sometimes employ corrupt individuals. Yet, individual misconduct does not lead us to condemn firms or governments as failed institutions. Scholars note that organizations begin to show failures when they internally (or in the external policy environment in which they function) do not have rule systems to detect and weed out bad apples (McDonnell & Rutherford, 2018). We suggest that the dominance of the “virtue narrative” probably creates conditions for monitoring and governance lapses, hindering the weeding out processes.

This paper offers a theoretical synthesis of the literature on NGO failure and not methodological strategy to measure it. We recognize that measuring organizational dysfunction poses conceptual and empirical challenges (Prakash & Potoski, 2016). Failure is a continuum, not a binary concept. Organizations typically face different levels of failures; the same organization might fail in one dimension but do quite well in others. All institutions, be it governments, markets, or NGOs, exhibit some level of failure. Thus, acknowledging NGO failure does not mean condemning NGOs as a category of policy actors. Just as a study of market and government failures provides ideas for their reform, we hope studying NGO failures will help reform the NGO sector to accomplish their social, political, and economic missions.

This paper has four sections. In section two, we examine different conceptions of NGOs. Section three discusses the concepts of market and government failure because they provide us with building blocks to conceptualize NGO failure. Section four identifies four (inter-related) dimensions of NGO failure: agency failure, NGOization failure, representation failure, and cooperation failure. Finally, in the concluding section, we identify issues for future research, including measures to mitigate NGO failures.

Defining and Describing the NGO Sector

Most countries have a history of civic activity outside the state system, undertaken via multiple actors (Dolšak, 2017; Martens, 2002; Salamon & Anheier, 1997, 1998). This diversity contributes to the multiplicity of definitions of what constitutes the civil society sector (or NGOs). The theoretical confusion (Vakil, 1997) is accentuated by different disciplines favoring different definitions. For example, the interest group perspective, which is well-established in political science, is not the dominant paradigm in sociology to examine civic actions.

Civil society or NGOs are defined by what they are not, as opposed to what they are. Scholars identify two defining features: the non-governmental character and the nonprofit status. These features are supposed to shape how NGOs emerge, raise resources, and govern themselves. NGOs cannot legally compel individuals to donate to them or participate in their activities, unlike the state.Footnote 2 There could be instances where a religious order (organized as an NGO) requires members to pay membership dues such as tithing. If members do not have credible “exit” options, the religious order shows features of a government instead of an NGO. Similarly, if a non-governmental militia compels residents living in a particular area to pay a tribute, this group acquires features of a state.

Broadly, the civil society sector includes (1) advocacy organizations and social movementsFootnote 3 that seek to influence public and business policy for broader societal goals, as opposed to particularist goals, (2) nonprofits that seek to supply goods and services, often to the underprivileged; (3) local community groups such as Bowling Leagues and Parent-Teacher Associations (Putnam, 1995) that provide local public or club goods, and (4) community groups (typically without a formal structure) to ensure sustainable use of common-pool resources by community members (Ostrom, 1990). This paper focuses on advocacy organizations and nonprofits only because they tend to constitute the modern NGO sector. Nevertheless, these organizations are the critical pillars of NGO research and the focus of the literature on NGO failure.Footnote 4

To explore NGO failure, we need to understand why NGOs emerge in the first place. Weisbrod (1988) suggested that service delivery NGOs (nonprofits) arise due to the twin failures of both markets and governments. Information asymmetries between the buyer and the seller cause market failures. While governments respond by providing these products, they fail when citizens have heterogeneous preferences about what government should do, how, when, and for whom. While the governments may meet the needs of the median voter, preference heterogeneity means that governments are not supplying products to the preferences of the non-median citizen: for them, the government has failed. As Tiebout (1956) noted on fiscal federalism, these citizens could exit the jurisdiction. Alternatively, citizens might look for other suppliers, namely NGOs, of this product. For example, instead of sending children to public schools due to unhappiness with the quality of education, families might send them to private or parochial schools that are registered as NGOs.

Advocacy NGOs can also be understood through the theoretical lens of “twin failures.” Governments are expected to enact regulations to correct market failures that emerge from externality problems such as pollution. Governments construct rules keeping in mind the preferences of the median voter or the most well-resourced ones. But this means that some citizens fail to receive the regulatory policies they desire, which they perceive as a government’s failure. When citizens believe that their inadequate political voice contributes to policy neglect, they seek a mechanism to make themselves heard. This is where NGOs, as policy advocates, come in. Working on behalf of their constituents, they tackle government failure by mobilizing, educating, and lobbying governments about citizens’ preferences on policy issues. A large number of NGOs in a given sector simply reflects the multiplicity of citizen viewpoints, analogous to the point Weisbrod had made about service delivery NGOs (Johnson & Prakash, 2007).

Hansmann (1980) offers another perspective on the emergence of service delivery nonprofits (but not advocacy NGOs). Firms seek to maximize profits for shareholders. While service delivery NGOs (or nonprofits) such as hospitals can generate profits, they do not have shareholders who can legally claim them. For Hansmann, this “non-distributional constraint” explains why consumers trust NGOs over firms in product categories where they face information problems in assessing product quality (such as education and health services). This is because consumers implicitly assume that in light of information problems, profit maximization on behalf of shareholders drives firms to do unethical things, such as providing substandard products or overcharging consumers. Thus, in the virtue narrative, NGOs are implicitly “good” (the behavioral logic), but in Hansmann’s approach, NGOs do good things because of the specific institutional constraint under which they function: non-distribution of profits (the institutional logic).

Salamon (1987) views market failures as motivating voluntary efforts to deliver public services (not advocacy), especially when these failures hurt the underprivileged. With the onset of industrialization and urbanization, social problems grew in scale and complexity. The voluntary sector simply did not have the means and expertise to address these growing challenges, including housing, health, or education. Moreover, because donors tend to support specific causes or population groups, some sectors or population groups are neglected. Thus, in Salamon’s narrative, governments intervene in response to voluntary failure, and not vice versa. Indeed, in his perspective, this probably contributed to the emergence of the modern welfare state.

Conceptualizing Institutional Failures

Theoretically, institutions fail when the conditions necessary for their efficient and effective functioning, as outlined in their ideal type, do not hold, and they do not work toward their missions. For example, most standard public policy textbooks (for example, Weimer & Vining, 2017) note that markets fail when: (1) there are too few buyers or sellers (as in monopolies), (2) products have public goods characteristics (such as national defense), (3) products create externalities (such as pollution) or show increasing returns, causing the divergence between social benefits/costs and private benefits/cost which guide private transactions (Baumol & Oates, 1975), (4) there are information asymmetries which do not allow buyers and sellers to assess product value before the transaction, and (5) costs of contracting, monitoring, and dispute resolution are high and restrict market exchanges.

There is an extensive literature on correcting market failures. Starting with Coase (1937), scholars suggest that firms emerge to economize on the transaction costs of arms-length market exchanges, a market failure. In perfectly functioning markets where commercial contracts can be developed and enforced at zero transaction costs, firms (as organizations) are not needed. But contracting costs cannot be wished away. Firms economize on contracting costs by replacing arm’s length transactions with hierarchical fiats about resource allocation (Williamson, 1973). Because firms are themselves susceptible to principal-agent conflicts (Berle & Means, 1932), they establish corporate governance institutions to mitigate these conflicts, as we discuss subsequently.

While firms address contracting failures, they do not address (sometimes even accentuate) other kinds of market failures such as externalities, public goods, information problems, and monopolies. This is where governmental regulations come in. Since the Depression of the 1930s, the notion of self-governing markets has come under severe challenge. Consequently, the government’s role in regulating economic activities has increased substantially. Although facing legitimacy problems of its own (Majone, 1999), the regulatory state is expected to enact and enforce regulations to correct market failures.Footnote 5

The belief that government interventions will correct market failure assumes that governments are motivated to act, are omniscient, and impartial. When these assumptions do not hold, governments fail through sins of omission and commission. Governments could fail to enact policies. Administrative agencies tasked with creating and enforcing regulations might have agendas that differ from that of the legislature. Governments can get captured by regulated firms (Stigler, 1971). When government regulations breed crony capitalism, government failure causes market failure instead of correcting it.

In 1979, Charles Wolf introduced the notion of nonmarket failure, analogous to the concept of market failure. For him, nonmarket failure encompassed both governments and NGOs (we disagree with this view and explain why below). In Wolf’s analysis, nonmarket failures arise because nonmarket outputs have distinctive characteristics, making it difficult to monitor the performance of the actors supplying them. These include: (1) supplying nontraded goods whose production is not guided by market signals, (2) product quality is difficult to assess because consumers often cannot vote with their dollars to signal their approval or disapproval of the product, (3) governments, in particular, tend to be monopoly suppliers, (4) unlike firms, nonmarket actors do not generate profits and therefore cannot make decisions about production levels. In Wolf’s view, other ways of preference aggregation to signal desired production levels, such as voting, function poorly.

Wolf’s key point is that governments and nonprofits suffer from an “internality” problem because organizational performance is evaluated by internal standards that cannot be benchmarked against market-based goals. In effect, the measurement problems drive nonmarket failures. As we note subsequently, Wolf has overstated the case about internal standards and puts too much faith in external standards. Donors, funders, and even rating organizations such as Charity Navigator routinely assess NGOs using external standards, and yet their measurement approaches may not correct governance challenges that NGOs face.

While Wolf incorrectly subsumed governments and NGOs under the same category, some elements of his discussion offer ideas for categorizing NGO failures. Drawing on the growing literature on NGOs’ dysfunctionalities, we offer a typology to understand NGO failures’ drivers and observable characteristics.

NGO Failures: Toward a Typology

The twin failure theory views NGOs as vehicles to correct market and government failures. But NGOs show some level of failure when they do not conform to the assumptions informing their governance design. These assumptions are: NGOs work for the community (and not donors), and their employees are aligned with the mission (agency issues); their activities reflect the needs of the communities they serve (representation issues); they function as organs of the community and not as professionalized organizations (NGOization failure); and they work cooperatively with each other to solve social, political, and economic problems (cooperation issues). Below we examine four NGO failure categories that have some overlap: agency failure, representation failure, NGOization, and cooperation failure. Given the variation in the functions NGOs perform, some NGO failures such as agency failure and NGOization failure are more prevalent among service delivery nonprofits, while representation failures are more salient for advocacy NGOs. Cooperation failures probably are equally salient for both.

Agency Failure

The modern NGO sector consists of formal organizations. The virtue narrative fails to adequately appreciate that NGOs face principal-agency issues like any other organization (Berle & Means, 1932). As locally rooted actors, NGOs are supposed to work for the community (Prakash & Gugerty, 2010a). If local rootedness is assumed to hold, NGOs should view the communities they serve as their true principals (Tocqueville, 1969 [1835-40]. But in reality, donors, the resource providers, tend to be viewed as the real principals.

Every organization requires critical resources to survive, which are often provided by external actors. Resource dependence theories note resource controllers’ influence over organizational structures and strategies (Pfeffer & Salancik, 1978). When donors reside outside the community, their lack of lived experience probably makes them inadequate judges of community needs.Footnote 6 Consequently, different incentives begin to shape organizational priorities (Abou Assi, 2013).Footnote 7 As Ebrahim (2005) notes, NGOs prioritize upward accountability to donors over downward accountability to communities.Footnote 8

Resource dependence issues are particularly salient in the international NGO sector (Khieng & Dahles, 2015). For example, scholars note the case of mission displacement. In the context of Malawi, Morfit (2011) notes how global AIDS funding has crowded out other health priorities for both governments and NGOs. In particular, the Gates Foundation, an NGO with enormous wealth and a technology-focused theory of change, has captured the global health agenda, often at the cost of local priorities (Birn, 2014; McCoy et al., 2009).

Monitoring costs also cause agency problems. Even if NGOs’ missions (or their donors’ priorities) were perfectly aligned with that of the community, it is not clear whether adequate governance systems are in place to monitor NGO action. This has two dimensions: internal monitoring by the board and external monitoring by donors. While (non-executive) NGO boards have a fiduciary duty to ensure that their organizations work as per their mission, there is considerable variation in how they interpret their roles. Scholars suggest that given the ambiguous accountability standards and unclear organizational performance measures, the level of board monitoring depends on how board members themselves interpret their role (LeRoux & Langer, 2016; Preston & Brown, 2004). Thus, the role played by NGO boards remains unclear.

Scholars also debate the role of external monitors. Firms have shareholders, the legally identifiable owners with a claim to the “residual.” Shareholders have strong incentives to monitor the firm. They are aided by the stock market and the army of well-trained financial analysts who carefully monitor firm performance and benchmark it against peer firms. For NGOs, donors have strong incentives to ensure that their donated monies are well spent (which leads to other problems, as discussed subsequently). Institutional donors such as governments and foundations typically outline performance criteria for service delivery NGOs—although it is not clear that the proliferation of accountability mechanisms leads to superior outcomes at the organizational or policy level (Park & Kramarz, 2019). While individual donors might have fewer tools to establish criteria and monitor NGO performance, online platforms such as Charity Navigator and Guidestar could be viewed as (imperfect) functional equivalents of financial analysts that work on their behalf. Thus, agency failures stem from NGOs viewing donors instead of communities as their principals, but not due to a lack of appropriate criteria to assess NGO performance. Given the critical role, donors play for service-focused nonprofits, agency failures are probably more salient for them in relation to advocacy NGOs.

NGOization Failure

External funding tends to create another type of NGO failure that transforms the NGO sector’s very nature. Scholars term this as NGOization (Alvarez, 2009; Hearn, 1998). While NGOs are supposed to be community initiatives managed by community leaders, the modern NGO sector has become professionalized with the attendant ills of managerialism. This is sometimes termed as the NGOization of the civil society sector, a phenomenon more salient for internationally focused service delivery nonprofits (Chahim & Prakash, 2014).

NGOization, in part, is an outcome of both governments contracting out public service delivery at the domestic level (Eikenberry & Kluver, 2004; Smith & Lipsky, 1993) and “aid fatigue” at the international level. Domestically, social service subcontracting begun in the 1980s as a consequence of the Thatcher-Reagan approach to limited government. It gathered speed in the 1990s with the “reinventing government” movement (Osborne, 1993). Internationally, donors saw little tangible progress in developing countries (Easterly, 2010) and found a politically attractive culprit: the governments of aid recipient countries (Gibson et al., 2005). Donors believe that NGOs have the motivation and knowledge to use foreign aid appropriately (Dietrich, 2013; Edwards & Hulme, 1996). Increasingly, donors are funneling aid to local NGOs via global charity chains (Bebbington, 2004; Wallace et al., 2007).

In charity chains, the key pillar of NGOization, donors typically contract with a prominent international NGO, who, in turn, seek to subcontract with local NGOs along with establishing local offices. As local NGOs compete to become subcontractors, they “professionalize” (Hwang & Powell, 2009). Donors emphasize “deliverables” and appropriate paperwork to prevent agency abuse (Baines et al., 2011). This is something that professional NGOs can handle. But professional NGO managers, as opposed to volunteers, tend to work within a different incentive structure. For them, NGOs are a career. They move between organizations, sometimes between government and the NGO sector, as we discuss below. Their focus is on bigger budgets, organizational growth, and other indicators of professional success.

Moreover, in a replay of Gresham’s Law, professional NGOs begin to drive out locally rooted NGOs. The former tend to be visible because they know how to work with media (especially social media) and receive publicity due to their international connections. In addition, their top managers probably have completed studies in elite western institutions and met international NGOs at conferences and professional development workshops (Kamstra & Schulpen, 2015). Hence, NGOization creates dualism in the local civic sector (Dogra, 2013) and marginalizes the homegrown, locally connected NGOs. As opposed to community-managed NGOs, the professionalized NGOs have fewer capabilities and incentives to understand and meet community demands, hence the failure.

Representation Failure

Previously we noted how resource dependence impedes NGOs from effectively working for the community. Some scholars challenge the notion of a community as an egalitarian, non-hierarchical collection of individuals (Agarwal & Gibson, 1999) because communities vary in the levels of their economic, linguistic, racial, and ethnic diversity. It is not clear that NGOs represent all interests equally (Salamon, 1987) or how they identify their “median” community member. Some suggest that advocacy NGOs, in particular, have an upper- or middle-class bias (Kamath & Vijayabaskar, 2009), undermining the voices of the poor and the minorities, a constituency that tends to be neglected by both the market and the state.

In many Western societies, upper-class white men tend to occupy leadership positions in leading NGOs. The Green 2.0 report documents the lack of diversity in US environmental NGOs.Footnote 9 Moreover, volunteers play an important role in many NGOs. Often only the middle or upper classes have the resources to volunteer their time, which has sizeable opportunity costs for less privileged households. Finally, to the extent advocacy NGOs influence policy agendas, they privilege certain perspectives within a community over others.

Salamon (1987) had also anticipated that certain NGO sectors or organizations might attract disproportionate resources. Donors may have idiosyncratic preferences that may lead them to focus on specific areas. NGOs favored by the rich and the powerful tend to get publicity and become identified as civil society representatives in the policy process. Consequently, inviting advocacy NGOs to sit at the policy table does not necessarily make the policy process more democratic or representative. Instead, it might reinforce the structural inequities and enhance the democracy deficit that NGOs are expected to correct.

Critical scholars see the NGO sector reproducing social inequalities and making citizens less willing to challenge the system. The Gramscian critics note that NGOs perform a system-maintenance function (Katz, 2006). The NGO sector’s status-quo bias is also encouraged by philanthropic foundations, many of which have been established by industry titans (Fisher, 1983; Francis, 2019). In sum, NGOs begin to show some level of failure when their agendas and advocacy do not represent the preferences of the underprivileged, who are often neglected by markets and states. This way, NGOs might reproduce inequalities and not playing the transformational role that is envisioned for them.

Cooperation Failure

Virtue narrative asserts that unlike markets, which foster atomistic, non-cooperative behavior and encourage instrumental actions, NGOs promote pro-social, cooperative behaviors that lead to the production of social capital (Keck & Sikkink, 1998). At a fundamental level, the cooperation assertion faces problems in pluralist democracies that encourage a clash of viewpoints via political parties, interest groups, or advocacy NGOs. Indeed, advocacy NGOs often represent different policy preferences: pro-gun versus anti-gun groups or pro-choice versus pro-life groups. While conservative NGOs tend to get labeled as counter-movements (Meyer & Staggenborg, 1996), their presence reveals that NGOs compete in pluralist systems, and cooperation is not necessarily a critical part of their institutional DNA.

But even ideologically aligned NGOs may not function cooperatively (Sell & Prakash, 2004). Scare resources that NGOs compete for—funding, grants, and media attention—are critical for organizational survival. Competition is more pointed for governmental and foundation grants, especially for service-focused nonprofits. In the context of the former USSR, Henderson (2002, p. 159) finds that ‘‘[g]roups…are not necessarily willing to share their grant ideas for fear that it would jeopardize their funding possibilities.” Cooley and Ron (2002) find that international aid agencies’ use of competitive tenders and renewable contracting encourages service-focused nonprofits to adopt firm-like competitive mindsets.

Both advocacy NGOs and service-focused nonprofits face another problem that encourages competitive behavior: low barriers to entry. In most countries, it is easy to establish an NGO. For example, there are at least 3.1 million legally registered NGOs in India, which is more than double the number of schools and about 250 times the number of government hospitals.Footnote 10

Resource competition means that NGOs have incentives to “capture” funding agencies. These dynamics are visible in government funding, which brings into question the assertion about their “non-governmental” character. When NGOs receive substantial resources from governments, as in many Western Democracies, they have incentives to capture the political system, especially the budget appropriation processes. Mosley (2012) reports NGO managers adopting several strategies to secure government funding, including lobbying. The budgetary capture is aided by the “revolving door” between NGOs and governments, with retiring politicians moving to top NGO positions and vice versa (Cada & Ptácková, 2014). This closeness potentially leads to another problem: reverse capture (Coen, 2005). Governments seek to “use” the favored NGOs for their agendas. In this way, NGOs begin to function as a de facto arm of the government, more interested in serving government interests and less focused on collaborating with other NGOs in the same sector.

Conclusion

In an indictment of the NGO sector, Sriskandarajah (2014), the secretary-general of Civicus, wrote: ‘We have become a part of the problem rather than the solution. Our corporatization has steered us toward activism-lite, a version of our work rendered palatable to big business and capitalist states. Not only does this approach threaten no one in power, but it stifles grassroots activism with its weighty monoculturalism.” If true, NGO failures have several unanticipated consequences for public policy. Perhaps the most glaring is how governments weaponize these failures to crack down on the NGO sector. Amnesty International notes that about 50 countries—about every fourth country—have enacted laws restricting NGOs (Christensen & Weinstein, 2013; Dupuy et al., 2015).Footnote 11 Governments cast NGOs as foreign agents disconnected from local communities (Hearn, 1998; Vogel, 2006). Managers of international NGOs often lead elite lifestyles and adopt a patron-client mindset when working abroad. In Angola, Peters (2013) reports that the European managers’ salary is typically three times of a local professional working in the same position, who probably has superior technical credentials and work experience. Pfeiffer (2003) notes how working for international NGOs in the health sector is often equated with winning a lottery. Cailhol et al. (2013) document how in sub-Saharan Africa, high salaries international NGOs pay caused an internal brain drain from the government health systems to the international NGOs.

Is this criticism of NGOs fair? Are not all institutions prone to failure because the assumptions about their ideal types do not hold? This is a fair point; all institutions are fallible. Yet, a focus on institutional failings could encourage institutional reform. Indeed, in response to some criticism, the NGO sector has adopted new governance mechanisms. Institutions can fail when their principals do not have low-cost mechanisms to monitor performance. The community members, the real principals, certainly lack such mechanisms. But donors, as resource controllers and defacto principals, might also face challenges on this count. While institutional donors can demand detailed reports from NGOs, most individual donors do not have such leverage—although the individual mega-donors surely do. This is where third-party rating platforms such as Charity Navigator play a role. They analyze NGO revenue, overheads, and expenditures information that US charities annually provide to the federal government on Form 990. Based on this information, Charity Navigator rates charities on how efficiently they manage their funds, penalizing them for high overheads, which are a proxy for agency problems. Potential donors can benchmark individual charities against others in the sector to assess how charities use their donated funds. Viewed this way, Charity Navigator mitigates information problems donors face and potentially controls NGOs’ agency problems. Yet, solutions sometimes create a new set of problems. In the context of Charity Navigator, it is not clear whether high charity ratings lead to more giving. Moreover, because Charity Navigator penalizes charities with higher overheads, it creates perverse incentives for charities to underinvest in organizational building over short-term results (Brown et al., 2017; Ling & Neely, 2013; Lowell et al., 2005; Sloan, 2009; Yörük, 2016).

Another important NGO innovation is the emergence of accountability clubs (Bies, 2010; Prakash & Gugerty, 2010a). NGOs seek organizational survival and growth. They might aim to address donor concerns about organizational performance proactively. They might also worry that scandals or media exposes about any NGO could damage the reputation of other NGOs in that sector. Anticipating such collective reputation problems (Grant & Potoski, 2015), NGOs come together to form self-regulatory clubs that ask members to adopt the best management and good governance practices. Some clubs also require that external auditors check members’ compliance with the club rules (Willems et al., 2017). Club membership could be expensive, but NGOs have incentives to participate in these clubs because their membership serves as a reputational signal to outside stakeholders. This allows the better-governed NGOs to differentiate themselves from others.

This paper has offered a theoretical synthesis of the literature on NGO failures. Future research should develop an empirical strategy to measure it, building on the growing literature examining both what types of NGOs undertake organizational evaluation (Mitchell & Berlan, 2016) and how they undertake it (Bryan et al., 2021; Carman, 2011; Hoefer, 2000). Broadly, measuring institutional failure poses substantial conceptual and empirical challenges. In some cases, scholars could identify “objective” dysfunction measures, while in others, scholars may rely on perception-based measures. For example, scholars calculate the loss of consumer welfare due to higher prices resulting from market imperfections in anti-trust issues, a market failure. Yet, Tinkelman and Donabedian (2007) note the problems of relying on financial measures alone to judge nonprofit performance. Charity Navigator outlines another way to measure nonprofit performance despite criticisms, such as the NGO starvation cycle and the difficulties of relying on overheads to assess performance (Gregory & Howard, 2009; Lecy & Searing, 2015).

Most measures of NGO failure examine how the inputs are used. As a result, there is a neglect of their outputs, outcomes, and impacts. Government agencies provide dashboards with information allowing citizens to assess how they use their inputs, what activities they undertake, what outputs they produce, and what outcomes result. Many NGOs, especially professional ones, follow this approach in their annual reports. Yet, output-based measures have limitations as well. If performance is assessed via specific indicators, the organization has incentives to focus on those metrics only, the problem of “commensuration” (Espeland, & Stevens, 1998). Therefore, community members must be consulted about the specific metrics to assess NGO’s organizational output. In this way, commensuration incentives can be mobilized to focus organizational attention on community needs. Thus, while recognizing its limitations, we believe the organizational output-based approach offers another possibility to measure different levels of NGO performance and failure.