2 - Theorizing NGOs and the state

Like any other group or organization, the state is constructed and reconstructed, invented and reinvented, through its interaction as a whole and of its parts with others. It is not a fixed entity; its organization, goals, means, partners, and operative rules change as it allies with and opposes others inside and outside its territory. The state continually morphs.

(Migdal 2001, 23)

A veritable ‘associational revolution’ now seems underway at the global level that may constitute as significant a social and political development of the latter twentieth century as the rise of the nation state was of the latter nineteenth century.

(Salamon 1993, 1)

This chapter provides the conceptualizations and theories that form the foundation for the arguments contained in the book. It addresses NGOs’ effects on the state as a whole, as well as on four component “elements of stateness” – territoriality, governance, administrative capacity, and legitimacy. Existing theories of organizational placement, governance and privatization, civil society, and the social contract between a state and society each help explain the role NGOs play in state building, and the organizational and institutional change we have witnessed over time.

A large body of literature has emerged to address NGOs and their impact. Detailed research on NGOs began in earnest around 1990. This initial research tended to focus on describing or conceptualizing NGOs (Vakil 1997; Uphoff 1993), and on explaining the growth of the sector (Salamon 1994), particularly as a response to state and economic crisis (Bratton 1989; Sandberg 1994; Tripp 1997; Kioko, Mute, and Akivaga 2002). Articles at the time also provided broad analyses of state–society or state–NGO relations (Bebbington et al. 1993; Bratton 1989, 1990; Clark 1995; Kameri-Mbote 2000; Clark 1993; Whaites 1998; Sen 1999; Sanyal 1994; Cannon 1996; Campbell 1996) and issues of donors and accountability (Ebrahim 2003; Edwards and Hulme 1996b; Van Der Heijden 1987).

Starting in the 2000s, studies began to disaggregate the effects in particular sectors. They have looked, for example, at the role NGOs play specifically in the health sector (Wamai 2004; Palmer 2006; Ejaz, Shaikh, and Rizvi 2011; Pugachev nd; Pick, Givaudan, and Reich 2008; Pfeiffer 2003; Leonard and Leonard 2004), in the fight against HIV/AIDS (Parkhurst 2005; Doyle and Patel 2008; Swidler and Watkins 2009; Chikwendu 2004), in education (Rose 2011; Seay 2010; Leinweber 2011), in water service provision (Rusca and Schwartz 2012), in agriculture (Puplampu and Tettey 2000; Igoe 2003), and in municipal government (Keese and Argudo 2006), including slum upgrading (Otiso 2003) and sanitation (Carrard et al. 2009).¹ Importantly, academic studies have also shifted toward assessing NGOs’ role in directly political outcomes like governance (Grindle 2004; Mercer 2003; Swidler 2007; Brass 2012a), regulation (Gugerty 2008; Gugerty and Prakash 2010), policymaking (Batley 2011; Wamai 2004), local politics (Boulding 2014), and collective or cooperative action (Fearon, Humphreys, and Weinstein 2011; Desai and Joshi 2011).

Although these studies explored different aspects of NGOs’ effects in the political sphere, very little research has measured the real feedback effects of NGO provision of services on state development.² Scholarship on NGOs has been rife with assertions that such organizations undermine, threaten, or regularly conflict with the state in developing countries. While there are circumstances in which adversarial relationships occur, which are discussed later in this chapter, in many places there has also been a change in the relationship between the state and society. The relationship has become more complex than common arguments suggest, as NGOs and their government counterparts have learned that it is in their mutual interest to work together, collaboratively, most of the time.

As a result of this collaboration, NGOs have become part of the organizational makeup of the state in many developing countries. NGOs and government agencies share complex interactions, resulting in a blurring of the boundaries between the two sectors.

Conceptualizing the state

What is a state? Research focusing on the state has long been a core component of the study of politics, which we can trace back to the writings of the earliest philosophers. From Hobbes to Weber, Smith to Marx, Durkheim to Gramsci, much of the canon in the field examines the nature of the state. Though some periods of research have dismissed the importance of focusing on the state, eventually the state is “brought back in” to the discussion (Evans, Rueschemeyer, and Skocpol 1985), because an analysis of political life without it would be impossible.

Research approaches to the state have varied, however. Sometimes, our focus has been on considering the state as an autonomous operator or set of players, acting on its own accord (Allison 1971; Skocpol 1979; Johnson 1982; Haggard 1990); at other times, research has focused on the “embedded” nature of the “state in society” (Migdal 1988, 2001) or the societal actors in the state (Evans 1995). We have also looked at the state through the lenses of its formal rules, function, behavior, and institutions. Those researchers focused on the international arena have historically examined states’ actions vis-à-vis other states. Others are interested in relations between states and markets. Whenever we study democracy, law, economic regulation, ethnic minorities, human rights, or civil society, we are analyzing some elements of the state.

More recently, many political scientists have debated whether or not the state still matters; whether it is weakening in relation to private enterprises, international organizations, or certain strong philanthropic organizations; and how it fits into a networked political economy characterized by global interactions and exchanges (Keohane 1984; Ohmae 1990; Strange 1996).³ Looking at Africa in particular, scholars question the very “stateness” of African states (Migdal 1988; Doornbos 1990; Sandberg 1994, 7; Herbst 2000; Young 1994; 2004; Callaghy 1987; Jackson 1990; Widner 1995; Herbst 1996–97). My research, however, demonstrates that the state is important and questions this pessimistic bent, opting to view states not as static, ideal-type Weberian institutions, but as mutable, diverse, and ever changing. I elucidate the ways in which the state has changed and is changing in response to an important new set of actors gaining strength in many areas of the world.

“The state” is nonetheless a difficult concept – understandable in common parlance, but slippery when one attempts to grasp it firmly and pin it down for dissection. Part of the reason for this difficulty is that the state is not one-dimensional, but complex and multifaceted, making pithy definitions inadequate (Turner and Young 1985, 12; Kjaer, Hansen, and Thomsen 2002). In “Politics as a Vocation” (1919), Weber famously argued that the state is “a human community that (successfully) claims the monopoly of the legitimate use of physical force within a given territory.” In work published later, Weber elaborated on the formal characteristics of the modern state:

It possesses an administrative and legal order subject to change by legislation, to which the organized corporate activity of the administrative staff, which is also regulated by legislation, is oriented. This system of order claims binding authority, not only over the members of the state, the citizens, most of whom have obtained membership by birth, but also to a very large extent, over all action taking place in the area of its jurisdiction. It is thus a compulsory association with a territorial basis. Furthermore, to-day, the use of force is regarded as legitimate only so far as it is permitted by the state or prescribed by it.

(Parsons 1947, 156)

Yet Weber's definitions are incomplete because they fail to recognize that, in addition to the physical buildings, written laws, and administrative offices of the state, there exists the abstract idea or image of the state – the construct that we draw to mind when thinking about the state, as well as the practices of the state – or the processes by which the state acts day-to-day (Young 1994; Weber 1919; Migdal 2001; Englebert 2000).⁴ In general, our image of the state is the Weberian archetype, but very few states have ever achieved a perfect resemblance to this bureaucratic-rational ideal in their day-to-day activities or practices. As Migdal put it, “While the image of the state implies a singular morality, one standard way, indeed right way of doing things, practices denote multiple types of performance and, possibly, some contention over what is the right way to act” (2001, 19, emphasis in original). Recognizing the possibility of multiple practices and viewpoints in our definition of the state allows us to admit as states those entities whose image does not perfectly align with their actual practices. Where nongovernmental actors have become part of the de facto organizational makeup of the state, as in Kenya, such is the case.

A state with blurred boundaries: theories on NGOs and the elements of stateness

What does the state look like, when the line between NGOs and the government blurs? What social scientific theories help make sense of the changes? Four elements of stateness have been particularly affected by the growth of NGO activity, and are the foci of this book: territoriality, governance, capacity, and legitimacy.⁵ Several key questions as well as pertinent theories arise when considering these components of the state. These elements are discussed by turn in this chapter and summarized in Table 2.1.

Table 2.1. Defining four elements of stateness

Element	Definition	Key questions	Crucial theories
Territoriality	The demarcation, occupation, and defense of a fixed geographical territory by governing institutions. Where states act.	What factors draw NGOs to work in particular locations? What does the resulting distribution of organizations indicate for how the state broadcasts power over the territory it claims?	Organizational placement
Governance	The patterns or methods by which governing occurs within a state. How states govern.	What changes have occurred in the way decisions about service provision are made and implemented as a result of NGOs? How have the relationships between NGOs and governments shifted over time?	Collaborative governance, Co-production, Synergy
Capacity	The ability to implement stated objectives. What states implement.	How have NGOs affected administrative capacity? Do NGOs allow government officials to shirk responsibility to their citizens, as some claim? Or do they enhance the government's ability to do its job?	Civil society, Social learning, Institutional isomorphism
Legitimacy	A generalized perception that the state has a right to govern and its actions are desirable, proper, or appropriate in its cultural context. How states are perceived.	Do NGOs providing services undermine the legitimacy of the state, or does NGOs presence bolster the way people view their state? Do citizens respond to NGO or government service provision differently?	Social contract

Territoriality

“The state, to begin with, is a territorially demarcated entity” (Turner and Young 1985, 12). Behaviorally, “The state seeks to uphold its hegemony over the territory it rules” (ibid., 15). Territoriality combines characteristic with practice, as it refers to the demarcation, occupation, and defense of a geographical territory by governing institutions. It concerns the “broadcasting of power” throughout a geographical space (Herbst 2000), or what was referred to in the 1960s as the penetration of geographical territory by governing authorities (Herz 1968). Territorial boundaries demarcate the lines of hegemony along which public authorities and peoples can make demands on each other, such as taxation, defense, security, or accountability. States in many developing countries struggle with territoriality. Peru's government in Lima, for example, fought the Sendero Luminoso (Shining Path) in the 1980s and 1990s, and the government in Pakistan today struggles to maintain territorial control over its own Northwest Frontier area (Naseemullah 2014).

What relationship exists between NGOs and territoriality, particularly in places with “geographically uneven patterns of state building” (Boone 2003, xi)? Some scholars argue that NGOs undermine the state by providing services in places where the state is weak on the ground. I find, however, that NGOs assist the state in creating organizational presence in remote areas by putting in place activities that locals associate with state jurisdiction. Instead of supplanting, NGOs supplement. Under the right conditions, NGOs impact the territoriality of the state by providing services in places that the government has been unable to reach, particularly in arid, sparsely populated areas where the NGOs-per-capita ratio is at its highest. In small market towns and villages, people expect either government or NGOs to be present, providing services, and see the two types of organizations as substitutes for one another. This finding accords with Foucaultian notions about shared authority and sovereignty, where nonstate actors undertake governmental tasks in cooperation with the government (Foucault 1991, Sending and Neumann 2006, Biersteker 2012, 250).

Understanding this finding requires knowledge of the factors that draw organizations to work in particular locations. Because patron–client relationships dominate the distribution of public goods in many developing countries, it is logical to expect that national patrimonial chains of distribution would also sway NGOs’ distribution of services. Clientelist distribution is therefore one possible pull factor for NGOs. The literature on nonprofit placement identifies two others as well: need and convenience (Bielefeld, Murdoch, and Waddell 1997, Joassart-Marcelli and Wolch 2003, Nemenoff 2008, Bielefeld and Murdoch 2004, Peck 2008, Büthe, Major, and de Mello e Souza 2012, Fruttero and Gauri 2005, Stirrat 2008).

In Kenya, NGO location is associated with both need for services and ease of accessing a particular location (Brass 2012b). On a per capita level, an objective need for services appears to be paramount – there are more NGOs per capita in outlying areas far from the capital, which tend to be poorer and to lack services. More importantly for our understanding of the politics of NGO placement, the organizations do not appear to be geographically distributed on the basis of national patrimonial dynamics: we are no more likely to find NGOs in the home region or support areas of powerful politicians than elsewhere. Instead, NGOs are relatively more prevalent in places where the state is weakly articulated. This observation aligns with recent evidence that donors bypass governments and channel money through NGOs in order to maximize the effect of aid (Dietrich 2013), which, in the past, was politically targeted in Kenya (Jablonski 2014, Briggs 2014).

It is important to remember, however, that NGO presence is not distributed evenly throughout the country, so NGO impacts are also uneven. Areas with high concentrations of well-funded and well-managed NGOs benefit more than others. Yet NGOs and government both see the role of NGOs as “gap-filling”: that is, complementing the state.

Governance

Along with territoriality, a state is defined by issues of governance: the patterns or methods by which governing occurs. In formal language, governance is “the modes of social coordination by which actors engage in rulemaking and implementation and in the provision of collective goods” (Börzel and Risse 2010, 114). Behaviorally, the imperative for the state is to make decisions and to implement them independently of other organizations or authority (Turner and Young 1985, 15). For a state to truly have complete sovereignty, in principle, its representatives alone must make decisions about the presence, distribution, and operation of service provision in the society.

In weakly institutionalized states today, however, NGOs have begun to make governing decisions and to implement them. They often do so alongside their government counterparts, contrary to both the normative argument that government should “steer” the ship of state (make policy) while private actors “row” (implement policy) (Osborne and Gaebler 1992, see also discussion in Stoker 1998, Peters and Pierre 1998, 231) and the belief that government is eroding or becoming irrelevant to the governance process (Rhodes 1996, Rosenau and Czempiel 1992).⁶ These latter opinions arose initially in the UK, Europe, and Australia following a neoliberal turn in the West, during which policymakers, scholars writing from the point of view of New Public Management in public administration, and international financial institutions on structural adjustment called for “new” and “good” governance. They sought the streamlining of public service provision, favoring third-party contracting and outsourcing to ostensibly more efficient and effective private organizations. Underpinning many of these calls was the notion that markets, freed from the oppressive hand of state intervention, would be able to supply goods and services faster, better, and cheaper than governments (Edwards and Hulme 1996b, 961). NGOs, for some, came to be “euphemisms for the private sector” (Puplampu and Tettey 2000),⁷ considered more efficient, effective, flexible, and innovative than government (Bratton 1989, Fowler 1991, Owiti, Aluoka, and Oloo 2004, World Bank 1989). Although most writers and practitioners in this vein do not call for the complete removal of the government from all economic life, they do call for its role to be reduced to creating an “enabling environment” in which private organizations execute service delivery (World Bank 1989).

Such privatization may be fruitful in some contexts, yet in many developing countries we instead find that NGOs join public actors at many levels in making decisions and policy regarding service provision (Abers and Keck 2006). On the steering side, they sometimes sit on government planning boards; governments integrate NGO programs and budgets into local and national plans; and NGOs help to write state legislation. Additionally, on the rowing side, NGOs extend the service arm of the state to places and locations for which government counterparts lack sufficient funds; they also provide indirect services that the government is not able to provide. Often, NGOs work collaboratively with government actors on programs neither could run alone. Furthermore, by way of positive example, NGOs influence government offices and employees to improve their quality of services. Governance under these conditions is not the removal of government, or “governance without government” (Rhodes 1996, Rosenau and Czempiel 1992), but the addition and acceptance of other actors, including NGOs, in the process.

Theories of collaborative governance, co-production, and synergies facilitate our understanding of these changes. Collaboration can be understood as the “process by which organizations with a stake in a problem seek a mutually determined solution [by pursuing] objectives they could not achieve working alone” (Sink 1998, 1188 cited in Gazley 2008). Scholarship on collaborative governance in developed countries points empirically to networks of public, private, and not-for-profit organizations proliferating to replace markets and hierarchies as problems become more complex. Such scholarship considers, for example, nonstate actors increasingly taking part in the policy decision-making process; a blurring of the boundaries between public and private actors or the advent of multi-centric, decentralized decision-making and authority; and more participatory governing processes (Cleveland 1972, Stoker 1998, Peters and Pierre 1998, Rhodes 2000, Rosenau 2000, Frederickson 2004, Bevir 2006). Here, governing patterns have moved from featuring government as the governing actor to incorporating government as the primary actor within a network that also includes nongovernmental actors (Kickert 1997, 736).

These relationships are sometimes discussed in terms of “co-production” and “synergy.” The former involves outputs, services, or policies produced by individuals across a range of organizations – governmental and not – usually including citizens (Ostrom 1996). Programs are designed to integrate nongovernmental actors, and particularly those receiving the service, into the process (Ostrom 1996). Co-production, according to Ostrom (1996: 1083), is a form of synergy necessary for achieving higher levels of welfare in developing countries. Synergy, moreover, creates conditions not only in which civic engagement is more likely to develop and thrive, but also where bureaucratic organizations become stronger (Evans 1996, Lam 1996).

Two features stand out to support the idea that patterns in weakly institutionalized countries are more likely to reflect a merging of government and nonstate actors, rather than an “enabling” government that makes policy for nonstate contractors to implement. First, for a government to be able to effectively steer, it must be strong (Chaudhry 1993). The case of Kenya demonstrates that even countries with a mid-range level of development can lack this strength. In Kenya, Parliament has historically been captured by the executive presidency (McSherry and Brass 2007), and patronage politics impede a hands-off approach. The weakness of the steering capabilities on the government side has hampered the development of the ideal-type role distinction between an enabling state that delegates and the largely private actors who provide services. As Peters and Pierre observe, “For weaker states (or cities), joining forces with private sector actors has been an established strategy to increase their governing capacity” (1998, 233).

Second, the “hollowing” (Rhodes 1994) of the state in the Western context has involved intentional third-party contracting to nonstate actors, while the government retains control and provides funding. In many weakly institutionalized countries, however, the state has not deliberately contracted service provision to NGOs.⁸ Instead, places like Kenya have witnessed “spontaneous privatization,” in which nongovernmental actors try to fill the gaps left by the state without the state explicitly prompting them to do so (UN Habitat 2010, 14), sometimes following a government's inadequate response to crisis (Hershey 2013). Funding patterns accord with this description – analysis of the NGO Board's data from 2006 shows that only 1 percent of funding for NGOs in Kenya comes from Kenyan government sources. More recent figures are even lower: in 2009, only 0.25 percent of NGO funding originated with the Kenyan government (NGOs Co-ordination Board 2009, 32). Thus, while many developing country governments have come to rely on NGOs for some service provision, we should not assume that governments always initiate these relationships.

Administrative capacity

“[Thirdly] the state is…a set of institutions of rule, an organizational expression of hegemony” (Turner and Young 1985, 13). Behaviorally, this characteristic expresses itself as administrative capacity, the state's ability to implement stated objectives and to realize goals (Evans 1995, Finegold and Skocpol 1995), often in opposition to powerful societal actors (Migdal 1988). Capacity is a slippery term, and is sometimes dismissed as tautological. It is true that it is easier to observe its consequences, such as levels of service provision or economic growth, than to observe capacity itself (Kjaer, Hansen, and Thomsen 2002: 7), but we can still understand it as the ability to move from a written goal on paper to a vaccine provided, a road built, or a school opened. As Huntington puts it, capacity can be thought of as a country's “degree of government” – more government presence and output equals more capacity (1978, 1).

This definition is consistent with that of donor organizations, who see capacity as “the ability of people, institutions and societies to perform functions, solve problems and set and achieve objectives” (DFID 2008). Capacity acts in many ways as the interface between states and their peoples; people continually make demands on the state for greater standards of life, and states, if they have the ability to do so, respond to these demands, making for good state–society relations (Kjaer, Hansen, and Thomsen 2002).

In this book, the question addressed is whether NGOs enhance or undermine the state's administrative capacity to provide services. I find that when NGOs and governments work collaboratively, NGOs are able to influence government performance within public administration. Individuals and departments in government have learned from NGOs, and have begun to mimic the tools they have seen NGOs use successfully for years, calling for participatory development and civic education so that their agencies can better serve the community. This behavior facilitates accountability, therefore reflecting a very slow process of change toward democratic governance practices among civil servants.

NGOs act, therefore, as agents of civil society. Although some scholars critique the labeling of NGOs as “civil society” actors (Mercer 2002, Banks, Hulme, and Edwards 2015), NGOs display many civil society characteristics: most are autonomous from the state, undertake common or public goods activities, facilitate individual-level participation and group association, and emphasize citizen rights (Peterson and Van Til 2004). Autonomy is particularly crucial for civil society to achieve its goals, including pressing the government for change (Tripp 2001, 2000). Indeed, NGOs empower civic actors around the world – from the Balkan states (Grodeland 2006) to South Asia (Doyle and Patel 2008, Nair 2011), Africa (Atibil 2012, Heinrich 2001), and Latin America (Boulding 2010).⁹

Two seminal texts on civil society, Alexis de Tocqueville's Democracy in America (1835) and Robert Putnam's Making Democracy Work (1993), help us to understand NGOs’ role in capacity-building.

Tocqueville wrote about nineteenth-century America, where government administration was extremely weak, even “absent” (1835, 72), as it is in many developing countries today. Tocqueville observed that although the government acknowledged its obligations to society and had laws detailing service provision (Tocqueville 1835, 44–45), civic associations actually carried out many tasks:

Americans of all ages, all stations in life, and all types of dispositions are forever forming associations.…Hospitals, prisons, and schools take shape in that way.…In every case, at the head of any new undertaking, where in France you would find the government or in England some territorial magnate, in the United States you are sure to find an association.

(1835, 513)

In effect, Tocqueville perceived the nineteenth-century equivalent of NGOs as extending, or even composing, the social service wing of the state. As he saw it, nongovernmental provision of services in America allowed patriotism to spread through the new states of the West in churches, schools, and policing (1835, 293). Tocqueville thus witnessed in Antebellum America a blurring of state and civil society that nevertheless remained unrecognized in much of the literature that followed.

Putnam and his co-authors followed in Tocqueville's footsteps, arguing that the higher the degree of “civicness” in a society, the higher state performance will be (1993, 98). Although Putnam has been critiqued for not explaining the causal mechanism between civil society and government performance, scholars have argued that as horizontal relationships of trust and interdependence develop through membership in all types of associations, citizens become more involved in their larger society. These active democratic citizens insist on effective and responsive service delivery, and press their politicians to achieve it.¹⁰ Thus, civil society “reinforces a strong state” (1993, 182), by increasing democratization and institutional accountability. This conclusion is echoed in later studies (Evans 1997). Clark, for example, argues that state–NGO interaction is important precisely because NGOs represent “genuinely participatory development” (Clark 1995, 600). Harbeson (1994, 1) asserts that civil society is necessary for sustained political reform, state legitimacy, and improved governance, among other benefits. Civil society groups can act as a buffer, broker, symbol, agent, regulator, integrator, representative, or midwife between states and society (ibid., 24). Frank Holmquist's (1984) work on harambee groups in Kenya supports this idea, showing how the self-help groups enhanced state effectiveness.¹¹ Ndegwa (1996, 4) also shows that although NGOs don't always push for democratization directly, their “mundane activities” can clearly empower communities.

In more recent years, development practitioners, including those at the United Nations and the World Bank, frequently elide the use of the terms “civil society” and “NGOs” (United Nations Non-Governmental Liaison Service 2011, World Bank 2011, 1989, Hyden 1983). Brautigam and Segarra (2007) refer to this inclusion of NGOs as the “partnership norm” that the Bank first made explicit in staff guidelines in 1988. According to these researchers, staff members were told to encourage recipient governments to collaborate with NGOs (ibid., 152). Reflecting this faith in NGOs, the World Bank increased the percentage of its projects involving civil society organizations (CSOs), including NGOs, from 21 to 72 percent between 1990 and 2006, and to 81 percent by 2009.¹² The Bank highlights the role that nongovernmental actors play in public service provision in order to make them more democratic, accountable, and transparent.

This is not to say that service provision outcomes and administrative capacity have drastically improved as a result of NGO involvement. The fact remains that relative to the government, NGO resources are small and their impacts are often indirect. Civil society, moreover, is not without its limitations. Civil society groups in Africa fall victim to the same sorts of ethnic tensions and competition that exist within society as a whole (Kasfir 1998), and NGOs are generally run by elites in society, which can sway their activities toward elite interests (Ndegwa 1996). Neither are civil society groups, including NGOs, immune from corruption or patronage (Smith 2010). Many groups have grown dependent on their international donors, in some cases swaying from their civic intent in order to meet donor demands (Igoe 2003). NGOs are often not democratic organizations (Edwards and Hulme 1996b; Bebbington 1997; Edwards 2000), and are frequently more accountable to their donors than they are to the groups they aim to serve (Ebrahim 2003; Mercer 1999). Ironically, NGOs activities designed to “build capacity” through training of local government agents are often not efficacious (Swidler and Watkins 2009; Watkins, Swidler, and Hannan 2012; Barr, Marcel, and Trudy 2005).

Because NGOs are able to pay higher salaries, moreover, they can draw the most competent employees out of the public sector (Chege 1999; Schuller 2009).¹³ Pfeiffer (2003) reveals this process at work in the health sector of Mozambique, where NGOs pulled government health workers out of important routine tasks. As a result, treatment quality decreased, as did the productivity and morale of staff remaining in government employ (Pfeiffer 2003).

Legitimacy

Finally, states are defined by their ties to citizens, ideally in “the supreme loyalty of the overwhelming majority of the people…” (Turner and Young 1985, 13). Such loyalty is needed for state legitimacy, a generalized acceptance of the state's authority to rule. Stated another way, legitimacy concerns a state's right to govern, based on the perception of its citizens that its actions are proper or appropriate in its cultural context (Suchman 1995).

I find that where citizens have contact with NGOs, state legitimacy is either unaffected or is actually higher than it is in the absence of these organizations. Likewise, having positive views of NGOs does not correlate with having negative perceptions of government. NGOs help the state to fulfill its end of the social contract, in which citizens give up freedoms in exchange for security and services.

Developing states have long predicated their legitimacy on the distribution of services and promises of economic development (Young 1988; Bratton 1989, 1989b; Fowler 1991; Kanyinga 1996, 71; Schatzberg 2001; Owiti, Aluoka, and Oloo 2004; Johnson 2015; Osodo and Matsvai 1997; Jackson and Rosberg 1984a). As Julius Nyerere, the first president of Tanzania and a pan-African leader, said in 1961, “Freedom to many means immediate betterment, as if by magic. Unless I can meet at least some of these aspirations, my support will wane and my head will roll.”

NGO presence also suggests to poor, rural people, who have great desires but little real expectations from government, that someone in the world outside their village cares about them – this improves their optimism about the future, and, by association, their support for government. People are thankful that services have been brought, and their thanks are diffuse. There exists variation in this effect, however: urbanites are more cynical than their rural counterparts. Even among city-dwellers, however, interactions with NGOs are not systematically associated with negative views of government.

These findings run counter to some of the dominant theories on NGOs, which assert that NGO provision of services can erode the social contract (Schuller 2009), thereby undermining government legitimacy. Fowler writes, “Of the five imperatives that are a constant source of political concern to African governments, legitimacy is potentially the one most susceptible to NGO expansion” (1991, 78). In a widely cited article, Michael Bratton explains that NGOs threaten legitimacy specifically because African governments rely on promises of service provision and economic development as their moral basis for holding power (1989). Legitimacy can suffer if NGOs offer services that the government cannot match (Martin 2004). In post-conflict countries, this is often the case – donors favor NGOs for service provision, with several unintended consequences for state building: low levels of capacity within local governments, unsustainable facilities, and insufficient upward and downward accountability among service providers (Batley and McLoughlin 2010, 132).¹⁴

Many of these arguments rely on an assumption that citizens in developing countries make comparisons between NGOs and the government, such that effective NGO action reflects poorly on the government. World Bank and UN development expert John Clark suggests NGOs themselves make this assumption; he writes that NGOs might not want to improve service delivery if it brings positive returns to the government by increasing its popularity (Clark 1995, 596). Others assert that NGOs’ participatory approach mobilizes people, encourages increased information sharing, fosters alternative political ideas, and empowers the disadvantaged, all of which can threaten extant political legitimacy, power, and order (Bratton 1989a, Fowler 1991, Boulding and Gibson 2009 citing Putnam 1993 and Putnam 2000, Martin 2004).

My findings, however, show that positive popular responses to NGOs do not equate to negative views of government, at least not in much of Kenya. Other recent research across Africa and in South Asia has come to the same conclusion (Sacks 2012, Dietrich and Winters 2015). If we consider NGOs to be part of the organizational form of the state, it is logical that NGOs do not undermine state legitimacy. Especially in rural areas, many people associate NGOs’ good deeds with their local administrators or politicians, who are often credited with having brought the NGOs to the community. This makes sense in patrimonial societies, where “big men” who have access to the resources of the state (as civil servants or politicians) distribute these resources among their clients in exchange for decision-making authority.¹⁵

This finding may be a mixed blessing, however. In democratic polities, NGOs bolstering state legitimacy is probably beneficial for all concerned. In authoritarian regimes, however, NGOs may (inadvertently) placate demand for change. NGOs may provide an excuse for the government to withdraw from service provision (Campbell 1996) or to become dependent on nongovernmental and international actors (Englebert 2009). If NGOs leave, the short-term service provision benefits they brought can evaporate if there has not also been investment in local government officials (Doyle and Patel 2008).

Allies or adversaries? Conditions facilitating blurred boundaries

NGO–government relations provide the backdrop against which NGO activities affect these four elements of stateness within a country. Their interactions can range from deeply tense, conflictual relations to collaborative relationships between NGOs and public managers as allies.¹⁶ Collaborative relationships are generally associated with several conditions, which are neither universal nor unwavering: high-performing, established NGOs that are focused primarily on service provision rather than human rights, democracy, or governance; a relatively open political climate, in which freedom of association is normally assured; a central government that is committed to development and grants autonomy to service providers; stable, predictable laws and regulations that provide positive incentives to the NGO sector; clear mechanisms for NGO–government engagement along multiple lines; deliberate efforts by NGO and government to create a positive working environment; and an alignment of interests among NGOs and government departments. These collaboration-promoting characteristics and their conflict-inducing counterparts can be discussed in four broad categories: (1) organizational attributes of both governments and NGOs; (2) characteristics of NGO regulation; (3) basis for and contours of NGO–government interaction; and (4) level of interest alignment between the organizations.¹⁷ These categories are summarized in Table 2.2.

Table 2.2. Determinants of NGO–government relationship

Category	Description under collaboration	Description under conflict
Organizational Attributes	NGOs Focus on service provision Established over many years High performing Leaders connected to government Government-funded or diverse funding sources	NGOs Seek political change Seek strong policy role Low capacity Work with politically sensitive groups Funded by distrusted (often foreign) donors
	Government New leaders (honeymoon period) Effective, skilled staff and capable organizations Committed to social change Space for organizational autonomy Economically and politically liberal	Government High bureaucratic red tape Low-skilled workers Weak commitment to poverty reduction Highly centralized; command-and-control administration Economically statist and politically authoritarian
Characteristics of NGO Regulation	Stable, predictable laws and regulation Straightforward; compliance is feasible Policies provide incentives to NGOs Goal is to facilitate NGO effectiveness NGOs accept state right to regulate NGO role in creating rules governing them	Laws and regulation unclear, sometimes intentionally Goal is to suppress NGO activity through rigidity, excessiveness, or interference in NGO activities NGOs avoid compliance with regulations
Basis for and Contours of Interaction	Basis for Interaction Informal, trust-based relations that have developed over time Donors require consultations and collaboration	Basis for Interaction Competition for resources Very rapid growth in NGO sector combined with limited government capacity to regulate
	Strategies of Engagement Deliberate efforts to create positive working relationship Government recognizes NGO contribution	Strategies of Engagement Deliberate efforts to avoid government regulation or monitoring Government does not recognize NGO contribution
	Features of Interaction Multiple forms of interaction occur NGOs avoid politically sensitive issues Mechanisms for engagement clearly defined Both organizations play role in writing formal agreements Some flexibility exists in relationship	Features of Interaction Political will for cooperation lacking NGOs unwilling to engage in dialogue NGOs deliberately not transparent Asymmetrical power relations Rigid agreements, written by one side
Level of Interest Alignment	High level of alignment Mechanisms exist for accommodation when there is conflict or tension	Low alignment of interests Low levels of trust

“Collaboration” and “conflict” are two ends of a spectrum that contains a range of possible interactions (Coston 1998). These relationships can be multifaceted as well – in some cases, an NGO and a government actor might collaborate even when each views the other as a “necessary evil” – what Bryant (2001) calls “critical engagement.” Organizational pairings can also exist at multiple points on the spectrum simultaneously (Batley 2011, 317). For example, a Ministry of Health might have a conflictual relationship with a particular organization, yet the Ministry of Water in the same country might engage that NGO in a collaborative partnership. This can be true of interactions among different parts of the same organizations, at different levels, or even in different interactions of the same actors over time (Fisher 1998, Najam 2000, McLoughlin 2011, Ramanath and Ebrahim 2010). Ergo, certain conditions are more likely to facilitate collaborative or synergistic partnering relationships between NGOs and governments, while other conditions are associated with adversarial, conflictual, or – at best – neutral relationships.

Organizational attributes of both governments and NGOs

A number of organizational characteristics affect the nature of NGO–government relationships. On the NGO side, these pertain to organizational mission and strategy, age and size, leadership, source of funding, and program performance. On the government side, leadership and organizational performance are also important, as are their structure, development strategy, and political regime type.

NGO characteristics

Both the mission and the strategies of NGOs affect their interactions with government (Ramanath and Ebrahim 2010). Where an NGO's mission is oriented toward service provision, relations tend to be less conflictual than where NGOs aim to bring about social change – promoting human rights, anti-corruption measures, and civil and political liberties (Bratton 1989a), or participation, empowerment, or democracy (Clark 1992, Sen 1999, Banks and Hulme 2012). This principle is especially applicable to non-controversial services, like immunizations, clean water, and agricultural extension services (Bratton 1989a, Sen 1999). Governments see these activities as non-threatening, whereas governance activities can be a direct challenge to their rule. The latter is particularly true where NGOs organize protests, legally challenge the state via the courts, join with opposition leaders, or use media and informal communications to oppose the state (Clark 1992, Sen 1999, Banks and Hulme 2012, Nair 2011).

Likewise, where NGOs do not seek to engage in development planning as a deliberate strategy, but instead see their role as “gap filling” for the government, relations tend to be collaborative (Clark 1992). The same is also seen to be true, however, when the NGO sector is very small or weak, offering few services (Bratton 1989a). In such cases, NGOs are more likely to defer to government policy and service provision decisions.

When NGOs perform these gap-filling activities well, displaying high capacity and professionalism, governments tend to appreciate them and want to work together (Brautigam and Segarra 2007, Clark 1993). Conversely, miscalculating what is feasible (Nurul Alam 2011) or demonstrating poor performance and low skills (Clark 1993) can lead to a more acrimonious relationship. Likewise, if NGOs wish to avoid conflict, they must be careful not to generate loyalty from the people that comes at the expense of loyalty to government (Bratton 1989a, Heurlin 2010).

Conflict tends to be the norm where NGOs compete to make policy (McLoughlin 2011) or are perceived as engaging in anti-governmental activities (Tripp 2001). State–NGO relations are more likely to be strained where the organizations’ missions are to provide services in disputed territories, in sensitive security areas, in places where rebellion could occur, or in refugee camps and cross-border operations (Bratton 1989a, McLoughlin 2011, Sen 1999), since these are sensitive or unstable areas for the government. Even if unintentional, NGOs working with a very narrow constituency or marginalized group in a country can raise concerns for the government (Clark 1993) by threatening to disrupt power dynamics.

Often, these factors interact with other characteristics of NGOs. Where NGOs have been present in a country for a long period of time, relations tend to be more collaborative, as longevity means that a degree of trust has developed with government and community leaders (Rose 2011, Jelinek 2006). Long-standing, strong leaders within NGOs reinforce these trends, since the government will have gained familiarity with them over time (Rose 2011). Likewise, NGOs whose leaders previously worked in government, or who have moved repeatedly between government and NGOs, are likely to be able to work collaboratively with government offices, since they have personal connections to them (McLoughlin 2011). Conflict can arise, however, if strong, capable leaders use their skills to challenge the government (Heurlin 2010).

Finally, an NGO's source and type of funding can influence how it interacts with the government, including public agencies. Organizations that are funded through government contracts tend to have collaborative relationships with government (Rosenberg, Hartwig, and Merson 2008), especially where few other sources of funding exist – though this collaboration can easily turn into co-optation of the NGO. NGOs with diverse funding bases also tend to be able to develop partnerships with government – and on a more equal footing – since the funding diversity facilitates autonomy from public agencies (Rose 2011, Nair 2011). In such cases, organizations can take advantage of even minimal room to maneuver to create “constructive reciprocities” with public offices (Tripp 2001, 105). Where NGOs and the government compete for resources, or where they are perceived by the government to be funded by foreign states or organizations that are not allies of the government (Clark 1993), distrust and apprehension can develop.

Government attributes

Some of the same types of organizational characteristics that matter for NGOs are also significant on the government side of the equation, though they can be important in different ways. Whereas leaders that have managed nongovernmental organizations for long periods of time are able to develop trusting relationships with their government counterparts over the years, new leaders in government can also have a “honeymoon period” that facilitates collaboration (Brautigam and Segarra 2007, Brass 2012a, Atibil 2012). This honeymoon is most likely to occur if the new government leader is seen as a change-oriented individual, has previously worked in NGOs, or came to power due to a crisis where uncertainty opens a space for learning.

As with NGOs, collaborative relationships are more likely to develop where government agencies have effective structures (Clark 1992) and knowledgeable and capable staff (Jelinek 2006, Mayhew 2005), as well as the capacity to comply with donor partnership requirements (Brautigam and Segarra 2007). Such government organizations are more likely able to complement the skills of their NGO counterparts. When government staff employees have few skills and organizations lack capacity, relations tend to be tense (Jelinek 2006, Bratton 1989a, Clark 1993, McLoughlin 2011). In these situations, NGO workers often become frustrated by government ineffectiveness, and public employees can be threatened by their NGO counterparts – particularly when there is a high level of NGO performance. This tension can be exacerbated where state legitimacy is low (McLoughlin 2011), excessive bureaucratic red tape slows interactions between the organizations (Pugachev nd), or a dominant but ineffective central government agency tries to maintain control over the service sector (Rose 2006).

Just as NGOs’ mission and strategy affects interactions with government, a government's development strategy – not only at the national political level, but also the strategies of line ministries and local administrations – strongly influences the relationship. Public organizations that have a positive social agenda will find partnering with NGOs easier than those whose commitment to poverty reduction is weak (Clark 1993). Likewise, government organizations that are open to institutional pluralism in general are more apt to create a space for NGOs in policy creation or implementation (Clark 1993, Mayhew 2005, Coston 1998) than those that are resistant to such pluralism. Oftentimes this openness corresponds to states with a liberal economic development strategy (Heurlin 2010, Bratton 1989a), since such governments are more prone to contracting out to private organizations and to exerting less centralized control than are those with statist or socialist economic development strategies (Heurlin 2010, Neal 2008). Openness to pluralism also corresponds to states with democratic governance (Brautigam and Segarra 2007, Clark 1992), meaning not only a multi-party political system and respect for civil liberties (Bratton 1989a), but also a focus on accountability in the public administration (Clark 1993). That said, stability in both the economy and the political system create space for collaboration with NGOs as well (Nair 2011, McLoughlin 2011), and this can occur in nondemocratic spaces.

Government leaders, including public administration leaders, who are focused on asserting authority or restricting organizational autonomy will generally have more conflict-laden relationships with NGOs (Atibil 2012, Clark 1992, Mayhew 2005) than their hands-off colleagues. This is true not only of military dictatorships and one-party states (Clark 1992), but also personalist authoritarian regimes (Heurlin 2010). In the latter case, leaders tend to have a shallow base of elite support and must rely on patronage to maintain the compliance of those outside their base. NGOs can be used in such regimes to challenge the ruler (ibid.). Authoritarian rulers can avoid such challenges, however, if they co-opt NGOs (Heurlin 2010, Mayhew 2005). Political actors in patronage systems can sometimes even grant NGO services to loyal clients (Bratton 1989a).

Characteristics of NGO regulation

In addition to public organization attributes in general, the manner in which governments regulate NGOs strongly shapes how public organizations and nongovernmental organizations interact. Just as stability in politics and the economy facilitate collaborative relationships, so too do stability and predictability in the regulatory and legal environment (McLoughlin 2011). NGOs are more interested in working with public agencies where registering as an NGO is straightforward and easy to accomplish (Batley 2006), reporting requirements are reasonable, and taxation policies provide incentives for activities that conform with development priorities (Clark 1993). Finally, the motives for regulation matter – where regulation seeks to promote effective operation of NGOs (Batley 2006, Clark 1993) or incorporation of NGOs into state strategies (Heurlin 2010), partnerships are likely.

The response to regulation is also important. Collaboration is more likely where NGOs broadly accept state control through laws and frameworks of collaboration, but remain unsubordinated to the state (Nair 2011). Involving NGOs in creating regulation and NGO–government agreements facilitates positive responses (McLoughlin 2011, Coston 1998, Gugerty 2010), although NGOs engage in self-regulation in a number of countries (Gugerty 2008).

Generally, the inverse is also true. NGOs and government officials are more likely to come into conflict where laws, regulation, and “rules of engagement” are unclear or disrespected (Batley 2006) (including intentionally informal rules (Heurlin 2010)), which is more likely in highly centralized, command-and-control environments (McLoughlin 2011). The same is true where regulation is designed to suppress NGO activity (Gugerty 2008; Mayhew 2005), is excessive or rigid (Bratton 1989a), or directly interferes in NGOs’ activities (Mayhew 2005) – even when the regulation is not applied or implemented (Batley 2006). In such cases, governments and line ministries face NGO counterparts unwilling to comply with the laws and regulations (Palmer et al. 2006). This conflict can be exacerbated further when NGOs and government sectors hold different opinions about regulation and how it is used (Mayhew 2005).

Basis for and contours of NGO–government interaction

The reasons that NGOs and government interact, as well as the contours of those interactions, together form a third factor affecting the likelihood of collaboration or conflict. Here, several elements are important: the basis or impetus for NGO–government interactions; strategies both types of organizations use in engaging with the other; and features of the actual interactions, such as the degree of formality and power dynamics.

Basis for interactions

One of the key ingredients for a collaborative relationship between NGOs and government agencies is a slow buildup of interactions over time. These processes allow personal, informal, and trust-based relationships to develop between NGO workers or leaders and their government counterparts (Rose 2011; Batley 2011; Pick, Givaudan, and Reich 2008; Nair 2011; McLoughlin 2011). As several scholars have noted, even where relationships were conflictual at some point in time, repeated interactions can mellow tensions (Bratton 1989a) by creating familiarity and facilitating social learning, often through key bridging individuals (Brautigam and Segarra 2007). The role of donors is often important in such interactions, since many donors require governments to partner with NGOs in order to receive funds. Donors often facilitate interactions by insisting that consultation and collaboration occur (Brautigam and Segarra 2007; Pugachev nd; Nurul Alam 2011; Batley 2011), and by working to create an environment that encourages NGO entrepreneurialism (Bratton 1989a).

On the other hand, conflictual relationships are likely when the basis for interaction between NGOs and government agencies or officials is due to competition over resources, including donor resources or clients (Pugachev nd; Mayhew 2005). In particular, in weak states, governments often feel threatened by NGOs when donors direct resources to the NGOs (Brass 2012a). Similar tensions arise when the NGO sector in a country grows more rapidly than the government's capacity to regulate, monitor, or keep track of organizations in the sector (Bratton 1989a). Both of these bases for interaction are more likely in relatively closed political systems and those with a high degree of government corruption and low capacity. However, if organizations initially interacted during a period of political repression, where conflict is high, such tensions can remain, even if democracy later develops (McLoughlin 2011; Sen 1999).

Strategies of engagement

Whereas I discussed NGOs’ mission and strategy generally in the section on organizational attributes, here I highlight the strategy and tactics NGOs use to engage the government specifically. When NGO leaders or workers prioritize positive working relationships with government, and vice versa, collaborative relationships become more likely. Putting in the effort to recognize and engage the other organizations produces results. NGOs often deliberately invest considerable time in relationships with their government counterparts to avoid tension (Rose 2011; Pick, Givaudan, and Reich 2008), confrontation, or conflict (Batley 2011; Bratton 1989a; Clark 1992). To promote positive working relationships, they: draw attention to the congruence between their goals and those of the government (Bratton 1989a); deliberately share information about plans, programs, activities, and results with government (Jelinek 2006); work to coordinate activities with those of government (Jelinek 2006; Clark 1992); position their work strategically vis-à-vis government, including helping government departments improve their services and strengthen their systems (Clark 1992); and share resources with their government counterparts (Coston 1998). NGOs also often invite government officials to their trainings, workshops, project inaugurations, and other events (Jelinek 2006). This inclusiveness is often a deliberate strategy for keeping relationships cordial.

On the government side of the coin, collaboration is more likely when government agencies or officials formally recognize the contributions that NGOs make (Rose 2011). In some cases, public actors merely become resigned to the presence and popularity of NGOs (Cannon 1996), while others explicitly create spaces for their mutual cooperation (Clark 1993; Mayhew 2005; Coston 1998). In co-production relationships, governments deliberately share responsibility and operation with other actors, including NGOs (Coston 1998).

Under certain conditions, however, governments are unlikely to recognize NGOs’ role in service provision. If NGOs’ role in a service area becomes very large, or if NGOs and the government hold divergent understandings of the proper role of NGOs (Atibil 2012), the government may see the organizations as a threat and may respond by increasing its efforts to exert control (Rose 2011). Likewise, NGOs are unlikely to take a deliberately collaborative approach where the government is restrictive and controlling.

Features of interactions

The manner in which government and NGO actors actually engage when they interact is also an important component of their relationship – both informally, and in formal contracts and agreements. Informally, interactions are more likely to be associated with partnership when NGOs and government are committed to working together (Pugachev nd, Mayhew 2005), and multiple forms of engagement occur between the organizations (Rose 2006; McLoughlin 2011). For example, an NGO that provides health services in a community alongside government health workers and is also invited to take part in creating policy or regulation in the health sector is more likely to work in collaboration with government actors. Over time, NGOs often learn what the political obstacles in their environment are – and how to avoid them (Brautigam and Segarra 2007; Clark 1992).

Tensions or outright conflict are more likely to appear, however, when the political will for cooperation is lacking on one or both sides (Mayhew 2005). Clashes can occur when NGOs are unwilling to engage in dialogue with the government (Clark 1993), when they are deliberately opaque regarding their activities or funding (Cannon 1996), or, more generally, when there are clearly asymmetrical power relations and the government holds more power (Coston 1998). Even when broad regional or countrywide conditions appear to be favorable (McLoughlin 2011), local-level informal political conditions can create tense interactions.

The character of formal contracts or agreements between NGOs and their government counterparts also shape the collaborative or conflictual nature of their relationships. Some scholars suggest that NGO–government relations are more likely to be collaborative where the mechanisms for engagement are clearly defined (Mayhew 2005) with each actor's role clearly delineated (McLoughlin 2011), and where both sides play a role in writing formal agreements (Rose 2011). Others, however, suggest that formal, contractual agreements can allow governments to subordinate NGOs (Batley 2011), particularly when the contractual relationship is rigid (Nurul Alam 2011) or designed by only one side (McLoughlin 2011). The degree of contractual formality may not matter as much (Coston 1998) as having a degree of flexibility or room to maneuver in the relationship (Rose 2011). NGOs, in particular, are more likely to take a collaborative stance when they maintain an ability to negotiate their position with the government (Nair 2011). NGOs maintaining a degree of informality (Batley 2011), or relationships growing out of informal relationships between NGOs and government officials, can facilitate such collaborative relationships (Rose 2011).

Level of interest alignment between the organizations

Finally, interest alignment among NGOs and their government counterparts is crucial for the establishment and maintenance of strongly collaborative relationships (Puplampu and Tettey 2000; Brautigam and Segarra 2007; Rosenberg, Hartwig, and Merson 2008; Clark 1992; McLoughlin 2011; Sen 1999). Alternatively, if their interests do conflict, collaborative relationships are also possible when mechanisms exist to accommodate both sides (Coston 1998). Such interest alignment is more likely to occur where NGOs and public officials share common beliefs or values (McLoughlin 2011), and where NGOs bring resources to the relationship that supplement those of government (Bratton 1989a). According to several scholars, many NGOs and governments have learned over time that it is in their interests to collaborate (Brautigam and Segarra 2007; Clark 1992).

On the other hand, discord between organizations is more common when their interests do not align (Brautigam and Segarra 2007; Najam 1996). This is more likely to happen when government leaders take steps to make sure donor resources remain in government hands (Brautigam and Segarra 2007), when NGOs subscribe to a development theory that is different from that of government (Clark 1992), or when NGOs demand greater participation in setting the agenda than the state would like (Sen 1999).

At the most basic level, NGO–government interactions are shaped by the perceptions that NGO workers hold of the government as a whole, or government actors in particular, and vice versa. In particular, as in any human relationship, trust matters. Government and NGO actors have difficulty engaging in collaborative relationships when they distrust one another (Bebbington et al. 1993; Clark 1993), or perceive the other to distrust them (Jelinek 2006). This distrust can occur at both the individual and organizational levels. Like distrust, resentment brings tension (Jelinek 2006, Nurul Alam 2011), as when governments resent NGOs’ access to resources (Clark 1993).

Conflict is more likely to arise when distrust is combined with perceptions among government actors or agencies that NGOs are competing with them for resources (Gugerty 2008; Coston 1998), for clients (Pugachev nd), or for political support (Bratton 1989a). Likewise, when government actors think NGOs are tools of foreign donors trying to change state policy (Pugachev nd), the likelihood of struggles increase. These perceptions can lead government to feel threatened or challenged by NGOs, which hampers collaboration (Clark 1992; Mayhew 2005).

On the NGO side, the possibility for collaboration or partnership is strained when NGOs perceive government officials to be incapable of implementing projects and programs (Jelinek 2006), to deliberately obstruct or interfere in NGOs’ work (Nurul Alam 2011), or to have too much red tape (Pugachev nd). On both sides, such discordant perceptions often grow out of a history of mistrust and accusations of corruption (Batley 2011), or out of simple misinformation (Jelinek 2006). Where NGOs and government perceive each other as having mutual goals and a similar understanding of the means for achieving goals (Pugachev nd, Clark 1993; Coston 1998), collaboration becomes much more likely.

NGO–government isomorphism: moving toward collaboration through social learning

The characteristics described in this chapter provide an overview of the conditions likely to make allies or adversaries of NGOs and their government counterparts, yet it is crucial to remember that relations are not static, and that these conditions are not universal. As Batley (2011, 308) explains in a comparative article on Bangladesh, India, and Pakistan, “Governments’ relationships with voluntary associations or NGOs have evolved historically and are affected by the institutional characteristics of states and by the practices of governments.” Indeed, in most places around the world, there has been a thawing of interactions from generalized tension or outright hostility toward something more collaborative or at least neutral in tone. NGOs and states have come to more collaborative arrangements over time through a process of social learning and isomorphism (cf. Rosenberg, Hartwig, and Merson 2008; Brautigam and Segarra 2007; Keese and Argudo 2006; Rose 2006; Zafar Ullah et al. 2006; Taylor 2006; Brinkerhoff 2002).

What is social learning? It is the process by which actors acquire new knowledge through interaction with other people, organizations, and institutions in an environment. Akin to socialization, social learning does not occur via deliberate education. Instead, as people are exposed to new ideas and are persuaded to adopt them, they slowly adapt their understanding of what makes sense, is appropriate, or is legitimate (Brautigam and Segarra 2007). These changes can be seen in the NGO-and-government context as government offices adopt strategies, techniques, and processes previously associated with NGOs, rather than with public administrations.

I am not the first to apply social learning theory to studies of NGO–government relations. Here, I draw on Segarra (1997) and Brautigam and Segarra (2007), whose study of Ecuador, Gambia, and Guatemala shows how collaboration between NGOs and government, initially rejected by both sets of actors alike, became a taken-for-granted, legitimate, normative practice through strategic social learning processes that occurred during the 1990s and into the 2000s. Despite significant initial resistance, governments in all three countries learned that partnering with NGOs was beneficial to them. Like Brautigam and Segarra, I draw attention to the role played by donors in these processes in Kenya. Donors not only encourage collaboration between NGOs and the government, but also fund certain types of NGO activities, thereby reinforcing the norms they see and approve. At the same time, social learning can occur as NGOs exert influence on policy and practice by example and interaction, rather than through direct confrontation with the government (Batley 2011).

In many countries, social learning has initiated a process of institutional isomorphism on the part of government line ministries (DiMaggio and Powell 1983). According to DiMaggio and Powell's (1983, 149) highly cited theory, isomorphism is the process by which organizations change strategies and structure such that they resemble others in their field – a process of homogenization in organizational characteristics. These changes involve social learning, as “organizational decision makers learn appropriate responses and adjust their behavior accordingly” (ibid., 149). Three means of isomorphism exist: coercive, in which pressure from other organizations and society causes change; mimetic, which involves modeling organizational behavior and structure on the example of other, successful organizations in a field; and normative, in which organizations change through professionalization and formal education (ibid.).

Government agencies in many developing countries have changed through mimetic isomorphism, in which they learn from and adopt models of programmatic behavior and organizational dynamics from NGOs. Government workers model – sometimes explicitly and other times unintentionally – their programs on those of NGOs, which they perceive to be more legitimate and successful, confirming DiMaggio and Powell's theory (DiMaggio and Powell 1983, 152). Public administrators are also affected by coercive isomorphism, stemming largely from donor governments and multilateral development institutions, as well as from NGOs. Here, administrative agencies have adjusted their internal processes in response to pressure from donor organizations and NGOs, on which they are dependent for resources. Following DiMaggio and Powell (1983, 150), I show that even when the changes made are largely ceremonial, they are still consequential, able to change dynamics in the long run.

Through these processes of social learning and institutional isomorphism, bureaucratic organizations of government line ministries begin to resemble NGOs in several key ways that suggest improvements in their administrative capacity. First, line ministries have increasingly taken participatory approaches, integrating the views of beneficiary communities into planning and decision-making processes. These “demand-driven” approaches reflect a new way of thinking for many civil servants. Second, government offices have increasingly focused on performance management approaches – holding annual performance reviews for employees, setting targets for provision of services, and evaluating their work against these targets.

Conclusion

This chapter has provided conceptualizations of key terms in this book, theories of the interplay between NGO service provision and the state, and conditions under which NGOs and their government counterparts collaborate or conflict. Four “elements of stateness” – territoriality, governance, administrative capacity and legitimacy – were explained, as well as how they have changed as a result of the proliferation of NGOs providing services around the world. Theories of civil society, governance, the social contract, and social learning help elucidate these changes.

Under the conditions detailed in the chapter, NGOs can comprise an integral component of the organizational form of the state in developing countries like Kenya. When NGOs and governments share largely positive working relationships, and when NGOs are given room to maneuver, NGOs expand the nature of state service provision in such a way that we can include nongovernmental as well as governmental actors under the aegis of “the state.” As NGOs and government work hand in hand on programs and projects, the line between public agency and private NGOs blurs. NGOs help to reconstitute the state through the creation of networks of actors undertaking functions that had traditionally been associated with the state. Symbolically, NGOs suggest to people that organizations are looking out for them. With the addition of these private actors, government performance improves.