Over the past decade, state legitimacy—understood in its simplest form as citizens' acceptance of the state's right to rule—has gained increasing prominence in state-building debates. The view that illegitimacy undermines the state's capacity and stability is advanced in much of the influential state-building literature (Kaplan 2008; Lake 2007; Rotberg 2004). This view is also recently echoed in development policy, where legitimacy deficits are seen as a key driver of state fragility, and the antithesis of its corollary—state resilience (Department for International Development [DFID] 2010; Organisation for Economic Co-operation and Development [OECD] 2010). Yet the manifest inability of many conflict-affected states to (re)gain legitimacy has been described as “disappointing” to say the least (François and Sud 2006). In recognition of this disjuncture between the significance of legitimacy and its elusiveness, in 2009 the president of the World Bank called for legitimacy to henceforth become the “strategic centre of gravity” for all state-building interventions (Zoellick 2009).
Alongside the rise of legitimacy as a core component of state-building, there has been a concomitant concern (at least at the rhetorical level) with understanding the sources of legitimacy available to a state, and how external aid interventions might influence them (Bellina et al. 2009; OECD 2010). In this vein, there is a striking trend toward framing the provision of vital public services—including health, education, water, and sanitation—as a key source of state legitimacy. The OECD (2011, 74) argues that because basic services represent a material expression of “reciprocal state-society relations” (the synonym for the social contract), aid interventions in this area “play a major role in enhancing state legitimacy.” So widespread is the idea of a direct causal link between service delivery and state legitimacy, some conclude it has achieved the status of a received wisdom in aid policy (Carpenter, Slater, and Mallet 2012).
There are at least two reasons to question this received wisdom. The first is to understand what happens to aid for service delivery, traditionally designed primarily to meet critical human needs, when it takes on the added, political imperative of state-building. The second is that, taken at face value, the notion that service delivery can instrumentally enhance state legitimacy appears something of a leap of faith. While a long scholarly tradition has unravelled the consequences of legitimacy for the state, much less is known about how legitimacy is accrued (Carter 2011). To the degree that social scientists have grappled empirically with legitimacy—arguably the “hot potato” of politics—there is no consensus around its origins, other than that these are multiple, interconnected, and context specific (Gilley 2006, 2009). Although illustrative cases suggest a link between declining or inadequate service performance and crises of legitimacy,1 the reverse proposition—that improved performance enhances state legitimacy—is not nearly so established (Brinkerhoff, Wetterberg, and Dunn 2012; Gilley 2006).
This article aims to take stock of the theoretical and empirical foundations of the received wisdom that service delivery improves state legitimacy. It finds that, in practice, the relationship between a state's performance in delivering basic services, on the one hand, and its degree of legitimacy, on the other, is nonlinear. A number of factors interrupt any guaranteed or straightforward causal relationship between these variables. Specifically, to the extent that citizens evaluate the state's right to rule through service delivery, this evaluation is likely to be affected by shifting expectations of what the state should provide, subjective assessments of impartiality and distributive justice, the relational aspects of provision, how easy it is to attribute (credit or blame) performance to the state, and the technical and political characteristics of the service.
Albeit indicative, the findings of this article together support the view that for any given institution to generate legitimacy, it must ultimately be justifiable by reference to core social values, and resonate with beliefs about what is right for society (Beetham 1991). These findings suggest a case for opening up the dominant institutional model underpinning the contemporary aid debate, which reduces the role of services in (re)building state legitimacy to an instrumental one. They call for a more rounded, norms-based account of the significance of service delivery for state legitimacy: one that looks beyond the material to the ideational and relational significance of services for citizens' evaluations of the state. Central to this is a deeper reading of the social construction of legitimacy that engages with the moral and normative criteria by which citizens individually and collectively judge state institutions (Saward 1992).
The article is structured as follows. It begins by setting out the key propositions underpinning the received wisdom that service delivery improves state legitimacy, asking to what extent these are informed by scholarly work on processes of legitimation. The next section goes on to illustrate that, in practice, a number of factors have been shown to interrupt any linear causal relationship between service delivery and state legitimacy. Based on these findings, the conclusion advances the case for a more rounded account of whether and under what conditions service delivery might alter citizens' perceptions of the state's right to rule.
Building State Legitimacy via Service Delivery: Foundations of a Received Wisdom
In prominent aid literature, the binding of states and societies through the exchange of services in return for citizen compliance often lies at the heart of state-building models. Central to this is the idea that basic services—a function states are universally expected to perform—signals state responsiveness, that is, both the willingness and capacity of states to respond to their citizen's basic needs (Whaites 2008). One particular value of service delivery is that as a signal or measure of state performance, it is highly tangible, in terms of both its physical apparatus and its acute value in everyday life. Accordingly, the OECD (2011) describe public services as the visible link between what citizens give the state (taxation) and what they expect in return (some degree of well-being). In the hierarchy of political goods, services are meaningful for state–society relations because they give content to the social contract between ruler and ruled (Rotberg 2004, 2–3). Expressed differently, they are “the glue that binds state and society together” (Milliken and Krause 2002, 761).
In some aid models, not only does service delivery have direct effects on state legitimacy, but in turn this legitimacy affords the state greater capacity to rule (Norwegian Agency for Development Cooperation [NORAD] 2009). The DFID (2010), for example, portrays a scenario in which states that respond to public expectations, including for basic services, set in motion a “virtuous circle” of state-building. In the DFID model, responsive services lay the basis for a more inclusive political settlement, strengthened state–society relations, and, over the long term, can address the underlying causes of fragility or conflict (DFID 2010). In the OECD's (2008) version of the virtuous circle, states with the requisite capacity to provide services in line with expectations are rewarded with increased citizen compliance with its laws and rules—crucially, tax compliance—which over time boosts state capacity to deliver services more effectively and, in turn, generates more legitimacy. In this way, the cycle of capacity, legitimacy and citizen compliance becomes self-reinforcing.
Another recurring theme is that among the multiple sources of legitimacy, service delivery falls squarely into the category of “performance legitimacy”—that is, a type of legitimacy dependent on the state's outputs (OECD 2011, 39). Some distinguish performance legitimacy from other sources of legitimacy derived from process, or how the state acquires power and conducts policymaking (Teskey, Schnell, and Poole 2012, 10). Writing for the Norwegian development agency (NORAD), Bellina et al. (2009) similarly categorize service delivery as falling into the discreet category of what the state produces, as distinct from how it functions, or the kinds of beliefs and shared community supporting the state's authority. Hence, to use Weber's (1962) well-known classification, service delivery is primarily seen as capable of building legitimacy of the “legal-rational” variety, that is, legitimacy derived not from charisma or tradition, but from functioning institutions.
While optimistic about the potential legitimizing effects of service delivery, development agencies are careful to acknowledge sources of legitimacy are context specific. Much of the aid literature adopts what might be termed an empirical definition of state legitimacy, one that stresses a regime is legitimate when people believe that no other authority would be superior, while avoiding venturing into the territory of what types of values and norms should underpin this belief (OECD 2010, 15).2 Accordingly, rather than there being a universal threshold of service access or coverage that can secure a state's legitimacy, it is the alignment between citizens' home-grown expectations of what the state should deliver and the state's capacity to meet these expectations that matters for legitimizing effects (Bellina et al. 2009; OECD 2008, 2011).
Questioning the Received Wisdom
The idea that services are a visible manifestation of the social contract resonates through research that positions service bureaucracy as the primary site where citizens are likely to encounter, and therefore subjectively judge, the state. Stuart Corbridge's (2005) seminal account of “seeing the state” in India is a widely cited case in point. He argues local services provide an opportunity for sightings of the state, and it is through these that people's expectations and interpretations of their broader rights and obligations with regard to the state are formed (Corbridge 2005). Others likewise see the state as largely meaningful to ordinary citizens when it is visible in localized practices (Gupta 1995). Studies of the attitudes and priorities of conflict-affected people have concluded that the degree to which state's meet citizens' everyday needs is an important component of their subjective assessment of it (Robins 2012, 4). From this perspective, positive encounters with frontline service officials might feasibly be a source of legitimacy for the state, particularly in fragile and conflict-affected situations where the state was previously mistrusted, or outright feared (Brinkerhoff, Wetterberg, and Dunn 2012, 279).
Likewise, one side of the virtuous circle posited in aid models is well established, that is, the proposition that legitimacy enhances capacity. The positive effects of the accrual of legitimacy on the capacity of states to govern and to generate development has been empirically demonstrated (Englebert 2002). Legitimacy enhances capacity because it makes citizens more likely to defer to decisions and it rules out of a sense of obligation, rather than through the threat (or exercise) of punishment or reward (Tyler 2006). Simply put, it makes ruling more efficient. The symbiotic relationship between a belief in the rightfulness of the state and citizen compliance underpins the stability of all political systems, and enables effective governance (Beetham 1991).
Arguments supporting the other side of the virtuous circle—that capacity necessarily enhances legitimacy—are typically more qualified. It is widely acknowledged that the significance of service delivery performance for state legitimacy is relative. While all definitions of legitimacy entail citizens evaluating whether or not the state is a proper, just, or rightful authority (Tyler 2006, 375), there are differing accounts of the importance of performance in this assessment. For Lipset (1984), legitimacy derives from a combination of effectiveness, the organization of political power, and how societies have historically resolved divisive issues. In his view, effectiveness and legitimacy are engaged in a continual balancing act: Powerful groups (e.g., military, business) may reject even a highly effective state if its basic values and symbolism does not fit with their own. This qualifies the legitimizing effects of service delivery by indicating these effects may depend on whose views count.
The relative effects of services on legitimacy can also be qualified in a temporal sense. Processes of legitimation can be thought of as the accrual of “goodwill” or loyalty to the state, which varies at any given moment in time (Gibson 2004, 289). A reservoir of legitimacy arguably affords the state better prospects of riding out periods of poor performance, without soliciting the withdrawal of consent (Gibson 2004, 289). From this perspective, whether or not a state's performance on service delivery affects its legitimacy may depend on the degree of legitimacy, or goodwill, it possessed in the first instance. Some maintain a chronic or acute breakdown in effectiveness would even endanger the stability of an otherwise highly legitimate state (Lipset 1984, 91). At a minimum, this is likely to be historically contingent on broader processes of legitimation.
The significance of service delivery for state legitimacy is also politically contingent. Scholars have questioned whether the means and ends of the “virtuous circle” are necessarily always virtuous (Brinkerhoff, Wetterberg, and Dunn 2012). Although from one perspective the physical presence of services might well signal state “responsiveness,” from another, more political perspective, they might form part of its coercive, political quest for preeminence. Tilly's (1992) seminal “coercion and capital” thesis illustrates that in Europe, the state's incentives to deliver basic services derived not from an altruistic quest for citizen well-being, but from the need to pacify them, so as to be able extract revenue to fund war. In Migdal's (2001) state in society approach, goods and services are conceptualized as part of a package of rewards, sanctions and symbols that the state ultimately provides in order to achieve social control. These coercive and controlling elements in the formation of social contracts are absent from mainstream development policy narratives. Van de Walle and Scott (2011) add another dimension, documenting how states have historically pursued service delivery for the purposes of penetration (establishing presence and visibility), standardization (quashing alternative power sources), and accommodation (creating loyalty, resolving disputes). Taken together, these insights call for a more political reading of the relationship between service delivery and state legitimacy, starting with the premise that legitimation processes are political struggles for order in social relations (Beetham 1991).
A further critique is that just as the institutional approach has dominated the state-building literature in general, the debate about service delivery as a source of state legitimacy has been colored by an institutional perspective. Situating service delivery within a hierarchy of state functions, as an output of an effective state apparatus, aligns with what Lemay-Hébert (2009) identifies as the dominant institutional approach to state-building, that is, one that views legitimacy as flowing automatically from functioning institutions. There are two potential challenges to this reductionist standpoint. The first is that pigeon-holing service delivery in the discrete category of performance legitimacy neutralizes it, underestimating the degree to which services might also conceivably act as a conduit for the state's norms and values, or what Gupta (1995) calls the “main myths and symbols” of the state. This has been argued with respect to health systems, for example, that “intentionally or not” may communicate the core values of the state to users, including its commitment to equity, transparency and accountability (Kruk et al. 2010, 94). In this way, service delivery is more than a question of the state's tangible functions and outputs, but also, to adopt a sociological perspective, a formative component of what Holsti calls the “idea of the state” (1996, 83). After all, as Migdal (2001, 6) argues, “societies are not, and cannot be bound only through material and instrumental relations” but also require “relational glue” in the form of common rules and meanings.
The second, related challenge to the reductionist approach is that ideology might shape citizens' perceptions of the state's performance. As vom Hau (2011) argues, an overemphasis on legitimacy as flowing directly from tangible outputs has diverted our attention away from this possibility. Legitimacy theory provides a strong foundation for the value of ideology in shaping belief; if legitimacy is fundamentally a belief in the rightfulness of the state's institutions (Beetham 1991), then those beliefs might conceivably be generated by deliberate cues and signals from those very institutions themselves. Lipset (1984, 86) specifically proposes legitimacy can be engineered by the state, dependent on its capacity to “engender and maintain the belief that existing institutions are the most appropriate or proper ones for the society.” In this interpretation, legitimation is in the hands of the powerful (Beetham 1991, 9). Particularly in the era of the electronic public sphere, perceptions of performance may rely on effective public relations, as much as material rewards (Beetham 1991, 9). This theoretically opens up the possibility that the discourse used by political leaders to frame and justify service delivery might influence its significance for state legitimacy.
The Difficulties of Questioning the Received Wisdom
Empirical studies that could more rigorously interrogate the underpinnings of the received wisdom that services enhance state legitimacy remain scant, particularly in fragile and conflict-affected states (Brinkerhoff, Wetterberg, and Dunn 2012). In part, these limitations reflect the methodological challenges inherent in measuring legitimacy.
One of the central unresolved issues relates to whether legitimacy is best measured through people's perceptions of the state, or through their behavior toward it. If legitimacy is conceived essentially as a belief in the rightfulness of the state, then researchers necessarily have to control for the subjective origins of beliefs, and their heterogeneity. Particularly but not only in fragile and conflict-affected states, there may be a dearth of reliable public opinion data and, more fundamentally, even where it exists, people may fear reporting their actual perceptions of the state (Call and Cousens 2008, 15–16). According to Beetham (1991, 13), asking people whether they believe a particular institution is legitimate is misleading anyway: Not only are people unlikely to understand what “legitimacy” means, but the more demonstrable, behavioral markers of legitimacy—consent and compliance—are observable in the public sphere, not “in the recesses of people's minds.” For practical purposes, these observables may be a more measurable effect of legitimacy (Gilley 2006, 49), but they nevertheless do not have a direct relationship with it. As Schaar (1984) argues, states can achieve compliant behavior through coercion, and people can consent/not consent out of fear, including in situations where the state is illegitimate by any other measure. In other words, the entitlement or right to rule is not equivalent to deference to power (Schaar 1984). Another problem with measuring legitimacy is that it is difficult to differentiate between people's support for an incumbent government, or individual leaders, and the more fundamental question of whether they accept the state's institutions as right and proper (Guerrero 2011).
To date, research on the determinants of state legitimacy has broadly segmented between a group of scholars seeking to uncover its universal correlates and others who maintain that legitimacy is more intuitively understood by observing the texture of citizen–state relations at the microlevel. In other words, legitimacy has been studied from above and below. The methodological hallmark of the former approach has been large-scale, quantitative studies, which tend to support the view that no state relies on a single source of legitimacy. In these studies, it is sometimes difficult to isolate the effects of measures of service performance on state legitimacy, particularly where these are analyzed as one component of a larger set of other indicators of socioeconomic development. In his analysis of data from 72 countries, Gilley (2006), for example, concludes that combined indicators of welfare gains, good governance, and democratic rights, were “important correlates, and probably causes, of legitimacy” (Gilley 2006, 48).3 Other studies have found the effects of service provision on approval of and trust in the state (used as proxy variables4) to be indirect, mediated by other markers of well-being, including food security (Sacks 2009). While these studies indicate correlations between well-functioning services and state legitimacy, they invariably conclude more research is needed to examine the causal mechanisms that underlie these correlations. In essence, they cannot explain why services might enhance legitimacy, or what factors might influence any causal relationship between them.
Factors Affecting the Relationship between Service Delivery and State Legitimacy
Notwithstanding the above limitations, emerging research indicates that the relationship between a state's performance in service delivery, on the one hand, and its degree of legitimacy, on the other, is nonlinear. In practice, a number of factors interrupt any direct, causal relationship between them. Specifically, this relationship is likely to be conditioned by shifting expectations of what the state should provide, subjective assessments of impartiality and distributive justice, the relational aspects of provision, how easy it is to attribute (credit or blame) performance to the state, and the technical and political characteristics of the service. Albeit indicative, these findings support the proposition that legitimacy is essentially socially constructed (Coicaud 2002; Holsti 1996). They suggest the significance of service delivery for state legitimacy depends at least partly on the locally determined normative criteria by which services are individually and collectively judged.
Citizen's expectations of what the state should provide appear to interrupt any straightforward relationship between objective service outputs and legitimacy gains.5 Highlighting the subjectivity of these expectations, Sacks's (2011) quantitative study across Africa, Latin America, and Asia finds weak correlations between objective measures of delivery (e.g., the mere presence of facilities) and citizens' satisfaction with services. In this case, citizens' assessment of performance appeared to depend instead on perceptions of how well government was “trying” to improve them. Recent Afro-barometer public opinion survey data similarly indicate the mere presence of physical infrastructure is not significant in shaping popular views about government performance (Asunka 2013). Rather, the quality of the experience (waiting times, availability of materials such as drugs/textbooks) and the accessibility of the service (capacity to pay fees, payment of bribes) are key (Asunka 2013). Indeed, perceptible improvements in performance may be more significant than absolute or verifiable measures of performance for legitimacy. In his study in Medellin, Colombia, Guerrero (2011) finds that a quick upgrading of basic services (infrastructure, health, education) in the city's less favored districts improved political support for and trust in government. Rapid improvements generated greater legitimizing returns than slower, less perceptible progress. Collectively, these studies indicate that subjective interpretations of quality and effort (rather than objective measures of them) are significant for the relationship between service delivery and state legitimacy.
Aligning citizens' expectations with state capacity, seen as the catalyst for legitimizing effects in aid debates, is unlikely to be straightforward. Particularly in fragile and conflict-affected states, citizens' expectations may be low, or nonexistent. In the Democratic Republic of Congo, researchers identified what they termed a “legitimacy threshold,” characterized by very low expectations of the state, and citizens' aversion to its intrusion into their everyday lives (Stel et al. 2012). The study concluded that to improve state legitimacy, expectations had to first be “stimulated,” through an initial show of performance (Stel et al. 2012). In contrast, citizens might in some cases have high expectations of services, and even be willing to pay for improvements, where they cite very low levels of satisfaction with them. This finding is made by Brinkerhoff, Wetterberg, and Dunn (2012) in Iraq, where satisfaction with water services did not correspond with reported levels of trust in the state. These contrasting findings—between an absence of any services and an aversion to the state, and poor services that have no direct relationship with levels of trust in the state—illustrate context specificity. They also signal that a foundational level of services may first need to be established before subsequent improvements are likely to affect citizens' belief in the rightfulness of the state. The dilemma is that particularly in postconflict situations, stimulating expectations may have to be balanced with the well-known risk of raising expectations above what is feasible, with potential negative implications for legitimacy if those expectations are not met (Brinkerhoff, Wetterberg, and Dunn 2012).
Expectations have also been shown to shift over time. Recent multi-country research in Nepal, Rwanda, and South Sudan indicates that in some fragile and conflict-affected states where services are poor or nonexistent, expectations can quickly graduate from initial concerns over access, to include concerns over quality and cost (Ndaruhutse 2012). Furthermore, there may be a “tipping point” in the legitimating returns from provision, once expectations are met. This was recently found in Colombia where, as service provision expanded, citizens' level of satisfaction diminished, partly because once desired improvements in one location were achieved, attention quickly turned to another area where performance was lagging (Guerrero 2011). In this case, the relationship between improvements in service delivery and enhanced state legitimacy appeared U-shaped; an initial flurry of trust-generating effects faded over time as citizens came to take service delivery for granted. These findings nuance the idea that it is the alignment of expectations and capacity that produce legitimating effects. They remind us that even where services are provided to the full extent of state capacity, expectations may still not be met. Just as expectations are a something of a moving target for the state, the effects of meeting them on state legitimacy may likewise be temporary.
Perceptions of Impartiality and Distributive Justice
Others have argued that certain normative characteristics of performance color citizens' evaluations of the state, affecting the degree to which services are likely to improve state legitimacy. Prominent among these is the issue of perceptions of equity in the distribution of services (Ndaruhutse 2012). Qualitative research in Liberia, Nepal, and Colombia found that unequal or exclusionary access to public goods was detrimental to citizens' views of the state's rightfulness (Dix, Hussmann, and Walton 2012). Particularly in postconflict contexts where horizontal inequalities prevail, a redistribution of services may therefore be important for (re)legitimizing the state among excluded groups, including those alienated through a period of conflict. However, this is likely to be a careful, political, balancing act. Citing the case of Iraq, Brinkerhoff, Wetterberg, and Dunn (2012) note that the redistribution of services to previously excluded groups in the postwar period diminished the state's overall legitimacy gains. This recalls the fundamental importance of understanding whose views count in the assessment of the state's performance. However, since most studies do not disaggregate citizens' perceptions of the state's performance along ethnic, area, or class lines, this is a widely neglected question. In a rare exception, Carter's (2011) research in South Africa observed how people's perceptions of state legitimacy differed according to race, age, and gender, between urban and rural populations, and by their “lived experiences” of apartheid. Through this albeit isolated study, attention is drawn to understanding the historic origins of the expectations of different groups. In practice, citizens' perceptions of the state can rarely be aggregated.
In some cases, the degree to which service delivery is perceived as procedurally fair has been shown to be significant in determining citizen's view of the state. In his widely cited research, Rothstein (2009) makes the case that because basic service delivery devolves significant discretion to low level public officials, the impartiality with which services are implemented on the ground is of acute significance for political legitimacy. His work empirically demonstrates that in developed states, greater impartiality in the exercise of state power—including through service delivery—is positively associated with higher levels of trust in government (Rothstein 2009). Other studies concur that whether or not services are operated through open and transparent practices may be seen as a tangible marker of the state's broader commitment to accountability (Ndaruhutse 2012). At a minimum, in order to solicit citizens' trust and approval, officials (e.g., teachers, nurses) may need to treat the individuals with whom they interact in a dignified manner (Sacks 2011). This is in line with one of the central tenets of legitimacy theory that the perceived fairness of the process through which authorities and institutions make decisions and exercise authority is a key aspect of people's willingness to comply with it (Tyler 2006). On the other hand, Blundo and de Sardan's (2006, 101) attention to the “real functioning” of public services illustrates that corrupt practices are not necessarily a source of illegitimacy; indeed, delivery that appears dysfunctional from a legal perspective may nevertheless be regulated and legitimated by local social and cultural logics. In this way, the relationship between the procedural fairness of services and legitimizing beliefs is best read in the context of local rather than universal norms.
Relational Aspects of Delivery
How services are organized and managed at the point of delivery could also condition the effects of provision on state legitimacy. Cross-country case study research looking specifically at multistakeholder processes6 concluded it was mainly the relationships formed through them that were significant for citizens' perceptions of the state (Stel et al. 2012). In particular, these processes created space for civil society organizations to articulate citizens' demands, and to directly engage with government agencies. They presented opportunities to build bridges between the state and social groups (Stel et al. 2012). In other isolated cases, coproduction7 has played a formative role in generating positive evaluations of the state on the part of citizens. Research from China reveals that state officials themselves understand the potential legitimizing returns from visible processes of coproduction. Tsai (2011) finds that local state bureaucrats viewed collaboration with nongovernmental organizations (NGOs) and local communities to implement local infrastructure services as a means of soliciting citizens' trust in them. Moreover, some officials believed collaboration would enhance their capacity to elicit greater overall levels of citizen compliance with state policies. This microlevel insight into the social contract being “articulated” through service delivery highlights services can bring citizens and states together not in a simple supply-and-demand transaction, but also in a coproductive one. It recalls the argument that in order to support their legitimacy, states require not just productive but also relational capacity (Robinson 2008; Stel et al. 2012). Early indications are that in the sphere of service delivery, the relational aspects of provision matter for citizens' evaluations of the state, and therefore also for any potential legitimizing effects.
Who is perceived to be delivering a particular service might also influence whether or not the state accrues legitimacy from services, though perhaps not in a straightforward way. The dominant position in aid policy has been that parallel (nonstate) service delivery structures undermine state legitimacy because they reduce the state's visibility as provider (Bellina et al. 2009). There is some supporting evidence of this from Zambia, where citizens who (rightly or wrongly) credited nonstate providers with service delivery were found to be significantly less likely to have confidence in their government, or to comply with taxes and regulations (Sacks 2009). The counter view, however, is that whether or not nonstate service delivery affects people's perceptions of the state logically depends on whether people expected the state to deliver services directly in the first place (Batley and Mcloughlin 2010). Stel (2011) makes this point in reference to perceptions of service delivery in Burundi, noting that because people did not expect the state to be involved at the point of delivery, they were not disappointed when it was not. Indeed, in fragile and conflict-affected situations where the state may have been repressive, citizens may not trust the state to deliver services, let alone expect it to (Brinkerhoff, Wetterberg, and Dunn 2012). These findings suggest there is no universally applicable rule that nonstate provision undermines state legitimacy. They indicate attention might more usefully be turned to whether forms of provision are seen as normatively appropriate for populations, as has been found recently in the case of international nonstate provision of basic services in areas of limited statehood (Krasner and Risse 2014).
Processes of attribution—specifically, how easy it is for people to credit or blame the state for the delivery of services—may also interrupt any linear causal relationship between service delivery and state legitimacy. Research has shown that to whom people attribute service performance does not always reflect who is actually delivering them in practice. Sacks (2009) finds that citizens were incorrectly attributing government services to nonstate actors where multiple other actors (NGOs, churches, and donors) were also providing services in the local area. Like others, she argues attribution is primarily a product of visibility. In this particular instance, nonstate actors were more effective at branding themselves than government (Sacks 2009). Mindful of this potential for misattribution, policymakers have called for measures to increase the visibility of the state's role in service provision at the point of delivery, including by reducing donor and NGO branding (Teskey, Schnell, and Poole 2012). At the same time, some research indicates that even where the state is not the frontline provider of services, citizens may not be blind to its less palpable, indirect roles in provision, which can include oversight, regulation, and facilitation. Some focus groups have demonstrated that citizens can evaluate this indirect role of the state beyond its physical presence at the point of delivery (Stel et al. 2012). In Ethiopia, even where citizens attributed service delivery to nonstate actors, they also understood that government agencies were likely to be involved in the design process and, moreover, they could differentiate between the quality of delivery and the degree to which the state was fulfilling its responsibility in a broader sense (Mandefro, Noor, and Stel 2012). Overall, these findings collectively draw attention to understanding how citizens are informed about the particular role of the state in provision, whether it is the direct deliverer or not, as significant for the potential legitimizing effects of services.
How services are attributed to different levels of the state is also likely to be significant. Isolated studies indicate that improved service outcomes at the local level can enhance perceptions of the legitimacy of those agencies directly responsible for them (Bukenya 2013). However, there is a paucity of studies that can illustrate how this “scales up,” to affect state legitimacy overall. Acknowledging this in relation to his research on HIV/AIDS health clinics in Uganda, Bukenya (2013) concludes that improved perceptions of local health clinics did not automatically translate into improved perceptions of the central state. Some go further and argue that highly localized, personalized relationships of exchange of services and favors between individuals and local officials may actually undermine wider legitimacy (Stel et al. 2012). Particularly in patron–client contexts, reciprocity and exchange may be important aspects of the legitimation of power only at the microlevel (Chabal and Daloz 1999). At a minimum, it may be important to consider which level of government (if any) people view as responsible for different services, and therefore where they are likely to assign credit (or blame) for more or less effective performance (Brinkerhoff, Wetterberg, and Dunn 2012). Given that the state is not a “coherent, integrated, and goal-oriented body” (Migdal 2001, 12), tracing these intervening links between local delivery, and aggregate perceptions of the broader institutions of the state is challenging. As illustrated here, it is conceivable that perceptions of local delivery and overall assessments of the state's legitimacy may or may not be connected. This further qualifies any straightforward causal relationship between local services and state legitimacy.
The different technical and political characteristics of services may also affect their potential as a source of state legitimacy. In particular, the characteristics of visibility and information asymmetry may affect how easy it is for citizens to assign credit or blame for performance (Mcloughlin and Batley 2012). It is logical that performance has to be seen, experienced, and appreciated in order for it to translate into legitimating beliefs, but different services offer different scope for this. For example, Guerrero (2011) finds that services have a greater effect on trust and legitimating beliefs when they are both critical (salient) and highly visible to the public. Moreover, his study suggests those that are easily measureable (“as easy as turning the lights or the tap on”) and are experienced homogenously by a community in a geographically contained area, are more capable of influencing levels of trust in the state than those that are highly heterogeneous, or experienced differently by different citizens (e.g., individual encounters with doctors). Likewise, to the degree that citizens' capacity to evaluate the quality of public services depends on the degree of information they have about them (Jilke 2013), then levels of information asymmetry may condition the legitimizing effects of performance (Mcloughlin and Batley 2012). As Sacks (2011, 5) notes, available information about service quality determine whether citizens can make objective assessments of performance, or whether instead they rely on subjective accounts.8 On this basis, the relationship between service delivery and state legitimacy is unlikely to be uniform across service sectors.
There are also normative reasons why particular services may be more or less significant for state legitimacy in any given context. Kruk et al. (2010, 91) afford health special status because it is universally seen as a “super-ordinate value,” prized irrespective of ideology or political affiliation. In her statistical analysis of cross-country public opinion data, Sacks (2011) also finds that basic health services potentially have a greater overall effect on the approval of the incumbent government, compared to water and sanitation services, because of the acute significance of health for people's daily lives. Education is sometimes singled out as the service most capable of generating “social trust,” particularly where it can convey values of equality, institutionalize transparency measures, and bring young people from different ethnic, religious, and social groups together (Rothstein 2011, 163). Likewise, where a service caters to the main constituency of support of the ruling party, it may come to be viewed as a particularly salient source of state legitimacy (e.g., food subsidies in Malawi) (Mcloughlin and Batley 2012). While not suggesting there is a universal hierarchy of services, these findings highlight that understanding the potential for service delivery to enhance state legitimacy depends on the normative value of different services in specific contexts, which is unlikely to be uniform across countries, or even over time within them.
Conclusion: Toward a Rounded Account of the Relationship between Service Delivery and State Legitimacy
Development policy literature on the role of public services in processes of state legitimation has been colored by an understanding of the state as a set of institutions that can be built through certain functions, and is yet to convincingly embrace the “idea of the state.” This narrow interpretation of the instrumental role of service delivery in processes of state legitimation reflects the difficulties donors face in “getting beyond capacity.”9 In essence, development agencies are locked in a tension between understanding legitimacy empirically as a belief based on people's perceptions and values, and pursuing it through the “one-size-fits-all” modus operandi of state-building: capacity development.
The paucity of empirical work in this area raises an immediate question over the received wisdom that service delivery necessarily enhances state legitimacy. Quantitative research, with its sophisticated analysis of correlations between services and indicators of legitimacy, cannot account for the causal processes through which services might influence citizens' perceptions of the state. In qualitative inquiry, there have been few historical case studies to trace how citizens adjust their perceptions of the state, and indeed their behavior toward it, in response to relative improvements in service delivery over time.
In spite of these limitations, it is clear that in practice a number of factors interrupt any direct causal relationship between a state's performance in delivering basic services, on the one hand, and its degree of legitimacy, on the other. Specifically, this relationship is likely to be conditioned by shifting expectations of what the state should provide, subjective assessments of impartiality and distributive justice, the relational aspects of provision, how easy it is to attribute (credit or blame) performance to the state, and the technical and normative characteristics of particular services.
The findings of this article illustrate there is no straightforward alignment between objective service outputs and legitimacy gains. Even where expectations are low, they may quickly shift over time, and are likely to vary between different social groups, creating something of a moving target for the state. Furthermore, beyond its instrumental value, performance is normative: Services can act as a conduit for the norms and values of the state. In some cases, perceptions of impartiality and distributive justice in allocations of services between groups have been significant for citizens' evaluations of its rightfulness. This implies that service delivery might be analyzed outside of the restrictive, instrumental bracket of performance legitimacy, as also a source of process and normative legitimacy.
The narrow focus on whether nonstate provision undermines state legitimacy has arguably masked the more pressing question of whom citizens expect to deliver services, and whether forms of provision are considered normatively appropriate in particular contexts. Related to this, understanding attribution—or how citizens come to credit/blame or reward/sanction the state for relative improvements or deteriorations in service levels—is an important through not well-understood link in the hypothesized causal chain between services and state legitimacy. To this end, there is a need for a more joined up analysis of the localized effects of services on trust in local bureaucracies and citizens' beliefs in the broader state's right to rule. Attribution is neither guaranteed nor always technically correct, depending on the visibility of and information available about the state's role in provision. Likewise, the relationship between service delivery and state legitimacy is unlikely to be uniform across services, because they offer different possibilities for attributing responsibility for performance to the state, and their significance to societies is likely to vary according to local norms, values, and priorities at particular points in time.
Taken together, the above findings call for a more rounded account of the relationship between service delivery and state legitimacy, one that starts from a recognition that this relationship is not automatic. They suggest research and policy might look beyond the material to the ideational and relational significance of services for citizens' evaluations of the state. Central to this is a deeper reading of the social construction of legitimacy that engages with the moral and normative criteria by which citizens' individually and collectively judge service delivery institutions. Future empirical inquiry could usefully focus on the texture of citizens' everyday encounters with service bureaucracy and how these affect citizens' perceptions of the state from the “bottom up.” More longitudinal and historical research designs could better capture how expectations and perceptions of levels and quality of provision alter over time. These strategies might advance a more nuanced and deeper understanding of the relationship between service delivery and state legitimacy in particular cultural and socioeconomic contexts. The significance of this agenda is to understand under what circumstances service delivery fulfills the dual imperative of meeting basic needs and state-building, and whether aid interventions can ever convincingly claim to do both.