Domestic reformers and external donors have invested enormous energy and resources into improving governance in developing countries since the 1990s. Yet there is still remarkably little understanding of how governance progress actually occurs in these contexts. Reform strategies that work well in some places often prove disappointing elsewhere. A close examination of governance successes in the developing world indicates that effective advocacy must move beyond a search for single-focus “magic bullet” solutions toward an integrated approach that recognizes multiple interrelated drivers of governance change.

Key Themes

  • Reform prospects rely on the interactions among three governance pillars: political commitment, bureaucratic capacity, and state-society relations.  Multiple factors shape each of these pillars and they can emerge in very different forms. It is nevertheless possible to identify common trends and emerging lessons.
  • Initial donor efforts to transplant Western institutions and best practices into developing countries largely failed. Governance advocates have subsequently turned to other solutions, such as finding individual political champions, encouraging citizen demand for good governance, establishing technocratic enclaves within bureaucracies, and devising flexible context-specific reform strategies.
  • The success of each of these approaches relies on supporting conditions. Political champions, for example, usually fall short without effective allies in government and civil society. Enclaves of bureaucratic excellence similarly require political support to maintain their independence.


Diane de Gramont
Diane de Gramont was a nonresident research analyst in Carnegie’s Democracy and Rule of Law Program.
More >

Political commitment, bureaucratic capacity, and state-society relations are deeply interdependent. This can lead to apparent vicious cycles of low commitment, low capacity, and unproductive state-society relations, but substantial progress can also be made on all three elements relatively simultaneously.

Individual political champions or external incentive structures alone are not sufficient to produce political commitment. For change to occur, key actors must recognize a serious problem or threat and decide that governance reform is an advantageous and feasible response. This requires the right incentives and a significant role for ideas and leadership abilities.

Bureaucratic effectiveness can be improved even where patronage politics are widespread. Flexible approaches to reform implementation, the creation of technocratic enclaves, and managerial efforts to raise civil service morale and accountability, among other strategies, can yield positive results.

Productive state-society relations can be exclusive or inclusive. Public participation and civil society advocacy are associated with governance improvements in some countries. In other contexts, however, exclusive public-private relationships have helped advance developmental policies. Both models rely to differing degrees on political commitment and a capable state. 


Over the past two decades, governance has moved to center stage in international development discussions. Scholars attribute development “miracles,” such as the cases of South Korea and Singapore, to particularly effective governments and often blame persistent development failures on incompetent or predatory rule. Domestic as well as transnational civil society groups focused on governance issues have multiplied throughout the developing world. International agencies have spent billions of dollars promoting better governance in poor countries in the belief that effective and responsive states are critical to socioeconomic development. These efforts cover a vast array of issues, from macroeconomic management to the quality of public service delivery and new spaces for citizen participation in development decisionmaking.

Despite all of this attention and investment, there is still remarkably little understanding of how governance progress actually occurs in developing countries. Reform advocates largely know what good governance looks like, but much less about how to get there. The best-known cases of effective governance draw on European and Asian experiences and often point to the importance of particular historical trajectories and structural factors not present in most low-income countries. Political economy studies of developing countries tend to have the opposite limitation, describing the myriad ways governments fall short of developmental ideals but not how they can overcome these obstacles.

Without clear reform examples to emulate, governance advocates have had to improvise. When development agencies decided to adopt governance as a priority, they initially sought to transplant the institutions and best practices of Western democracies into developing country contexts. These efforts largely failed to take root, forcing aid providers and domestic reformers to search for new approaches to address governance deficiencies.

Reformers have pursued a variety of strategies, such as finding political champions who can push through policy change, encouraging bottom-up pressure for better governance, and establishing special technocratic enclaves within problematic bureaucracies. In addition, a growing chorus of scholars and practitioners advises moving away from preset prescriptions and instead focusing on incremental, context-specific “good enough governance” solutions to specific development problems that “work with the grain” of domestic structures.1

Each of these approaches highlights promising reform options and can point to some striking success stories. Yet not one strategy is the proverbial magic bullet that can deliver governance transformation across all developing-country contexts. All of them rely on other supporting conditions. Reform champions usually fall short without allies in government and civil society as well as a competent bureaucracy to implement new policies. Public demands have limited impact when the state cannot respond. Enclaves of bureaucratic excellence require robust political support to maintain their independence. Incremental advances can unravel in the face of political instability.

Encouraging lasting reform means moving beyond single-focus approaches toward integrated, strategic thinking built on a strong understanding of the interactions among three key governance pillars: political commitment, bureaucratic capacity, and state-society relations. Substantial research and experience has begun to accumulate in each of these areas, including a growing number of real-world accounts of governance advances in developing countries. Existing case studies are limited in number and geography, with countries such as India, Brazil, and Colombia attracting substantial attention and others much less. Diverse studies also tend to focus on separate aspects of governance and highlight different reform drivers. As a result, the insights that can be gleaned from this learning process remain relatively scattered and sometimes point in contradictory directions. They nonetheless provide a crucial starting point to understand the dynamics of reform. 

This paper brings together these varied streams of research and practice to identify and critically examine the central debates and emerging lessons of governance change related to political commitment, bureaucratic capacity, and state-society relations. Political will cannot be reduced to either personal agency or external incentives alone; it is instead shaped by a combination of leadership threat perceptions, political security, and elite ideas. Civil service weakness is a stubborn challenge for governance reformers but several strategies have shown promise in improving bureaucratic effectiveness. These include flexible approaches to reform implementation, the creation of technocratic enclaves, and managerial efforts to raise civil service morale and accountability. Productive state-society relations can involve both inclusive citizen participation and exclusive relationships between the state and private sector. These three governance pillars are not independent and are deeply affected by the extensive interactions among key drivers of reform.

Inside the Black Box of Political Commitment

Whether a governance reform effort succeeds or fails, subsequent accounts will almost always explain its fate at least partly in terms of the political will of relevant power holders. This makes clear sense. On the one hand, active resistance by powerful interests or simple political apathy frequently prevents policy changes from being approved or fully implemented. Committed leaders, on the other hand, can sometimes initiate and carry through far-reaching governance improvements even in the face of considerable opposition. But why do some leaders and elites stake their political capital on improving governance effectiveness while others appear indifferent to bureaucratic decay and developmental failures?

Aid practitioners have responded to this question in two major ways. The first focuses on individual political champions who have the beliefs and skills necessary to lead reform efforts. The second discounts personalities in favor of incentives, considering the winners and losers of reform and assuming that power holders will act rationally based on their self-interest. These two approaches represent opposite perspectives on individual motivation. They are most useful for analyzing the likely drivers of political will when considered together.

Governance reform studies indicate that robust political commitment relies on key actors recognizing a serious problem or threat and deciding that governance reform is an advantageous and feasible response. These steps often involve conflicting pressures. The more severe a problem, the more difficult it can be to reject short-term crisis management in favor of transformative institution-building. To navigate these tensions and come up with positive governance responses, power holders must have not only the right incentives but also the right mix of ideas and leadership abilities.

Political Champions and External Incentives

Faced with political leaders who appear unable or unwilling to deliver effective governance and public services, a natural first reaction among both ordinary citizens and foreign donors is to look for better leaders. The result is enthusiasm for seemingly exceptional individuals who, because of their education, rhetoric, or other personal characteristics, generate high hopes that they can personally deliver governance improvements. Such personalities range from populist leaders who rally electorates with promises of rapid socioeconomic progress to foreign-educated technocrats who speak the reform language of Western donors and move easily in international and professional circles.

The “reform champion” narrative was clear, for example, in the praise and attention former U.S. president Bill Clinton’s administration offered to what it saw as a new generation of progressive African politicians in the 1990s, including the heads of Eritrea, Ethiopia, Rwanda, and Uganda.2 More recently, leaders such as Liberian President Ellen Johnson Sirleaf and Nigerian Finance Minister Ngozi Okonjo-Iweala have garnered similar reputations as indispensable African reformers.

The search for political champions is not restricted to prominent national figures; it also occurs with low-level officials and civil society leaders responsible for specific policy areas. A program aimed at improving teacher performance, for instance, is likely to rely heavily on apparent reformists within a state’s ministry of education and the teacher’s union.

This focus on leadership has some scholarly support. Marcus Andre Melo, Njuguna Ng’ethe, and James Manor describe the critical roles played by Ugandan President Yoweri Museveni, former Brazilian president Fernando Henrique Cardoso, and former chief minister of the Indian state Madhya Pradesh Digvijay Singh in skillfully pushing through reforms that benefited the poor.3 In a comparative study of Mexican municipalities that gained new authority due to decentralization, Merilee Grindle concludes that the quality of leadership exhibited by mayors was the most important factor in explaining differences in local government performance.4

Yet the central appeal of a reform-champion approach—the idea that particular individuals have the will and ability to overcome political constraints to development progress—is also a major limitation. Numerous seemingly promising leaders retreat from reform promises when changes begin to threaten their hold on power. Others deftly project themselves as reformists in some areas while resisting change in others. Before the outbreak of civil war in Syria, many international observers had hoped that President Bashar al-Assad and his Western-educated wife, Asma al-Assad, would be leaders of a new guard who could reform the country.5 Those hopes quickly faded in early 2011 with his government’s brutal crackdown on dissent. In Uganda, Museveni became a favorite of foreign donors due to his support for economic reform and action on HIV/AIDS even while he was consolidating his authoritarian rule.

The emphasis on uniquely capable leadership reinforces some of these tendencies. Politicians can become increasingly convinced that their continued rule is necessary for their countries’ development and refuse to yield control.6

Moreover, even steadfast reformists usually lack the power to push through policy change by themselves. Catherine Weaver criticizes the tendency of external actors to market policy ideas to perceived political champions who are then unable to overcome powerful interests with a stake in the status quo.7 Matt Andrews similarly highlights that even high-ranking reform champions have limited de facto authority. He emphasizes that policy change is a collective endeavor and requires multiple leaders with complementary roles.8

In response to the shortcomings of a champion-based approach, development analysts have increasingly employed political economy studies to take a more pragmatic view of the power and interests of key actors. As a 2014 World Bank publication stresses, the purpose of political economy analysis is not “to identify who are the ‘good guys’ and the ‘bad guys’ in a specific situation. Rather, it is to understand the underlying drivers that shape the incentives of decision makers.”9 The basic intuition behind this approach is that individuals pursue their self-interest in response to external incentives.

This can explain the seemingly contradictory behavior of policymakers who appear to be champions of reform in one moment and obstacles to developmental change in the next. For example, Chief Minister Nara Chandrababu Naidu in the Indian state of Andhra Pradesh successfully projected an image of good governance to external donors through reforms in areas like urban water management and e-governance. At the same time, he worked to solidify his patronage-based political control over the countryside.10 This behavior may seem like a betrayal of reform principles, but it follows the clear logic of a politician using multiple strategies to maintain power.11

Analyzing the incentives of key actors adds realism to the study of political commitment, but this approach also risks neglecting other important factors. While people may act rationally, they are not robots. Individuals face multiple and often competing interests and can respond to the same environment in different ways. For this reason, David Hudson and Adrian Leftwich criticize donors’ intense attention to incentives and self-interest in political economy studies, stressing that “agents have to interpret the opportunities and risks facing them, often in a situation of uncertainty. As such, agents’ beliefs, values, and other cognitive filters are central to understanding how they act.”12 Dani Rodrik similarly cautions against privileging interests over ideas in political economy, contending that leaders can be motivated by reputation and glory as much as economic gain. He notes that some politicians are better than others at identifying creative institutional solutions that serve multiple interests.13

Focusing exclusively on either exceptional leaders or objective incentive structures obscures a full understanding of the likely drivers of political will. But each approach can add nuance to the other. The rational actor perspective demonstrates the importance of paying attention to external incentives rather than considering reform commitment an intrinsic characteristic of certain individuals. Even the most enlightened politicians are unlikely to embark on reform efforts unless they believe change is possible given the political forces and funding arrayed for and against it. They must also consider whether the effort to enact a particular reform will cost them their power—and therefore their ability to enact future policies. The political-champion lens highlights how leadership beliefs and talents influence the way key power holders define their interests and legacy, as well as which solutions they consider viable.

Common Governance Threats

Before committing resources and energy to advance reform agendas, key power holders must decide that governance change is a priority. This requires more than just an awareness that governance problems exist. Multiple issues constantly jockey for the attention of leaders and they cannot all be satisfactorily addressed. Moreover, governance changes are almost always politically risky as current elites tend to have a vested interest in the status quo. Even a seemingly commonsense reform to improve public services for the majority of citizens can create as many political headaches as it solves. Scholarship on past reform processes indicates that sustained political commitment tends to arise in response to issues that are not only serious but also pose a direct security, financial, or political threat to power holders.14

Historically, security threats have been powerful drivers of one of the most difficult and far-reaching types of governance reform—basic state building. Charles Tilly first developed this theory with relation to early modern Europe, arguing that the constant danger of war forced rulers to undertake risky tax collection and other state-building efforts in order to provide for collective defense.15 Jeffrey Herbst draws on this logic to explain state weakness in Africa, contending that a historical lack of competition over land and modern protections for international borders have diminished the incentive for African leaders to build effective states to control and defend their territory.16  

Security threats can be particularly strong reform drivers because they tend to affect entire groups rather than individual leaders, potentially uniting elites behind more effective governance. This has been a recurring theme in studies of political and economic reforms in Asia.17 Ziya Öniş notes the danger that countries such as Taiwan and South Korea faced from communist rivals, contending that “the extraordinary security threat faced by the East Asian states helped to bolster the nationalistic vision inherent in these states and the unique commitment to the long-term transformation of the economy.”18 David Booth draws on the Asian literature to make a similar argument regarding Rwanda and Ethiopia, contending that in both countries, the experience of large-scale internal violence galvanized elites into overcoming collective action problems and uniting behind pro-developmental leaders.19

These examples all involve authoritarian regimes, but security crises can also spark reform commitment in more pluralistic settings. For example, the city of Medellín in Colombia and Bihar state in India were well-known in the 1990s for their high crime levels, political corruption, and economic problems—appearing to be the worst governed parts of countries already struggling with serious political challenges. Yet both places experienced sweeping reforms in the mid-2000s that helped deliver security, public services, and more transparent governance. Studies of each transformation emphasize how an acute sense of crisis and loss of competitiveness—even relative to serious governance problems in the past—combined with electoral opportunity to spark unusual coalition building among economic elites and popular sectors of society. This enabled opposition leaders to win a strong mandate for reform.20

Fiscal threats also impact governance reform commitment. Political leaders need resources to accomplish their objectives, regardless of whether those goals involve private interests or public goods. A financial shortfall is thus a strong reason to question the status quo. In early European countries, for example, security threats had not only the direct effect of motivating state building but also indirect consequences on the relationship between citizens and rulers. Leaders in need of revenue to provide for collective defense were often forced to make concessions to citizens reluctant to pay taxes. As a result, expanded tax collection was frequently associated with increases in popular representation and the provision of public goods.21 This link was particularly strong in the United Kingdom and its colonies, explaining the resonance of the U.S. revolutionary protest, “no taxation without representation.” Mick Moore and other researchers at the International Centre for Tax and Development are exploring the possibilities for and obstacles to the emergence in the developing world of a similar social contract linking broad-based tax collection to effective and responsive governance.22

Researchers have come to a strong consensus on the negative version of this relationship: governments tend to perform poorly when leaders possess large revenues from natural resources—especially oil rents—and so do not need to tax their own citizens. This so-called resource curse has several causes. Because resource revenues usually go directly to the central government, they contribute to the centralization of power and create easy opportunities for corruption.23 Resource wealth also removes key incentives for politicians to attempt governance reforms: Why take risks to improve government effectiveness and promote broader prosperity if it is possible to reap huge profits from the status quo?24

Unstable global resource prices can create urgent fiscal needs, but the instability is often too brief to sustain reform commitment. In Angola, for instance, a decline in oil revenues sparked high-level interest in reforming taxation of the non-oil economy, but political will appears to have declined as oil prices rebounded.25

Foreign aid can play a similarly detrimental role for governance reform commitment, though research findings are mixed. Some scholars emphasize that development assistance comes with many more conditions than natural resource rents and is unlikely to create the same problems; others find evidence of aid’s negative effects on institutional development and government accountability.26 In a study of the surprising peace-building and governance successes of Somaliland, a breakaway territory from Somalia, Sarah Phillips contends that a lack of external involvement and aid to the government forced former president Mohamed Ibrahim Egal to bargain with local elites for resources and promoted locally driven problem solving.27 Ken Menkhaus contrasts Somaliland’s successes with the experience of Somalia, where, he argues, aid dependence has created disincentives to govern well.28

In addition to security and fiscal needs, political competition has become an increasingly potent threat to leaders across the developing world, and thus an important driver of reform. The threat of losing an election can create incentives for politicians to invest in effective governance to deliver public services. André Borges, for instance, contends that education reform was more successful in Brazilian states with higher political competition. Jos Mooij similarly attributes reformist commitment in Andhra Pradesh to the rise of the middle class and increasingly assertive lower castes, which convinced leaders they could not depend only on preexisting patronage networks.29 Elections can also bring new leaders to power who are more responsive to reform-focused constituencies and interests, as occurred in Medellín and Bihar.

Not all governance reforms are equally difficult to enact, and leaders may commit to policy change even when they do not face an imminent loss of power or threat of insolvency. Political will to reform property-tax collection, for instance, could arise from a predicted reduction in revenues rather than an acute fiscal crisis. Yet case studies suggest that the most difficult and transformative governance reforms require correspondingly serious threats to mobilize sufficient elite support. Relatively moderate forms of “demand-side” political pressure—such as publicizing report cards on public service quality—have gained substantial donor support but in many cases do not constitute a sufficient threat to force a response from power holders, especially in nondemocratic settings.30

Balancing Threats and Feasibility

Threats are powerful but volatile drivers of reform commitment. Severe security challenges, economic crisis, and political competition can create strong incentives to make governance improvements, but they also tend to shorten time frames and make institutional reforms appear less feasible. Conversely, leaders who enjoy enough political dominance to plan far into the future and to push through significant policy changes may feel secure in their current positions and possess fewer incentives to transform the status quo. Political commitment does not arise from either threats or political security alone but rather from a dynamic tension between the two.

Leaders are more likely to invest in effective governance if they expect to maintain power long enough to see the benefits of their reforms.31 Mancur Olson illustrates this point through the analogy of stationary and roving bandits. Both are predatory, but the stationary bandit continues to exploit one territory for a long period and thus has an interest in encouraging economic prosperity to increase his future revenues.32 Peter Lewis and Tim Kelsall both make a similar argument that well-organized, elite rent seeking that is oriented toward the long term can be compatible with development-enhancing policies, respectively pointing to Indonesia under former president Suharto and certain periods in African history such as Kenya under former president Jomo Kenyatta.33

In addition to long time frames, political commitment requires some sense that proposed reforms are viable. As Derick Brinkerhoff notes with regard to anticorruption efforts, key actors may come to the conclusion that it is “best not to try if we aren’t sure we have the means to make progress.”34

Some of the most important factors affecting feasibility are the resources available, the prospects for getting policy changes approved by other parts of the political system (such as the legislature), and the implementing bureaucracy’s competence and discipline. Atul Kohli, for example, describes how a strong and disciplined Communist Party in India’s West Bengal state enabled the creation of an extensive land-reform initiative and helped facilitate and monitor the program’s rollout.35 Other case studies in Latin America and India highlight the advantages of strong executives with extensive powers and leverage over disciplined parties and legislatures.36

Political insecurity, on the other hand, can inhibit commitment to governance improvements. Nicolas van de Walle and William Reno both explain how the fiscal constraints imposed by structural adjustments in African countries led some rulers to hollow out their administrations and withdraw from development promises in order to redirect scarce resources toward elite rent seeking. Amid weak bureaucracies and neopatrimonial politics, leaders calculated that it was politically easier and safer to maintain power by cutting deals with dominant political and social actors instead of attempting the much more difficult task of winning public support through effective delivery of public services.37 Impending elections can create similar pressure for politicians to focus on the short-term distribution of political benefits rather than long-term institution building.

Unstable political contexts can also promote the rise of predatory actors. Those who prosper amid weak governance and insecurity tend to develop a stake in the continuation of those problems. Pierre Engleburt and Denis Tull, for instance, argue that external state-building programs in postconflict countries tend to fail in part because donors assume that domestic power holders agree that state failure is a problem and that peace and good governance are the solutions. Yet warlords and politicians in conflict countries often profit from the corruption opportunities provided by governance deficiencies, and they may actively stall reforms in order to maintain foreign aid revenues.38 In these situations, actors interested in better governance—such as the business leaders and civil society groups that played a key role in Medellín’s transformation—may exist but tend to lack institutionalized paths to power or the capacity to rein in warlords.39

Yet just as crisis situations can hinder reform commitment, so too can political dominance. Leaders with extensive powers and financial resources that face no serious threats can easily entrench predatory governance and dismantle independent institutions. While Kelsall decries the “competitive clientelism” of electoral politics, he acknowledges that centralized rent seeking without developmental commitments can result in kleptocracy.40 The “stationary bandit” argument that long-term rent extraction requires investments in internal economic prosperity fails if political leaders gain revenue from external sources, such as natural resource exports or foreign aid.

Moreover, the benefits of strong executives and disciplined parties only apply if these actors are the main proponents of reform, which is not always the case. While Atul Kohli considers the land reform efforts in West Bengal in the 1970s and 1980s a success, for instance, Sumir Lal argues that the same powerful party structure that enabled that reform currently stifles the ability of rural citizens to demand better schools.41

These tensions help explain why neither democracy nor authoritarianism per se leads to better governance. The most promising formula for political commitment instead appears to be a serious and persistent threat alongside sufficient institutional capacity and political continuity to make governance reform appealing. This can manifest in multiple ways. East Asian developmental states, for example, had external security concerns along with strong bureaucracies and authoritarian rule. Reformists in Medellín and Bihar lacked authoritarian control but faced both security and public pressures to deliver governance improvements; they had democratic pathways to power as well as the political support and executive authority to make reforms feasible within their terms in office.

The Role of Leadership and Beliefs in Reform Commitment

Political will is strongly shaped by external incentives such as the threat environment and existing levels of political stability and institutional capacity. Yet each of these factors also leaves substantial room for interpretation. Actors and social groups may not agree about which threats are most pressing, which solutions are required, and how likely reforms are to succeed. Some politicians are willing to take large risks to transform governance because they have high confidence in their personal leadership abilities, expect that they will maintain power in the future, and wish to leave a lasting legacy. Other leaders in similar circumstances are more cautious and may not plan beyond the next election.

Countries with weak governance tend to face multiple threats. It is often not clear which one should take precedence. If a politician is grappling with a severe shortfall in government revenue but also needs to stand for election, does it make sense to advocate increased taxes? Or, if a leader faces both the threat of an external attack and a potential military coup, is it rational to strengthen the army? The answers to these questions depend on the specifics of each circumstance as well as the beliefs of decisionmakers.

The potential for diverging interpretations of threats is clear in the contrast between Somalia and Somaliland mentioned above. Both territories share a history of state predation and clan violence but have reacted in very different ways. Ken Menkhaus describes how Somalia’s experiences have fostered a zero-sum belief among the population that whoever gains control of government will use that power against everyone else. The central government, as a result, is considered a bigger threat than decentralized civil conflict. Local power holders are willing to invest in local order and services but oppose and sometimes actively undermine a strong national government.42 By contrast, Sarah Phillips argues that in Somaliland, similar experiences have fostered shared beliefs in the overriding importance of peace. This has encouraged ordinary citizens and key power holders to support a government capable of mediating between factions. The perceived threat of civil conflict is so serious that people are reluctant to criticize the government.43

Even when actors agree on a threat, they may have very different ideas as to the best solution. For example, popular social mobilization and unrest helped motivate multiple Latin American coups in the 1960s and 1970s. Most of these coups were conservative right-wing responses to elected left-leaning leaders. The 1968 Peruvian coup was led by military officers who shared the same concern with social unrest as their colleagues elsewhere in the region but crafted a very different response. The Peruvian officers decided that maintaining social order required embracing rather than repressing leftist ideas. They believed elected officials were incapable of promoting the social inclusion and economic transformation that the country required. Alfred Stepan attributes this reaction to the Peruvian military’s internal education programs.44

Similarly diverse reactions are evident in responses to economic problems. James Scott describes the “high-modernist” aims of state leaders from China to Tanzania who believed they could develop economic self-sufficiency with ambitious—and often disastrous—technocratic state planning to reorder and modernize society.45 Peter Evans argues that governments in countries like South Korea and Japan, by contrast, recognized the limits of state power and only selectively intervened in the economy to promote industrial transformation.46 Other developing country governments have adopted more explicitly pro-poor policies, and scholars often highlight the importance of leaders’ ideological and moral commitments to fighting poverty.47

Just as importantly, leaders differ in their feasibility calculations. A governance problem that one politician may dismiss as unsolvable might appear to another as a good opportunity. Part of this has to do with differences in actual and self-perceived political abilities among leaders. Skilled leaders can sometimes rally other parts of the political system around their plans. Subnational executives in Bogotá, and in the Indian states of Andhra Pradesh and Madhya Pradesh, for instance, faced legislative obstacles to reform but ultimately secured support for their priorities by mobilizing public pressure for change and convincing lawmakers that reform would enhance both their access to rents and their reelection prospects.48

The political viability of reforms also depends on the quality of proposed solutions. Dani Rodrik notes that innovative ideas can create benefits for both established interests and ordinary citizens, reducing reform constraints and increasing the political gains for leaders. He points to the success of elite efforts in Japan, the United Kingdom, and Germany in encouraging economic transformation through industrialization while maintaining political power.49 In analyzing the success of pro-poor reforms in Brazil, Uganda, and the Indian state of Madhya Pradesh, Melo, Ng’ethe, and Manor observe that all three cases involved realistic and well-sequenced proposals promoted by skilled politicians. Leaders in these places were capable of building public support, outmaneuvering political opponents, and shaping reforms in a way that enhanced their own power.50

Skillful and innovative reform leadership does not have to come only or even primarily from political officials. Social actors can employ similar techniques to gain support for desired policy changes. Raul Fabella and Jaime Faustino, for instance, describe the role of “development entrepreneurs” in identifying “technically sound, politically possible” reform ideas and working politically to mobilize coalitions for economic policy reforms in the Philippines. These development entrepreneurs included scholars, civil society leaders, lawyers, and other social actors.51

Political commitment results from a complicated mixture of threats and political security, incentives and beliefs, and contextual factors and individual abilities. Yet it is not impossible to analyze. Existing scholarship and case studies of governance reform efforts highlight the role of serious threats in motivating commitment as well as the importance of feasibility concerns, ideas, and leadership skills in translating those perceived threats into developmental governance solutions.

Improving Bureaucratic Effectiveness

Even the most talented and committed leaders cannot simply will policy change into force. They need financial and technical resources and a competent administrative apparatus to carry out their decisions. Resources clearly assist governance but are not usually the most important binding constraints on reform. If they were, traditional donor projects providing financial and technical assistance would have a better track record. Weak bureaucracies, meanwhile, pose a persistent and often seemingly intractable obstacle to effective governance. New initiatives with substantial resources and political support can easily founder if they are entrusted to a poorly managed, corrupt, and underqualified civil service.

Yet civil service overhauls have consistently proved to be among the most difficult targets for international donors and domestic reformers.52 Patronage-based hiring in government undermines the creation of meritocratic, professional bureaucracies but also serves as a key source of political support for politicians. And—as seen above—low administrative capacity reduces the feasibility of reform success and can diminish political commitment. In an apparently vicious cycle, the weaker the bureaucracy, the less likely it can deliver quality public goods and the more leaders may depend on patronage rather than public-service promises to maintain power.

To escape such a low-capacity equilibrium, policymakers must find ways to work with the administrative resources they can access, imperfect as those may be. Scholars have documented multiple strategies to navigate unfavorable bureaucratic conditions. Some approaches focus on the reform process itself rather than any specific endpoint, emphasizing the importance of moving away from best-practice blueprints toward more flexible, context-specific approaches. Others examine a specific type of institutional coping mechanism—the creation of high-performing technocratic enclaves insulated from the rest of the bureaucracy. Enclaves cannot substitute for a more broadly functional civil service, but some of their successful management strategies have been usefully applied in wider bureaucratic settings.

From Best Practice to Flexible Process

While development programs are frequently controversial, few aid approaches have garnered more widespread criticism than the extensive donor efforts to replicate institutional best practices that exist in Western countries without regard for the needs of specific developing-country contexts. Capacity-building efforts often focus on training bureaucrats on proper procedures, for example. These programs implicitly assume that corruption and other deficiencies arise from a lack of knowledge rather than deeper political economy drivers.53 This approach tends to result in reform efforts either bouncing off adverse circumstances or ending in superficial measures that appear to comply with best practices but do not lead to real internal change—what Matt Andrews, Lant Pritchett, and Michael Woolcock term “isomorphic mimicry.”54

Development agencies’ policies—if not always their practices—have increasingly recognized the need to move beyond best-practice approaches. The World Bank’s 2000 public sector strategy, for instance, acknowledges that “too often . . . the Bank’s efforts at reform have relied on foreign or ‘best practice’ models that do not necessarily fit well with country circumstances and capabilities. . . .  This strategy emphasizes the need to start with what exists on the ground and to clarify which reform options ‘fit’ well in specific settings.”55 Subsequent World Bank governance policy documents have echoed the same ideas.

A 2006 reference document on capacity building from the Organization for Economic Cooperation and Development (OECD) notes that “until recently, capacity development was viewed mainly as a technical process, involving the simple transfer of knowledge and organizational models from North to South.” The paper argues that a “new consensus” has emerged around the idea that capacity development must be a locally driven process sensitive to contextual differences, and that domestic political leadership has an important role to play in that process.56

What does a context-sensitive approach to governance mean in practice? One common feature is a willingness to experiment with different institutional forms. Dani Rodrik argues that developing-country economic success stories from Botswana to China have relied on institutions that departed from established practices but accomplished their functions effectively because they fit well with the particular local context.57 David Booth similarly advocates for a “best fit” approach and a focus on “practical hybrids” that mix elements of local cultures and modern states.58 The downside of context-specific institutional solutions is that, as Rodrik warns, replicating successful innovations in other places is difficult.59

In addition to keeping an open mind about what good governance may ultimately look like, some scholars and practitioners advocate for a flexible reform process that responds to the inevitable obstacles that crop up when implementing new programs in weak bureaucratic environments. As early as 1989, Derick Brinkerhoff and Marcus Ingle proposed a “structured flexibility” approach that included built-in expectations of uncertainty and provided room for iterative learning and changes in strategy during the program.60 Matt Andrews, Lant Pritchett, and Michael Woolcock have more recently put forth what they call “problem-driven iterative adaptation” as a means of improving state capabilities. This approach involves several components, including starting from a locally determined problem, creating an environment conducive to experimentation, learning from built-in feedback mechanisms, and engaging a broad range of actors.61

Matt Andrews contends that reform projects with these characteristics perform better on average than inflexible ones. He provides multiple examples of how flexible program designs allow implementers to deal with constraints as they arise and take advantage of new opportunities. Andrews cites, for instance, an HIV/AIDS initiative in Pakistan that used regular progress reports to identify and act on emerging obstacles to program success, such as inadequate contractor reporting mechanisms. He also points to an HIV/AIDS program in Burundi that was flexible enough to introduce new antiretroviral drugs when they became available.62

Flexible approaches have significant promise, but as Brinkerhoff and Ingle emphasize, they depend on strong commitment to reform from key decisionmakers as well as the political space and continuity for the programs to develop.63 Externally supported governance programs that attempt a flexible approach require not just domestic political will but also the willingness of aid providers to fundamentally rethink how projects are designed, funded, and monitored.64 Absent robust commitment to reform and to monitoring mechanisms, overly flexible programs can fall into what one World Bank review calls “ad hoc incrementalism”—using gradualism as cover for superficial changes that do not link up to broader reform efforts.65

Enclaves of Excellence

While context-specific institutional design may be necessary in certain cases, some responses to weak bureaucracies are widely replicated. One of the most popular adaptive solutions involves setting up special technocratic enclaves to implement priority initiatives. These units, often referred to as “islands of excellence” or “pockets of effectiveness,” are supposed to be insulated from wider bureaucratic problems and usually offer higher salaries and more meritocratic hiring and staff management.

This enclave strategy is common in high-stakes and specialized areas of state administration, such as central banks and revenue agencies, where a well-trained professional staff is indispensable. Following this trend, countries across Anglophone Africa have established semiautonomous revenue administrations in an effort to professionalize tax collection.66 Improved performance in these elite bodies does not necessarily translate into broader governance improvements, however. They can instead attract skilled staff away from other government departments and may generate revenues that are redirected into patronage politics.67

Ricardo Soares de Oliveira describes an extreme example of these problems in Angola, where despite the government’s severe bureaucratic deficiencies, the state oil company is staffed with respected technocrats able to deal credibly with international commercial partners. The organization’s profits, however, fund a predatory state, creating “a strange marriage of the latest expertise and market savvy with the narrow enrichment goals of a failed state leadership.”68 This combination may appear incongruous if outsiders assume that bureaucratic efficiency represents a generalized commitment to good governance, but it is not surprising when considered within the context of Angola’s highly oil-dependent and authoritarian political economy.

Bureaucratic enclaves can also be charged with broader reform efforts that contribute more directly to government service delivery. In the Indian state of Madhya Pradesh, for example, former chief minister Digvijay Singh set up Rajiv Gandhi Missions to lead reforms in areas such as primary education and watershed management. These gave an elite group of civil servants under the chief minister’s direct supervision special authority to bypass other parts of the bureaucracy.69

Enclaves can be found across the globe, but their efficacy varies widely. Simply proclaiming that an administrative unit will be independent and meritocratic is insufficient to guarantee good performance. Scholars stress the importance of pairing these basic characteristics with sustained high-level political support, organizational features such as autonomy and clearly defined goals, and good management.70

Successful managerial reforms often involve increased accountability and performance incentives for employees. Gael Raballand and Anand Rajaram, for instance, describe how the Cameroonian customs agency improved the attitude and effectiveness of its agents through performance contracts that clearly spelled out expectations and targets. The agency also combined positive incentives such as congratulatory letters with negative ones like the threat of transfer to a less financially lucrative position. The authors note that these efforts depended on the political commitment and demand for reform from the head of customs.71

Reviews of successful enclaves also highlight what Merilee Grindle calls “organizational mystique.” In a comparison of bureaucratic performance in six developing countries, Grindle finds that neither the task an organization performed nor its salary levels are good predictors of effectiveness. She emphasizes instead the importance of autonomy for managers and high performance expectations for staff. Grindle puts particular emphasis on organizational mystique; that is, widely shared beliefs among employees that they and the organization have an important mission to fulfill and can make unique contributions to the cause.72

More recent research by Michael Roll similarly describes how organizations can improve their effectiveness by instilling a greater sense of purpose and pride in employees, assisted by positive feedback and improved benefits.73 Both Roll and Grindle explain that this strong organizational identity relies partly on employees’ perceptions that they are part of a select elite separate from the rest of the bureaucracy.

Beyond Enclaves

Despite this emphasis on exclusivity, similar management strategies aimed at increasing accountability and instilling organizational pride in civil servants have proved effective outside of elite technocratic enclaves.

In a study of a program from the 1980s and 1990s that extended health services in the Brazilian state of Ceará, Judith Tendler attributes successful implementation in large part to the unusual dedication of community health workers to their jobs. She believes this dedication resulted from several factors. First, while most of these workers were unskilled and lacked the benefits of ordinary civil servants, they were recruited through a transparent and competitive process that lent prestige to their positions. Second, the state government publicized the program widely and actively promoted the idea of civil service as a moral calling, which encouraged greater public respect for health workers. Third, the government urged citizens and civil society groups to monitor the program’s performance and demand improved services.74

Leaders in several Brazilian and Indian states overcame the lack of performance incentives in the regular civil service by hiring contract workers who brought needed skills and manpower but enjoyed less job security than ordinary bureaucrats.75 Some Indian states also shifted vital services such as employment and land records to electronic systems to facilitate implementation and reduce corruption.76 Mathew Devlin and Sebastien Chaskel describe a campaign by then mayor Antanas Mockus’s administration in Bogotá to do all of the above: increase performance monitoring of bureaucrats, remove some of their discretion in taking cash payments, and simultaneously raise morale by publicly revaluing their roles and contributions as civil servants.77

The best strategies for improving the performance of bureaucrats depend on multiple factors, including the task involved, the potential for corruption, and the number of employees required. But these cases indicate that meritocratic processes, increased performance monitoring, and an emphasis on a public service mission can play a crucial role in a variety of bureaucratic settings. These approaches rely on sustained political commitment from top political officials as well as talented managers.

Models of Productive State-Society Relations

Political leaders and state administrations are not the only actors that influence governance reform. Citizens play central roles in shaping government priorities and resisting or facilitating state actions. Public involvement in governance can also constitute a key check on state power. A cohesive political leadership that is motivated by shared threat perceptions and enabled by an efficient administration yet lacks public accountability may use its power to repress citizens and enrich a privileged group.

For these reasons policymakers and scholars have reached widespread consensus on the importance of state-society relations to governance outcomes. Donors initially worked in this area through programs that encouraged citizen participation in development decisionmaking processes and fostered civil society development. But over the past decade, major development actors have shifted from discussing state “supply” and social “demand” as two separate needs toward cultivating interactions between state and society. A domestic accountability report by the governance network of the OECD’s Development Assistance Committee, for example, emphasizes the “need to move beyond a narrow focus on supply-side versus demand-side accountability support, or a focus on formal institutions, and instead to look more closely at the linkages among actors and how these can be strengthened over time.”78

What stronger or more productive state-society relations mean in practice is less clear, however. Close linkages between political elites and citizens are not always beneficial to development. Indeed, patronage relationships are among the most frequently cited governance problems in developing countries. Politicians use state resources to favor particular constituencies in exchange for political support, undermining bureaucratic effectiveness and the provision of broad-based public services.79 Furthermore, some types of social organizations, such as criminal networks or extremist groups, can seriously undermine good governance, especially if they find allies in state institutions. 

When development agencies discuss strengthening state-society relations, they have a very different interaction in mind. In the mainstream donor framework, public-spirited civil society organizations and individual citizens participate in inclusive consultations with the government and hold state officials accountable for delivering developmental policies. This focus on inclusion and participation seems to promise more attention to public rather than private goods.

Some scholars, however, point to a more exclusive mode of productive state-society relations. Based on case studies of some Asian and African countries, they argue that strong informal linkages between state officials and select private actors can form the basis for robust alliances that support developmental policies. Both this framework and the inclusive participatory approach rely on certain supportive conditions, above all political commitment and a competent and responsive state administration.

The Promise and Limits of Inclusive Participation

The inclusive participatory model of state-society relations is powerfully attractive. It appears as the most promising way to build not only political commitment for reform but the “right” kind of political will. Security or fiscal threats can sometimes encourage authoritarian state building and the political threat of elections can intensify patronage distribution. But pressure from citizens and nongovernmental organizations (NGOs) for better state services seems to be narrowly targeted at encouraging public-goods-oriented responses from government. Moreover, citizen-led groups can contribute to improved service quality by monitoring bureaucratic performance. In some cases social actors also substitute for weak state capacity by directly aiding the government in delivering public services.

Advocates of this approach can point to some notable success stories. Among the most famous achievements are the participatory budgeting experiment in Porto Alegre, Brazil, and the launching of citizen report cards for public services in Bangalore, India, which highlighted serious problems in service delivery and led to improvements in quality.80 These initiatives have been widely replicated in other developing countries and have had some success in increasing the proportion of government budgets devoted to social spending and improving state responsiveness to public complaints about services.81 Civil society advocacy has also played a key role around the world in encouraging greater budget transparency, advocating for the rights of marginalized groups, exposing government corruption, advising legislators on policy reforms, and much more.

The role of inclusive participatory ideas is clear in many case studies of governance reform described above. Medellín’s transformation relied on the rise of a new political coalition formed by academics, community associations, NGOs, business groups, and other civil society actors strongly committed to positive change.82 The city government later engaged in several transparency and participatory budgeting efforts and advocated “civic pacts.” The mayoral administration consulted with communities on proposed projects and promised resources in exchange for communities’ involvement and support for the projects.83

Judith Tendler considers public monitoring of community health workers to have been a key success factor in Ceará.84 Reforms in Bihar and Madhya Pradesh similarly relied on enhanced government transparency and efforts to encourage local monitoring of state services, such as making teachers more accountable to local elected councils.85

The developmental potential of citizen advocacy has led to efforts in many countries to pass new laws that increase government transparency and public participation. It has also sparked extensive donor funding for civil society development and other efforts to promote social accountability. Yet as these participatory efforts have expanded, their limitations have also become increasingly clear.

Critiques of the simplistic application of the inclusive participatory model in divergent contexts have accumulated into a substantial body of literature.86 One major concern is that citizen-participation programs are often conceived as both advancing a universally accepted public good and as advocacy initiatives designed to challenge the status quo. This is a difficult balance to maintain. Inclusive consultation rarely yields a general consensus. Even reform proposals that seem to serve clear developmental ends are usually hotly contested within societies. Improving the reliability and efficiency of electricity provision, for instance, could also mean increased costs and reduced access for some people. Moreover, civil society organizations involved in policy advocacy have their own interests and may not be as public-interested or representative of ordinary citizens—especially the poor—as they appear to external funders. Even broad coalitions do not speak for everyone and clash with other segments of society they believe oppose good governance.

While participatory programs are designed to shift leadership incentives and power relations in favor of public-goods-oriented policies, they are deeply constrained by existing power structures. As mentioned earlier, the pressure that ordinary citizens can apply to politicians through formal participatory mechanisms often appears as less of a threat than other demands on leaders. Some forms of participation can actually be disempowering if they legitimize processes in which citizens have little say. Transformative social change tends to come not from apolitical and technocratic NGOs but from politically influential actors, such as social movements or religious groups that external donors are usually uncomfortable supporting. More assertive actions tend to be very risky because they can spark state retaliation and endanger activists as well as donor-recipient relations.

The success of public participation efforts in contributing to better governance thus seems to depend on supporting political conditions. These can include either state officials’ prior commitment to reform or a political system that allows organized citizens to influence and potentially replace political leaders. Many of the successful participatory programs described above were initiated by reformist leaders who actively encouraged public participation while implementing their plans.

In a review of transparency and accountability initiatives, Rosie McGee and John Gaventa note that the original participatory budgeting experience in Porto Alegre relied on “a political leadership highly committed to its success.” They further emphasize that transparency and accountability initiatives more generally rely on relatively open and democratic political environments that include freedom of expression.87 David Booth focuses even more strongly on elite political will, arguing that “both voter pressure and client pressure are weak factors for improving performance unless there is some kind of shift in incentives from the top down.”88

Bottom-up pressure also requires a competent state to respond. Strong citizen movements or popular revolutions can in some cases replace leaders and motivate greater political will, but they often find it much more difficult to create effective bureaucracies. In her comparison of bureaucratic successes and failures, Merilee Grindle notes that in some instances “demand was so overwhelming on poverty-stricken and hopelessly understaffed and undersupplied organizations that it resulted in even poorer morale among employees and a greater propensity to shirk.”89 Sparking governance transformation through popular pressure is especially difficult using the moderate forms of citizen participation and monitoring that attract the vast majority of external support. In the most extensive review of local participatory development programs to date, Ghazala Mansuri and Vijayendra Rao stress that “the evidence overwhelmingly suggests that effective community-based interventions have to be implemented in conjunction with a responsive state.”90

These findings do not negate the valuable role that citizen participation can play in contributing to better governance practices. Given the frequent unevenness of political commitment, bottom-up pressure can promote the rise of new leaders and reinforce the developmental efforts of reformists within government against opposition from other state actors.91 Citizen monitoring can fill some gaps in state accountability systems. But civil society activism and spaces for public participation are far from a governance panacea. By themselves, activism and formal participation are unlikely to overcome either a hostile political environment or very weak administrative capacity.

Developmental Patronage?

Some critics of donors’ intensive focus on civil society and citizen participation point out that many significant developmental success stories occurred amid more exclusive state-society relations, including patronage ties. Such dynamics gained prominence in studies of successful Asian developmental states and are increasingly being explored in African cases.

The East Asian tigers—South Korea, Taiwan, Singapore, and Hong Kong—and other Asian developmental states successfully transformed their economies not just despite exclusive state-society relations, but by some accounts because of them. In influential contributions, Ziya Öniş and Atul Kohli both argue that public-private cooperation in the East Asian developmental states relied on a strong political commitment to economic growth paired with selective access to the state that blocked labor interests and other social groups from influence.92 Mushtaq Khan distinguishes between the state actions required to maintain efficient markets and those needed to promote rapid economic growth. He contends that growth requires interventionist state policies that direct resources toward priority sectors and promote technological learning—giving some social actors privileged access to economic rents.93

Productive state-society cooperation in these countries relied on strong informal relationships between the state and private actors. Peter Evans highlights the “embedded autonomy” of states like Japan and South Korea, arguing that their successes drew on their ability to both maintain strong and coherent bureaucracies and encourage deep linkages between bureaucrats and business leaders. These relationships were forged through shared educational experiences and other personal ties.94 Social ties helped bureaucrats understand economic needs, limited arbitrary policies, and facilitated the implementation of state plans. Peter Lewis contends that state–private sector coalitions also enabled economic growth in Indonesia by providing informal guarantees for business.95 Researchers at the Centre for the Future State similarly highlight how informal linkages between politicians and economic actors in Egypt, Indonesia, and China reduced investment risks.96

Drawing on these experiences, scholars at the Africa Power and Politics Program argue that it is possible to identify examples of “developmental patrimonialism” in Africa, particularly in Ethiopia and Rwanda. The scholars contend that economic transformations in these cases rely on particular forms of centralized rent seeking that are compatible with economic growth. David Booth and Frederick Golooba-Mutebi describe an example of close state-business ties in Rwanda, where the ruling party, the Rwandan Patriotic Front (RPF), owns an investment company that is a major shareholder in leading Rwandan businesses. The authors argue that this arrangement provides capital to jump-start new industries in the country and creates a legitimate source of financing for the RPF, decreasing its incentives to engage in state corruption.97

The developmental contributions of exclusive state-society relationships can be significant, but they generally rely even more heavily on robust political will and bureaucratic capacity than inclusionary participation. Preventing privileged relationships between government and business from turning into destructive rent-seeking arrangements requires a high degree of political and bureaucratic discipline—what Mushtaq Khan calls “governance capabilities”—to ensure that state resources go to the right actors and do not fund nonperformers.98

Drawing on this experience, Tim Kelsall contends that developmental patrimonialism requires strong leadership with a long-term orientation, a single or dominant party system, a top-down patron-client network, and a competent and disciplined economic technocracy.99 The other scholars cited above similarly stress the importance of steadfast political commitment, elite cohesion, and capable bureaucracies. Peter Lewis is somewhat less strict, arguing that informal state-society guarantees can partially compensate for weak institutions, but he too emphasizes the need for political resolve and organizational capacity.100

The potential pitfalls of this exclusive model are evident even where it seems to have been beneficial. Booth and Golooba-Mutebi acknowledge that some critics consider the RPF-owned company an anticompetitive instrument that crowds out private investment.101 The historical examples of developmental patrimonialism cited by Kelsall—including Côte d’Ivoire, Kenya, and Malawi in the mid-1960s to mid-1970s—all subsequently broke down into more typically destructive rent seeking.102

The case of Somaliland presents an interesting example of the potential and limitations of building effective governance through exclusive deals with important social groups. Sarah Phillips highlights the role of preexisting elite networks in enabling trust and cooperation among key social and political actors, including shared educational experiences. She also explains how former president Mohamed Ibrahim Egal used personal ties with business leaders to secure loans for the government in exchange for favorable state treatment. This deal limited state power over business and violated norms of state impartiality, but it also helped political leaders exert influence over clan elders and constituted an early building block of capable governance.103

Aspects of both the inclusive participatory and exclusive developmental models of state-society relations can be combined in the same political system. Elected governments frequently find it more productive to coordinate development policies with a select group of important social partners than to attempt to forge a consensus among all elements of society. Citizens in democracies may also wish to avoid responsibility for monitoring service delivery, preferring that state agencies institute efficient top-down controls. And even exclusive authoritarian regimes can sometimes welcome citizen feedback to discipline bureaucrats and build their popular legitimacy.

Peter Evans rejects the idea that embedded autonomy has to be based on restricted relations between the state and business elites. He argues that informal state-society networks can connect governments with unions and other social groups and that contemporary political leaders are likely to need more inclusive embedded autonomy to deal with their varied constituencies.104


Governance reform efforts in developing countries are at a crossroads. In some respects, reform advocates—both in international aid organizations and in developing countries—have made major strides in the past two decades. Governance is now broadly accepted as a key determinant of development outcomes and unprecedented resources are available for governance programs. At the same time, the governance field is plagued by uncertainty about the central drivers of progress and the difficulty of demonstrating clear development impacts from reform efforts. These issues have gained increasing attention as aid agencies put ever-greater demands on staff and grant recipients to provide measurable short-term results.

This pressure understandably contributes to a search for magic-bullet solutions that offer simple, targeted interventions to unlock governance progress. Reform strategies, such as creating bureaucratic enclaves or formal participatory processes, are widely attempted even where they fit poorly with underlying incentives and power relations. This trend partly reflects an insufficient understanding of context among reform advocates. Yet it also stems from the absence of other straightforward solutions. Enclaves may not work well without political commitment, but neither will other bureaucratic reform strategies. If the state is too weak to deliver public services, community consultations may end in frustration, yet exclusive ties with economic elites are unlikely to yield better results.

In these situations, effective reform advocacy requires much more than simply studying the context and selecting the best approach off a menu of strategies. It entails acknowledging that in many cases no single solution can deliver lasting governance progress and that an integrated approach is required to address obstacles on multiple fronts. This in turn relies on a solid understanding of the drivers of political commitment, bureaucratic effectiveness, and productive state-society relations, as well as the interactions among these governance areas.

A number of emerging lessons can contribute to a better understanding of the dynamics of governance reform. Political commitment arises from multiple factors, including the pressures that politicians face from public demands and other serious threats, the political and administrative feasibility of reforms, and leadership skills and beliefs. Advances in bureaucratic capacity rely on political commitment and in some cases social monitoring as well as good management and institutional innovations. Popular pressure and strong formal or informal relationships between state and society can each help improve political will and bureaucratic performance but also depend on adequate levels of both.

The interdependence among governance areas can create seemingly vicious cycles of low political commitment, low bureaucratic capacity, and unproductive state-society relations, whereby progress in any area is blocked by deficiencies in the other two. Yet the relationships between different aspects of governance can also inform positive reform approaches. Recognizing the need for specific governance conditions that enable reform, for example, can help identify where certain strategies will be most valuable. Creative solutions such as participatory budgeting should not be dismissed simply because they do not work everywhere. When political conditions for governance transformation are clearly unfavorable, programs can work strategically to lay the necessary groundwork to improve long-term reform prospects. This could include, for instance, experimenting with new administrative strategies at lower levels of government or building citizen groups’ capacities to push for policy changes—both steps that are likely to make it easier to take advantage of political opportunities in the future.

Under the right conditions, it is possible to advance relatively quickly and simultaneously on political will, state capacity, and state-society relations. Lower starting levels of governance create many obstacles but they also offer some opportunities. In places like Medellín or Bihar, the perception that governance had reached a nadir helped mobilize political will around reform and offered rewards to politicians who demonstrated that the state could do something for its citizens. When public expectations are very low, even modest improvements in governance can spark positive cycles of increased citizen engagement, higher bureaucratic morale, and continued pressure for progress.

To deepen the development community’s understanding of reforming governance in developing countries, more research is required into successful transformations in seemingly unfavorable contexts. Additional studies will be especially valuable if they avoid focusing on just one reform driver—such as a talented political champion—and instead explicitly consider previous research findings and identify the background conditions that made successful strategies possible.

Governance reform is a multifaceted and often frustrating endeavor. Simply understanding the key obstacles to and potential facilitators of governance progress in any given context is a huge task for researchers and practitioners alike. Devising integrated strategies to promote governance improvements is even more difficult. Yet reform advocates have little choice but to grapple with these challenges. The cost of relying on an apparent magic bullet—in all but the most favorable contexts—will likely be reform failure. Successful governance transformation, meanwhile, can yield large payoffs in the form of significant and sustainable improvements in development outcomes.


About the Author

Diane de Gramont is a nonresident research analyst in Carnegie’s Democracy and Rule of Law Program. Her research specialties include international support for democracy and governance, the use of political analysis in development assistance, and the politics of governance reform. She holds a master’s in comparative government from the University of Oxford and is currently a JD candidate at Yale Law School. She is co-author (with Thomas Carothers) of Development Aid Confronts Politics: The Almost Revolution (Carnegie Endowment, 2013).
This research was completed with the support of the University of Oxford Clarendon Fund. I am grateful to Nic Cheeseman for helping me develop many of my ideas on this subject. I also thank Thomas Carothers and Rachel Kleinfeld for their invaluable feedback on earlier drafts of this paper.

The Carnegie Endowment gratefully acknowledges the support from the UK Department for International Development that helped make the research and writing of this paper possible. The views expressed herein are the responsibility of the author alone.

1 Merilee Grindle, “Good Enough Governance Revisited,” Development Policy Review 25, no. 5 (2007): 533–74; David Booth, Working With the Grain and Swimming Against the Tide: Barriers to Uptake of Research Findings on Governance and Public Services in Low-Income Africa,Working Paper 18, Africa Power and Politics Program, Overseas Development Institute, 2011; and Matt Andrews, The Limits of Institutional Reform in Development: Changing Rules for Realistic Solutions (Cambridge: Cambridge University Press, 2012).

2 Murithi Mutiga, “Uganda’s Stubborn Autocrat,” New York Times, February 24, 2014,; and Susan Rice, testimony, hearing on “Democracy in Africa: The New Generation of African Leaders,” U.S. Senate Subcommittee on African Affairs, Washington, D.C., March 12, 1998,

3 Marcus Andre Melo, Njuguna Ng’ethe, and James Manor, Against the Odds: Politicians, Institutions and the Struggle Against Poverty (New York: Columbia University Press, 2012).

4 Merilee Grindle, Going Local: Decentralization, Democratization, and the Promise of Good Governance (Princeton, N.J.: Princeton University Press, 2007).

5 Evan Osnos, “Reform Hinges on Syria’s Leader,” Chicago Tribune, April 22, 2005,; and Charles Krauthammer, “Syria’s ‘Reformer,’” Washington Post, March 31, 2011,

6 I am grateful to Rachel Kleinfeld for this point.

7 Catherine Weaver, Hypocrisy Trap: The World Bank and the Poverty of Reform (Princeton, N.J.: Princeton University Press, 2008).

8 Andrews, The Limits of Institutional Reform in Development: Changing Rules for Realistic Solutions.

9 Verena Fritz and Brian Levy, “Problem-Driven Political Economy in Action: Overview and Synthesis of the Case Studies,” in Problem-Driven Political Economy Analysis: The World Bank’s Experience, ed. Verena Fritz, Brian Levy, and Rachel Ort (Washington, D.C.: World Bank, 2014), 8.

10 James Manor, “Successful Governance Reforms in Two Indian States: Karnataka and Andhra Pradesh,” Commonwealth and Comparative Politics 45, no. 4 (2007): 425–51; and Jos Mooij, “Hype, Skill and Class: The Politics of Reform in Andhra Pradesh, India,” Commonwealth and Comparative Politics 15, no. 1 (2007): 34–56.

11 James Manor notes that Naidu’s attempt to use a women’s credit program to advance his political control in rural areas ended up backfiring and hurting him politically. It nevertheless seemed like a rational political decision at the time. 

12 David Hudson and Adrian Leftwich, “From Political Economy to Political Analysis,” Research Paper 25, Developmental Leadership Program, June 2014, 75.

13 Dani Rodrik, “When Ideas Trump Interests: Preferences, Worldviews, and Policy Innovations,” Journal of Economic Perspectives 28, no. 1 (2014): 189–208.

14 Matt Andrews reviews some of the literature on the role of crises and shocks in provoking a shift in attitudes toward institutional reform, noting that “most models of institutional change suggest that this reflection has to be provoked and usually emerges only when forced by a major disturbance to the status quo.” See Andrews, The Limits of Institutional Reform in Development: Changing Rules for Realistic Solutions, 141.

15 Charles Tilly, Coercion, Capital, and European States: AD 990–1992 (Hoboken, N.J.: Wiley-Blackwell, 1992).

16 Jeffrey Herbst, States and Power in Africa (Princeton N.J.: Princeton University Press, 2000).

17 See, for example, Dan Slater, Ordering Power: Contentious Politics and Authoritarian Leviathans in Southeast Asia (Cambridge: Cambridge University Press, 2010).

18 Ziya Öniş, “The Logic of the Developmental State,” Comparative Politics 24, no. 1 (October 1991): 116.

19 David Booth, Development as a Collective Action Problem: Addressing the Real Challenges of African Governance (London: Africa Power and Politics Program, Overseas Development Institute, 2012), 49.

20 On Medellín, see Kate Maclean, The ‘Medellín Miracle’: The Politics of Crisis, Elites and Coalitions, Research Paper 24, Developmental Leadership Program, March 2014; and Matthew Devlin and Sebastian Chaskel, “From Fear to Hope in Colombia: Sergio Fajardo and Medellín, 2004–2007,” Policy Note, Princeton: Innovations for Successful Societies, 2010. On Bihar, see Vikram Chand, “Transforming Governance in a Lagging State: The Case of Bihar,” in Public Service Delivery in India: Understanding the Reform Process, ed. Vikram Chand (Oxford: Oxford University Press, 2010); Rohan Mukherjee, “Coalition Building in a Divided Society: Bihar State, India 2005–2009,” Princeton: Innovations for Successful Societies, 2010; Arun Sinha, Nitish Kumar and the Rise of Bihar (New Delhi: Penguin Group, 2011); K. P. Narayana Kumar, “A Person of the Year: Nitish Kumar,” Forbes, 3 January 2011,; and “India’s Poorest State: A Triumph in Bihar,” Economist, November 25, 2010.

21 Deborah Brautigam, “Introduction: Taxation and State-Building in Developing Countries,” in Taxation and State Building in Developing Countries: Capacity and Consent, ed. Deborah Brautigam, Odd-Helge Fjeldstad, and Mick Moore (Cambridge: Cambridge University Press, 2008); Margaret Levi, Of Rule and Revenue (Berkeley, Calif.: University of California Press, 1988); and Charles Tilly, “War Making and State Making as Organized Crime” in Bringing the State Back In, ed. Peter Evans, Dietrich Rueschemeyer, and Theda Skocpol (Cambridge: Cambridge University Press, 1985),169–91.

22 See International Centre for Taxation and Development,; Mick Moore, “Revenue, State Formation and the Quality of Governance in Developing Countries,” International Political Science Review 25, no. 3 (2004): 297–319; Brautigam, Fjeldstad, and Moore, Taxation and State Building in Developing Countries.

23 For reviews of the multiple governance consequences of resource wealth, see Michael Ross, “The Politics of the Resource Curse: A Review,” in Handbook on the Politics of Development, ed. Carol Lancaster and Nicolas van de Walle (Oxford: Oxford University Press, 2013); and Terry Lynn Karl, The Paradox of Plenty: Oil Booms and Petro-States, (Berkeley, Calif.: University of California Press, 1997).

24 Ricardo Soares de Oliveira explores this phenomenon in Africa’s Gulf of Guinea. See Ricardo Soares de Oliveira, Oil and Politics in the Gulf of Guinea (London: Hurst and Co., 2007).

25 See Odd-Helge Fjeldstad, Soren Kirk Jensen, and Francisco Miguel Paulo, “Reforming the Non-Oil Tax System in Angola—Mission Impossible?” presentation at 2nd International Conference on “Angola: New Directions in Angolan Research,” St. Peter’s College, Oxford University, March 17–18, 2014.

26 For a review of the evidence in favor of aid undermining governance, see Todd Moss, Gunilla Pettersson, and Nicolas van de Walle, “An Aid-Institutions Paradox? Aid Dependency and State Building in Sub-Saharan Africa,” Working Paper no. 74, Center for Global Development, 2006. For more skeptical views on the negative effects of aid, see Mick Moore, “Obstacles to Increasing Tax Revenues in Low Income Countries,” Working Paper no. 15, International Centre for Tax and Development, 2013; and Nathan Jensen and Leonard Wantchekon, “Resource Wealth and Political Regimes in Africa,” Comparative Political Studies 37, no. 7 (2004): 816–41. 

27 Sarah Phillips, Political Settlements and State Formation: The Case of Somaliland, Research Paper 23, Developmental Leadership Program, November 2013.

28 Ken Menkhaus, “Governance Without Government in Somalia,” International Security 31, no. 3 (2007): 74–106.

29 André Borges, “State Government, Political Competition and Education Reform: Comparative Lessons From Brazil,” Bulletin of Latin American Research 27, no. 2 (2008): 235–54, and Mooij, “Hype, Skill and Class: The Politics of Reform in Andhra Pradesh, India,” 34–56.

30 Rosemary McGee and John Gaventa, “Shifting Power? Assessing the Impact of Transparency and Accountability Initiatives,” IDS Working Paper 383, Institute of Development Studies, 2011.

31 Margaret Levi makes this argument with reference to tax policy, arguing that leaders maximize revenue within the constraints of their bargaining power, transaction costs, and discount rates (time-horizons). See Levi, Of Rule and Revenue.

32 Mancur Olson, “Dictatorship, Democracy, and Development,” American Political Science Review 87, no. 3 (1993): 567–76.

33 Peter Lewis, Growing Apart: Oil, Politics, and Economic Change in Indonesia and Nigeria (Ann Arbor, Mich.: University of Michigan Press, 2007); and Tim Kelsall, Business, Politics, and the State in Africa: Challenging the Orthodoxies on Growth and Transformation (London: Zed Books, 2013).

34 Derick Brinkerhoff, “Unpacking the Concept of Political Will to Confront Corruption,” U4 Brief no. 1, Chr. Michelsen Institute, May 2010, 1.

35 Atul Kohli, The State and Poverty in India: The Politics of Reform (Cambridge: Cambridge University Press, 1987).

36 Manor, “Successful Governance Reforms in Two Indian States: Karnataka and Andhra Pradesh,” 425–51; and Barbara Geddes, Politician’s Dilemma: Building State Capacity in Latin America (Berkeley, Calif.: University of California Press, 1994).

37 Nicolas van de Walle, African Economies and the Politics of Permanent Crisis, 1979–1999 (Cambridge: Cambridge University Press, 2001); and William Reno, Warlord Politics and African States (Boulder, Colo.: Lynne Rienner Publishers, 1999).

38 Pierre Englebert and Denis Tull, “Post-Conflict Reconstruction in Africa: Flawed Ideas About Failed States,” International Security 32, no. 4 (2008): 106–139.

39> William Zartman discusses the basic dilemma that in a post-conflict situation the people with the most power tend to be warlords while more good-governance oriented outsiders from civil society may be preferable leaders but usually lack authority. See William Zartman, ed., Collapsed States: The Disintegration and Reconstitution of Legitimate Authority (Boulder, Colo.: Lynne Rienner Publishers, 1995).

40 Kelsall, Business, Politics, and the State in Africa.

41 Sumir Lal, “Drivers of Reform: Examining a Puzzle in West Bengal,” in Public Service Delivery in India: Understanding the Reform Process, ed. Vikram Chand(Oxford: Oxford University Press, 2010).

42 Menkhaus, “Governance Without Government in Somalia,” 74–106.  

43 Phillips, Political Settlements and State Formation: The Case of Somaliland.

44 Alfred Stepan, The State and Society: Peru in Comparative Perspective (Princeton, N.J.: Princeton University Press, 1978).

45 James Scott, Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed (New Haven, Conn.: Yale University Press, 1998).

46 Peter Evans, “The State as Problem and Solution: Predation, Embedded Autonomy, and Structural Change,” in The Politics of Economic Adjustment: International Constraints, Distributive Conflicts, and the State, ed. Stephan Haggard and Robert R. Kaufman (Princeton, N.J.: Princeton University Press, 1992).

47 See Kohli, The State and Poverty in India: The Politics of Reform; and Melo, Ng’ethe, and Manor, Against the Odds: Politicians, Institutions and the Struggle Against Poverty.

48 See Mathew Devlin and Sebastien Chaskel, “Conjuring and Consolidating a Turnaround: Governance in Bogotá 1992–2003,” Innovations for Successful Societies, 2010; Mooij, “Hype, Skill and Class: The Politics of Reform in Andhra Pradesh, India,” 34–56; and Melo, Ng’ethe, and Manor, Against the Odds: Politicians, Institutions and the Struggle Against Poverty.

49 Rodrik, “When Ideas Trump Interests: Preferences, Worldviews, and Policy Innovations.”

50 Melo, Ng’ethe, and Manor, Against the Odds: Politicians, Institutions and the Struggle Against Poverty.

51 Raul Fabella and Jaime Faustino, eds., Built on Dreams, Grounded in Reality: Economic Policy Reform in the Philippines (Makati City, Philippines: Asia Foundation, 2011).

52 See, for example, DFID, “Governance Portfolio Review Summary: Summary Review of DFID’s Governance Portfolio 2004–2009,” Department for International Development, July 2011; and World Bank Independent Evaluation Group, “Public Sector Reform: What Works and Why? An IEG Evaluation of World Bank Support,” World Bank, 2008.

53 Thomas Carothers notes a similar tendency toward “institutional modeling” and over-reliance on training in democracy assistance. See Thomas Carothers, Aiding Democracy Abroad: The Learning Curve (Washington, D.C.: Carnegie Endowment for International Peace, 1999).

54 Matt Andrews, Lant Pritchett, and Michael Woolcock, Escaping Capability Traps Through Problem-Driven Iterative Adaptation, Working Paper 299, Center for Global Development, June 2012.

55 World Bank, “Reforming Public Institutions and Strengthening Governance: A World Bank Strategy,” Public Sector Board, Poverty Reduction and Economic Management Network, World Bank, September 2000, vi.

56 OECD, “The Challenge of Capacity Development: Working Towards GoodPractice,” DAC Reference Document, Organization for Economic Cooperation and Development, 2006, 15–16.

57 Dani Rodrik, ed., In Search of Prosperity: Analytic Narratives on Economic Growth (Princeton, N.J.: Princeton University Press, 2003).

58 Booth, Development as a Collective Action Problem.

59 Dani Rodrik, “Growth Strategies,” in Handbook of Economic Growth, Volume 1A, ed. Philippe Aghion and Steven Durlauf (Elsevier B.V., 2005).

60 Derick Brinkerhoff and Marcus Ingle, “Blueprint and Process: A Structured Flexibility Approach to Development Management,” Public Administration and Development 9, no. 5 (November–December 1989).

61 Matt Andrews, Lant Pritchett, and Michael Woolcock, Escaping Capability Traps Through Problem-Driven Iterative Adaptation, Working Paper 299, Center for Global Development, June 2012.

62 Andrews, The Limits of Institutional Reform in Development: Changing Rules for Realistic Solutions, 139.

63 Brinkerhoff and Ingle, “Blueprint and Process: a Structured Flexibility Approach to Development Management.”

64 For more in-depth discussion of the obstacles to implementing flexible process approaches within current donor structures and practices, see Thomas Carothers and Diane de Gramont, Development Aid Confronts Politics: The Almost Revolution (Washington, D.C.: Carnegie Endowment for International Peace, 2013), chapter 6.

65 World Bank, Economic Growth in the 1990s: Learning From a Decade of Reform (Washington, D.C.: World Bank, 2005), 301.

66 Mick Moore, Revenue Reform and Statebuilding in Anglophone Africa, IDS Working Paper 2013, no. 428, Institute of Development Studies, 2013.

67 For a discussion of the advances and limitations of semi-autonomous tax administration in Africa, see Moore, Revenue Reform and Statebuilding in Anglophone Africa; and Christian von Soest, The African State and Its Revenues: How Politics Influences Tax Collection in Zambia and Botswana (Baden-Baden: Nomos, 2009).

98 Ricardo Soares de Oliveira, “Business Success, Angola-Style: Postcolonial Politics and the Rise and Rise of SONANGOL,” Journal of Modern African Studies 45, no. 4 (2007): 611. 

69 See Melo, Ng’ethe, and Manor, Against the Odds: Politicians, Institutions and the Struggle Against Poverty, and Manoj Srivastava. “Crafting Democracy and Good Governance in Local Arenas: Theory, Dilemmas, and Their Resolution Through the Experiments in Madhya Pradesh, India?” Crisis States Research Centre, London School of Economics, April 2005.

70 For a review of different hypotheses on this subject, see David Leonard, “Where Are ‘Pockets’ of Effective Agencies Likely in Weak Governance States and Why? A Propositional Inventory,” Working Paper 306, Institute of Development Studies, 2008. See also Michael Roll, ed., The Politics of Public Sector Performance: Pockets of Effectiveness in Developing Countries (New York: Routledge, 2014).

71 Gael Raballand and Anand Rajaram, “Behavior Economics and Public Sector Reform: An Accidental Experiment and Lessons From Cameroon,” World Bank Policy Research Working Paper 6596, September 2013.

72 Merilee Grindle, “Divergent Cultures? When Public Organizations Perform Well in Developing Countries,” World Development 25, no. 4 (1997): 481–95.

73 Roll, ed., The Politics of Public Sector Performance: Pockets of Effectiveness in Developing Countries.

74 Judith Tendler, Good Government in the Tropics (Baltimore: Johns Hopkins University Press, 1998).

75 Chand, “Transforming Governance in a Lagging State: The Case of Bihar”; Melo, Ng’ethe, and Manor, Against the Odds: Politicians, Institutions and the Struggle Against Poverty; Tendler, Good Government in the Tropics.

76 Carolyn Elliott, “Moving From Clientelist Politics Toward a Welfare Regime: Evidence From the 2009 Assembly Election in Andhra Pradesh,” Commonwealth and Comparative Politics 49, no. 1 (2011): 48–79; Manor, “Successful Governance Reforms in Two Indian States: Karnataka and Andhra Pradesh,” 425–51; Mukherjee, “Coalition Building in a Divided Society: Bihar State, India 2005–2009.”

77 Devlin and Chaskel, “Conjuring and Consolidating a Turnaround: Governance in Bogotá 1992–2003.”

78 OECD DAC, “Accountability and Democratic Governance: Orientations and Principles for Development,” DAC Guidelines and Reference Series, Organization for Economic Cooperation and Development, Development Assistance Committee, 2013, 31. For more on the evolution of donor policy from demand-side aid to state-society relations, see Carothers and de Gramont, Development Aid Confronts Politics: The Almost Revolution, 171–74.

79 There is now an extensive literature on patronage and the related concepts of neopatrimonialism and clientelism and their impact on state effectiveness. See, for example, Gero Erdmann and Ulf Engel, “Neopatrimonialism Reconsidered: Critical Review and Elaboration of an Elusive Concept,” Commonwealth and Comparative Politics 45, no. 1 (2007): 95–119; Francis Fukuyama, State Building: Governance and World Order in the 21st Century (London: Profile Books, 2005); and Joel Migdal, Strong Societies and Weak States: State-Society Relations and State Capabilities in the Third World (Princeton: Princeton University Press, 1988).

80 Manor, “Successful Governance Reforms in Two Indian States: Karnataka and Andhra Pradesh,” 425–51.

81 Mark Robinson, ed., Budgeting for the Poor (Basingstoke: Palgrave Macmillan, 2008); and Rosemary McGee and John Gaventa, Review of Impact and Effectiveness of Transparency and Accountability Initiatives: Synthesis Report (Brighton: Institute of Development Studies, 2010), 21.

82 Maclean, The ‘Medellín Miracle’: The Politics of Crisis, Elites and Coalitions.

83 Devlin and Chaskel, “From Fear to Hope in Colombia: Sergio Fajardo and Medellín, 2004–2007,” and Maclean, The ‘Medellín Miracle’: The Politics of Crisis, Elites and Coalitions.

84 Tendler, Good Government in the Tropics.

85 Chand, “Transforming Governance in a Lagging State: The Case of Bihar”; and Melo, Ng’ethe, and Manor, Against the Odds: Politicians, Institutions and the Struggle Against Poverty.

86 Some major recent reviews of the limits and potential of participation and transparency initiatives include: Development Research Centre on Citizenship, Participation and Accountability, Blurring the Boundaries: Citizen Action Across States and Societies (Brighton: Institute of Development Studies, 2011); Samuel Hickey and Giles Mohan, eds., Participation: From Tyranny to Transformation? Exploring New Approaches to Participation in Development (London: Zed Books, 2005); Ghazala Mansuri and Vijayendra Rao, Localizing Development: Does Participation Work? World Bank Policy Research Report (Washington, D.C.: World Bank, 2013); McGee and Gaventa, “Shifting Power? Assessing the Impact of Transparency and Accountability Initiatives”; and Alina Rocha Menocal and Bhavna Sharma, “Joint Evaluation of Citizens’ Voice and Accountability: Synthesis Report,” Evaluation Report EV692, Department for International Development, 2008. Also see Carothers and de Gramont, Development Aid Confronts Politics: The Almost Revolution, chapter 5.

87 McGee and Gaventa, “Shifting Power? Assessing the Impact of Transparency and Accountability Initiatives,” 19–21.

88 Booth, Working With the Grain and Swimming Against the Tide: Barriers to Uptake of Research Findings on Governance and Public Services in Low-Income Africa.

89 Grindle, “Divergent Cultures? When Public Organizations Perform Well in Developing Countries,” 486.

90 Mansuri and Rao, Localizing Development: Does Participation Work? 292.

91 This is a key dynamic in Judith Tendler’s examination of Ceara as well as Melo et al.’s account of Madhya Pradesh. See Tendler, Good Government in the Tropics; and Melo, Ng’ethe, and Manor, Against the Odds: Politicians, Institutions and the Struggle Against Poverty.

92 Öniş, “The Logic of the Developmental State,” 109–26; and Atul Kohli, State-Directed Development: Political Power and Industrialization in the Global Periphery (Cambridge: Cambridge University Press, 2004).

93 Mushtaq Khan, “Governance, Economic Growth and Development Since the 1960s: Background Paper for World Economic and Social Survey 2006,” United Nations Department of Economic and Social Affairs, 2006.

94 Evans, “The State as Problem and Solution: Predation, Embedded Autonomy, and Structural Change.”

95 Lewis, Growing Apart: Oil, Politics, and Economic Change in Indonesia and Nigeria.

96 Centre for the Future State, An Upside Down View of Governance (Brighton: Institute of Development Studies, April 2010), 6.

97 David Booth and Frederick Golooba-Mutebi, “Developmental Patrimonialism? The Case of Rwanda,” African Affairs 111, no. 444 (2012): 379–403.

98 Khan, “Governance, Economic Growth and Development Since the 1960s: Background Paper for World Economic and Social Survey 2006.”

99 Kelsall, Business, Politics, and the State in Africa, 26.

100 Lewis, Growing Apart: Oil, Politics, and Economic Change in Indonesia and Nigeria.

101 Booth and Golooba-Mutebi, “Developmental Patrimonialism? The Case of Rwanda,” 379–403.

102 Kelsall, Business, Politics, and the State in Africa.

103 Phillips, Political Settlements and State Formation: The Case of Somaliland.

104 Evans, “The State as Problem and Solution: Predation, Embedded Autonomy, and Structural Change.”