Abstract: This paper reviews research on the political economy of industrial policy and discusses its implications for the formulation of development strategies. The research has been marked by a succession of paradigms, which have brought new issues to light, but often to the detriment of the construction of a cumulative research programme. We survey the main contributions in each research paradigm, including ‘the developmental state’, ‘neo-patrimonialism’, and ‘political settlements’, as well as an Africanist literature on the specificities of African political economies. We then discuss the role of concepts and concept formation in the evolution of thinking on the political economy of development. In the two final sections, we outline what we see as the main findings from the literature. Firstly, we highlight the role of organizational strength and political alignment in development strategies. Secondly, by putting together these political economy factors with a conception of ‘political resources’ and ‘coalitional demands’, we come to formulate the problem of the developmentally-oriented leader as a series of decisions over time on how to allocate political resources so as to reconfigure interests and strengthen organizations, subject to political opposition and coalitional demands. We argue that this formulation can provide the needed structure for leaders’ strategies, help us make sense of past development experiences, and provide a basis for the more systematic study of development strategies.

Download PDF version here: Political Strategies for Industrial Development

  1. Introduction

In recent years, the development policy community has come to pay greater attention to the role of politics in undergirding the success of economic development efforts. The field of industrial policy has experienced a particularly lively debate on political economy[1], largely motivated by the impressive manufacturing-led performances of Northeast Asian ‘miracle economies’ such as South Korea and Taiwan[2]. The debate has led to the development of a plethora of typologies and analytical frameworks aiming to “interpret complex historical evidence” (Khan 2017) as a means of capturing the essence of what makes the politics of a given country more or less conducive to sustained economic growth. However, much of the field has moved towards the incorporation of additional variables in such frameworks, giving rise to more complex schema and at times compromising on analytical parsimony and tractability.

This piece surveys the literature on the political economy of industrialization, with the aim of extracting useful lessons for policymakers aspiring to promote industrialization[3]. Although a number of scholars have recently provided similar reviews of the literature (eg. Hudson and Leftwich (2014), Whitfield et al. (2015)) we differ from previous reviews on two counts. Firstly, we differ from much of the literature in making a clear analytical distinction between leaders’ motives for pursuing industrial policy and the obstacles they are likely to face in implementing their programmes. Second, and building on the first distinction, rather than pursuing another class of explanations for differential developmental performances, we choose to extract lessons from the literature pointing to the main political obstacles likely to be faced by aspiring industrializers. These can help us think on how to go about formulating strategies for dealing with such obstacles. Thus, in the final section we propose framing the challenge of industrialization in terms of maximizing the leadership’s objective (in this case industrialization) using the political resources they can muster, while subject to a set of coalitional demands. We argue that framing the question in this way can provide helpful insights and allow us to sidestep the structure-agency issues that bedevil much of the research on the political economy of industrialization.

Our categorization of the literature is otherwise relatively standard, and we start by discussing the literature on the ‘developmental state’, which is largely fixated with the Northeast Asian experience and operates at a fairly aggregate level. This is followed by a discussion of work that sought to generalize the findings of the first generation of studies on developmental states, and by a brief discussion of a literature on the particularities of African politics and African states. In the second half of the paper, we deal with the emerging ‘political settlements’ paradigm, largely inspired by the work of Mushtaq Khan, which relativizes the political conditions for successful industrial policy, examining the issue through a more disaggregated and explicitly political lens. In the final two sections, we first provide a critical assessment of the literature, before outlining how some of its main findings can be combined for thinking about political strategies for industrialization.


  1. The Developmental State paradigm

2.1 The Northeast Asian Developmental State

The term ‘developmental state’ was originally put forth by Chalmers Johnson (1982) in his book MITI and the Japanese Miracle: The Growth of Industrial Policy, 1925-1975, and referred to the main characteristics of post-war high-growth Japanese political economy. Although Johnson was careful not to make causal arguments about the effects of state intervention on economic growth, but rather sought to understand the context under which policymakers operated (Woo-Cumings 1999), his book had a strong impact. It subsequently sparked the development of a literature concerned with understanding why Northeast Asian states had been so much more successful in promoting industrialization than other states in the developing world, and the concept of ‘developmental state’ came to be applied to the experiences of South Korea (Amsden 1989), Singapore (Rodan 1989), Taiwan (Wade 1990), and Hong Kong (Haggard 1990).

The ‘developmental state’ was a middle-range concept, in the sense that it sprung from an intra-area comparison among relatively homogeneous cases (Sartori 1970). It served a descriptive and heuristic purpose, meant to illustrate the shared institutional and political characteristics of the East Asian high-growth political economies. Moreover, in the context of debates in development between adherents of structuralism and of liberal orthodoxy, the concept introduced a third perspective combining important roles for both the state and the market. Onis (1991) summarizes the basic traits of the developmental state, and is worth quoting at length:

“Economic development, defined in terms of growth, productivity, and competitiveness, constitutes the foremost and single-minded priority of state action. Conflict of goals is avoided by the absence of any commitment to equality and social welfare. Goals formulated specifically in terms of growth and competitiveness are rendered concrete by comparison with external reference economies which provide the state elites with models for emulation. There is an underlying commitment to private property and market, and state intervention is firmly circumscribed by this commitment. The market, however, is guided with instruments formulated by a small-scale, elite economic bureaucracy, recruited from the best managerial talent available in the system. Within the bureaucracy, a pilot agency plays a key role in policy formulation and implementation. Close institutionalized links are established between the elite bureaucracy and private business for consultation and cooperation…Yet another crucial component is a political system in which the bureaucracy is given sufficient scope to take initiatives and operate effectively. The politicians “reign” while the bureaucrats “rule”. The objective of the political elite is to legitimize the actions of the elite bureaucratic agencies and make space for the latter’s actions.”

Onis’ account evidences how the Northeast Asian developmental states had specific institutional underpinnings – property rights, market institutions, and an elite economic bureaucracy with an insulated key economic agency – as well as a nationalistic, developmental ideology; an authoritarian (or semi-authoritarian in the case of Japan) political system; and an alliance between the state and the business elite. Another important feature of the period were the geopolitical considerations that both contributed to the formation of developmental states (Doner, Ritchie and Slater 2005) and assured that Japan, Korea and Taiwan (the countries on which the literature tends to concentrate) could interact in growth-enhancing ways with each other and with the United States (Cumings 1984).

2.2 Developmental Stateness as an Attribute

The seminal work of Peter Evans (1989; 1995) is the first major contribution to the developmental state literature to embark on a cross-regional comparison. He attempts to explain the differences in the developmental experiences of Brazil, India and South Korea by building two ideal types of (what he refers to as) ‘Third World states’: developmental states and ‘predatory’ states. While the former are associated with industrial transformation, the latter “extract at the expense of society”. Of course, being ideal types, most existing states fell at some point of the continuum in between the two extremes. It is important to note that Evans employs the term ‘developmental state’ differently to the original literature on Northeast Asia. Instead of a middle-range concept, Evans’ notion of developmental stateness – and its converse, predatory stateness – can be better conceived of as an ideal type, which existing states can approximate to different degrees.[4] Under his classification, Japan, Korea and Taiwan would fit closer to the developmental end of the scale, the DRC would be an almost pure form of predatory state, and India and Brazil would fall somewhere in between.

For Evans, a state more closely resembles the ideal type of a developmental the more it is endowed with ‘embedded autonomy’. This is the property of forming a coherent entity, possessing its own esprit de corps and responding to internal incentives, typifying an ideal type Weberian bureaucracy (Weber 2009), while simultaneously being connected to society by a set of social ties “that binds the state to society and provides institutionalized channels for the continual negotiation and re-negotiation of goals and policies.” (Evans 1995, p. 12). Embedded autonomy is key for industrial transformation because it allows the state to effectively induce entrepreneurs to enter a new sector (‘midwifery’) and to subsequently assist in industrial upgrading (‘husbandry’) without it being captured by pre-existing or emerging vested interests. In this way, Evans’ is one of the first accounts to modify the concept of ‘developmental state’ so that it can ‘travel’ to other contexts and serve as a universal explanator of developmental performance.[5]

Atul Kohli (2004) differs from Evans in adding Nigeria to his list of case studies, also comprising South Korea, India and Brazil. He offers a more fine-grained conceptual framework, distinguishing between state capacity and coalitional underpinnings in the ideal types of cohesive-capitalist and fragmented-multiclass states. Cohesive-capitalist states share many of the traits of developmental states, and the terminology only differs due to Kohli’s observation that “the idea that developmental states facilitate development” is “too obvious for both analytical and normative comfort” (Ibid. p. 10). These states are cohesive because of their “centralized and purposive authority structures that often penetrate deep into society” and capitalist because of political leaders’ alliance with business leaders and their repression of organized labour. He identifies Korea under Park Chung Hee and Brazil during the Estado Novo and the military dictatorship as examples of cohesive-capitalist states.

A second ideal type of state organization is that of a fragmented-multiclass state. In these states, public authority is more dispersed, and political power depends on a broader coalition of class interests. For this reason, policies are less coherent, as they simultaneously seek to satisfy more constituencies and consequently cannot treat economic growth as their sole aim. Therefore, their efforts to promote industrialization are less successful than those of cohesive-capitalist states. We can see that compared to Evans, Kohli’s framework more consciously differentiates political coalitions (capitalist or multiclass) from the structure of state power (coherent or fragmented), producing a more granular typology of states. In theory, coherent-multiclass or fragmented-capitalist states could also present themselves to analysis, but he remarks that these state forms do not seem to appear among developing countries.[6]

The third ideal type of state organization in Kohli’s framework is composed of neo-patrimonial states. These are states that have a ‘modern’, façade but in practice function according to a ‘patrimonial’ logic – as opposed to a ‘rational-legal’ one – whereby no distinction is made between the public and private realms.[7] As compared to Evans’ predatory state ideal type, which attempts to capture the ‘logic’ of the state apparatus as a whole (ie. the state is either trying to predate or to develop), Kohli’s categorization more closely captures the subjective logic of individuals’ actions, and is closer to Weber’s original conception of how to interpret the meaning of social action (Weber 1991). Kohli traces how a neo-patrimonial logic governed the actions of the Nigerian state and led to disappointing developmental outcomes in areas such as the promotion of entrepreneurship, technology and infrastructure, as well as the forging of a disciplined, productive working class. Throughout the entire post-independence period, Nigerian institutions failed to transcend divisions based on ethnicity and kinship, and could not stop the pursuit of personalistic interests. The relative unimportance of who exactly was at the helm of the state underscores the fact that out of Kohli’s three ideal types of states, the neo-patrimonial state is the only one that does not distinguish between state capabilities and political alliances.

In addition to the elaboration of a conceptual framework to span the types of ‘Third World states’, Kohli analyses the historical roots of variations in state capacity. This constitutes an innovation in relation to the earlier literature on the developmental state, for which the historical horizon did not extend much further back than World War Two. Kohli sees colonialism as the most critical juncture determining the type of state that would be formed in the developing world. The type of colonial state created by colonial powers persisted due to the path dependence of state institutions; thus, the independent Korean state inherited features of the powerful Japanese colonial state, while in Brazil a patrimonial bureaucracy and the power of regional elites remained almost intact even after close to two centuries of independence. However, European colonial powers did not erect their states over a tabula rasa, but had to contend with pre-existing social structures. The latter, together with the type of colonialism practiced in each time period, affected the choice of colonization strategy. An example is the case of Nigeria, which was initially far more fragmented and technologically backward than Korea or India, and where Britain’s ‘colonialism on the cheap’ failed to lead to the emergence of a public sphere, perhaps even accentuating the preponderance of personalistic networks. Finally, Kohli notes the ability of nationalist movements to alter the institutional legacies of colonizing states, albeit in an incomplete manner, as in India. We revisit the theme of historical legacies in later in this paper, when we turn to a deeper exploration of the concept of ‘neo-patrimonialism’ in the study of African states.

2.3 The Politics of the Developmental State

Expectedly, soon after its inception, the developmental state paradigm was subject to academic criticisms. Some noted the “thin politics” of the paradigm (Wade 1992), which failed to “uncover the complex and dynamic internal workings of the state structure by depicting the state as an internally cohesive, unitary actor” (Moon and Prasad 1994, p. 364). Instead, they proceed to unpacking the various components of developmental states, such as “executive leadership, executive-bureaucratic nexus, intra-bureaucratic dynamics, and bureaucratic constituents”. A second flaw of the developmental state literature, according to Moon and Prasad, is its uncritical acceptance of a notion of technical rationality in economic policy, which ignores its inherently political character, and makes unwarranted causal links from state structure to economic performance. They claim that “Even in South Korea, economic policies have been politicized”, as “sectoral policies have been to a great extent subjected to political calculus of legitimacy building, power consolidation, and regime survival” (Moon and Prasad 1994, p. 368). Similar claims on the primacy of politics for understanding the genesis of developmental states are echoed by Leftwich (1995), although he extends the definition of developmental states beyond its classic application to East Asia, englobing states in Southeast Asia, and even Botswana as an African case.

Calls for an analysis of developmental states that is more alert to political rationality are heeded by Kang (2002), who in a comparative case study of industrial policy in South Korea and the Philippines, credits the superior economic success of the former to a balance of power between politicians and business leaders, owing to a high concentration of both political and economic power. As a result, government and business were constrained from acting in growth-preventing ways, as the encompassing nature of business groups, combined with the tight control of rents by government, allowed rents to be dispensed in ways that promoted economic growth.

If Kang (2002) tries to dispel the myth of a depoliticized economic policy, Ben Ross Schneider (1991) meets the first of Moon and Prasad’s challenges through an in-depth analysis of the career incentives of bureaucrats involved in industrial policy in authoritarian Brazil. He makes a distinction between bureaucratic appointments who are técnicos, for whom career aspirations are aligned with the organizational interests of their bureaucracies; politicos, who are primarily guided by political considerations and try to divert resources towards their regions; and political técnicos, técnicos with political contacts who are well-placed to navigate the nexus between the bureaucratic and political worlds. Schneider argues that the appointment of one or another type of bureaucrat to a particular post mattered for the outcomes of industrial policy, as depending on their network of contacts and their career aspirations, they were more or less likely and able to push for successful policies. Schneider (1993) generalizes these arguments on the link between career incentives and bureaucratic performance through a comparison between bureaucratic careers in Brazil, France, Japan, Mexico and the USA.

Schneider’s work is part of a broader political science literature dealing with business-state relations, which differs from the earlier literature on the developmental state in the level of detail with which it studies the institutional arrangements of the state and of business associations, as well as their agency in the political process. The literature finds that the establishment of ‘Weberian’ bureaucracies depends on politicians’ abilities to insulate bureaucrats from external pressures, and on the aforementioned patterns of recruitment and socialization of civil servants (Geddes 1990; 1994; Schneider 1993). Even where the bureaucracy cannot entirely be characterized as Weberian, successful industrial policy might often follow from the establishment of ‘pockets of bureaucratic efficiency’ in agencies responsible for specific issue areas (Geddes 1994; Evans 1997). State organization can also affect the way business is organized (Skocpol 1985; Haggard et al. 1997). The latter matters for the outcomes of industrial policy, as business’ organizational capacity determines its ability to overcome market failures and, in conjunction with pressures from government and the market, lobby for productivity-enhancing policies (Doner 1992; Doner and Schneider 2000). As a consequence, the conditions under which diverse business interests come to act collectively and the specific institutional arrangements that they adopt assume great significance, and scholars investigate the organization of business associations and business-government councils in detail, without losing sight of historically rooted patterns of interaction (Maxfield and Schneider 1997; Schneider 1998; Doner and Schneider 2000; Schneider 2015).These studies also single out the importance of more narrowly political factors for patterns of business-government collaboration, and many of them find that both business and government are more likely to sustain collaboration when they feel threatened (Doner 1992; Maxfield and Schneider 1997; Doner et al. 2005).

Overall, the literature on business-state brings studies on developmental states one step further, broadening the geographical focus and further disaggregating analytical categories. It highlights the doubly technical and political nature of industrial policy – as seen for example in Schneider’s distinction between the different types of bureaucrats active in authoritarian Brazil – and brings to the fore the importance of understanding both the politics and the organizational structure underpinning business-government collaboration. As we will see below, these themes resurface in the more recent literature on political settlements.


  1. Understanding African Political Economies

3.1 The African State

In Evans’ and Kohli’s accounts of state-led development, African states are deemed different – and inferior in terms of developmental success –  to other states in the developing world. The perceived distinctiveness of African states has sparked a voluminous literature, which we cannot hope to summarize comprehensively here. Nonetheless, borrowing from Centeno and Ferraro’s (2013) analytical scheme for understanding state power, most observers would agree with the idea that the majority of African states are weak in the territorial, economic, infrastructural, and symbolic dimensions of state power. Explanations for these deficiencies commonly invoke the concept of ‘neo-patrimonialism’, indicating the presence of an ‘incomplete’ form of modern stateness, or the inter-penetration of modern and traditional forms of domination. Herbst (2014) provides the best-known account of why this might be so. He frames his argument as an exploration of why the factors that led to the emergence of modern states in Europe, captured in Tilly’s (1992) famous dictum that states make war and war makes states, did not present themselves in the same way in Africa. For Tilly, the emergence of a ‘national state’, and the development of ‘rational-legal’ means of administration as its corollary, was spurred by the requirements of mobilizing resources for war-making in Europe. In contrast, Herbst argues that in Africa, the state-building process was (and still is) premised on a fundamentally different logic. Because of Africa’s geography, which makes it difficult to sustain high population densities (with the exception of the Great Lakes region and the Ethiopian highlands), African states were never forced to undergo the same process of consolidation of territorial control that Tilly described for Europe, since “the point of war was to take women, cattle, and slaves” (Ibid. p. 20).

Importantly for Herbst’s argument, and in a distinct manner to other underpopulated regions of the world such as Latin America and Central Asia, the same logic that hampered the consolidation of territorial control by indigenous rulers governed the efforts of European colonizers. Given the imperative for colonial African states to be self-financing, at least in their British, French and Belgian incarnations, the main mission of the state was the control of African labour. The few decades of colonial rule were followed by a post-colonial order where this basic logic continued to hold, perhaps aggravated by the artificial geographies of the polities inherited by African leaders. As such, most African states failed to increase their territorial reach. Moreover, an international system where the principle of state sovereignty became inviolable (a system which was in African heads of state’s interests to maintain) guaranteed that these artificial borders would persist and ensured that the imperative to increase state capacity in preparation for war would not arise. As a result, Africa failed to develop modern, Weberian states, and this failure continues to plague developmental efforts to this day.

Herbst’s arguments are echoed by other scholars, though with differences in emphasis. A literature in economics documents the effects of historical factors on the contemporary development of African countries. Examples include the slave trade (Nunn 2008; Nunn and Wantchekon 2011); the lack of precolonial political centralization (Michalopoulos and Papaioannou 2014); ‘extractive’ colonial institutions (Acemoglu et al. 2001); and even excessive genetic diversity (Ashraf and Galor 2013). On the more qualitative side, Kohli (2004) notes the rudimentary nature of technologies in pre-colonial Africa, as well as the absence of a written language in most of the continent, which he sees as one of the explanations for the lack of larger-scale states. Hyden (1980) attributes Africa’s technological backwardness to its geography, as the dearth of large rivers meant that agriculture was mostly rain-fed. These same geographic conditions also resulted in a peasantry mostly made up of small-scale, quasi-independent units operating in a land-abundant environment. The availability of ‘exit’ options prevented African states from ever ‘capturing’ the peasantry and extracting resources from it, a condition Hyden deems essential for the development of larger, centralized political units.

With regards to the colonial state, although the extent to which it represented a discontinuity from indigenous forms of political organization is a matter of historical controversy, there is a broad agreement that the states created during the colonial period were mostly artificial, and that they affected the political and economic viability of independent African countries (Nugent 2004; Alesina et al. 2011; Michalopoulos and Papaioannou 2016). The imperative to above all control African labour also meant that the colonial state imposed a brutal superstructure of domination, even if by means of a thinly stretched state apparatus, which was simultaneously perceived as alien by the citizenry (Young 1994). Moreover, it was an externally-oriented, ‘gatekeeper’ state (Cooper 2002), which relied on controlling African countries’ relationships with the external world, rather than penetrating the territory (Bayart 2000). For Young and Cooper, this legacy would prove ominous in the post-colonial period, as the African leaders that came to power perpetuated, and in many cases aggravated, many of the negative aspects of colonialism. Finally, as the post-World War II international system came to recognize the inviolability of national sovereignty, these birth defects became very hard to change.

3.2 The Nature of African Politics

Even if African states suffered from a number of birth defects, there still is a need to explain why these have persisted. Although such an overall diagnosis runs the risk of incurring into “an excessive levelling of the African political and economic landscape” (Mkandawire 2000; 2015), few would disagree with the idea that, in general, African countries have less developed states and less prosperous economies than most other countries in the world. Even if at times they have had quite significant growth spells (Jerven 2015), few, if any, have achieved the kind of structural transformation of their economies generally held to be the key to sustained development. Any plausible explanation for Africa’s developmental shortfall must at some point account for why political well-placed actors have failed to change the conditions at the root of these disappointing results.

Many have sought the explanation in theories of how African politics ‘works’. Jackson and Rosberg (1984) highlight the personalistic nature of politics in African countries, where politics does not consist of conflict between well-defined classes or social groups, but of “palace politics” and intrigue among “powerful and wilful men” and associated clans, factions, and political networks, which are “largely indifferent to the interests, concerns, and problems of social strata beyond the political class” (Jackson and Rosberg 1984, p. 424). In this scenario, politicians’ ultimate objective is not to implement any social or economic programmes, but merely to “seek access to the resources and honors of the state” (Ibid. p. 426). This characterization of African politics resonates with other prominent interpretations, such as Bayart’s (1993) “politics of the belly”, or Chabal and Daloz’s (1999) “instrumentalization of disorder”, which depict African politicians as exclusively concerned with the use of state resources for personal ends, as according to the ‘logic’ of neo-patrimonialism.[8]

Joseph (1987), writing on Nigeria, and Van de Walle (2007), further specify the distinctive way in which clientelism operates in Africa. As opposed to the image of a distributive process through which those with access to state resources reward their followers, Van de Walle argues that, like most early states, clientelism in postcolonial Africa has most frequently taken the form of ‘prebendalism’, whereby “an individual is given a public office in order for him/her to gain personal access over state resources (Van de Walle 2007, p. 51). Prebendal systems have very little clientelist redistribution and, as emphasized by Jackson and Rosberg, the political sphere is largely autonomous from societal concerns, except through “the less tangible bonds of ethnic identity” (Van de Walle 2007, p. 55).

Jackson and Rosberg give two reasons for the persistence of personal forms of rule in Africa. The first is an outcome of a perverse equilibrium, where, given their expectation that all other actors in the political system will act according to the dictates of personal rule, no individual actor has an incentive to deviate from these modes of conduct. A second reason relates to the artificiality of the states imposed by colonial powers and inherited by African leaders, which had no roots in indigenous institutions. The essentially alien nature of the colonial state and of the liberal democratic institutions introduced towards the end of colonial rule, combined with the social diversity of most African countries, hindered the formation of the minimal level of consensus required to consolidate rule-bound political institutions.

The link between the colonial period and current patterns of politics is also emphasized by other scholars. For instance, Berman (1998, p. 329) points to the legacy of “bureaucratic authoritarianism, pervasive patron-client relations, and a complex ethnic dialectic of assimilation, fragmentation and competition” left by the colonial powers, which in combination “make the full-scale development of both capitalism and the nation-state difficult, if not unlikely”. These legacies derive from the difficulty encountered by colonial powers of simultaneously controlling their African subjects and extracting an economic surplus (Ibid, p. 315):

“The dilemma, in short, was that the more colonial states pushed the development of the central features of the colonial political economy, using extra-economic coercion to establish the basis of labour and commodity markets, the more African societies were disrupted and the maintenance of law and order threatened.”

This central contradiction led to the strengthening of patron-client ties found in rural institutions, a process which “constrained the willingness and capacity of regimes to promote the capitalist transformation of agricultural production” (Boone 1994, p.110). These colonial modes of governance and exploitation led to a fragmentation of social power in African societies. As African countries gained independence and indigenous elites appropriated the institutional apparatus of the state, the building of ruling coalitions had to confront this structural reality. In most cases, they were unable to overcome the pre-existing fragmentation, largely due to the absence of organizationally strong political parties (Van de Walle 2007) and elites had to rely on the “lowest common denominator” of material resources to bind coalitions together. In the process, the same contradictions that had hindered the economic performance of colonial regimes came to affect their postcolonial successors.

The difficulty of building solid ruling coalitions in independent Africa is picked up by Allen (1995), who places it at the core of his model of postcolonial political trajectories. Like Jackson and Rosberg, Boone, and Berman, he sees clientelism as a device for dealing effectively with decolonisation strategies. However, he does not see clientelism as destiny, as many postcolonial regimes managed to stabilize their power through the establishment of “centralised-bureaucratic” politics, even as others degenerated into systems of “spoils politics”. Although the stability of centralized-bureaucratic polities would be put to the test during the economic crisis of the 1980s, and although Van de Walle 2007 qualifies the extent to which they were able to overcome prebendal forms of clientelism and establish patronage-based political orders, Allen shows that there exist means of establishing relatively stable ruling coalitions in postcolonial Africa. The fact that some of these regimes, most prominently Cote d’Ivoire and Kenya, managed to achieve relatively high growth rates over this period – even if their development models had inherent limitations[9] – also testifies to the possibility of building reasonably developmental regimes in the continent.


  1. The Political Settlements Paradigm

4.1 Rethinking the Role of Governance in Development

Both the developmental state paradigm and mainstream accounts of African politics can lead to a degree of pessimism with regards to the possibilities of achieving transformational growth of the sort experienced in many Asian countries, and a few in Latin America, over the second half of the twentieth century. At the same time as accounts of developmental states highlight their demanding institutional and coalitional requirements, we are told that African states are extremely fragile, while ruling coalitions are devoted exclusively to appropriating the resources of the state, lacking any transformational project whatsoever.

However, a contemporary literature in African political economy questions some of these conclusions, noting the diversity of conditions under which economic development can occur. At the base of this literature is a critique of the concept of ‘neo-patrimonialism’, which has been highly influential in the study of African politics. One of the most prominent elaborations of this critiques comes from Thandika Mkandawire (2001). Reviewing different ideas that have been put forward to explain African states’ developmental shortcomings, he notes that they suffer from an excessive levelling of the African development experience and ignore many instances in which there was robust economic performance in Africa. He argues that the neo-patrimonial school has ‘vilified’ African states, based on comparisons with idealized versions of East Asian states. In particular, the view that African political economies are riddled by excessive rent-seeking is troublesome because by trying to explain everything, the concept of ‘rent-seeking’ loses its analytical value. He questions the literature’s failure to account for the differential impacts of rent-seeking, and neo-patrimonialism more generally, in Asia and in Africa, and argues that instead of thinking whether there are rents in the economy, precedence should be given to identifying who the recipients of those rents are, and whether they have the incentive to invest in new activities.

Moreover, it is not clear whether patrimonial state-society relations are a function of the level of development, or whether there is a distinct African variety of neo-patrimonialism. Thus, argues Mkandawire, a deeper understanding of the political conditions for development in Africa requires a more in-depth look at the policies that sustained growth in Africa and elsewhere, beyond generic, catch-all concepts such as ‘rent-seeking’ or ‘neo-patrimonialism’. This critique is extended in Mkandawire (2015), where he goes through a number of explanations for distinct economic and political phenomena that invoke ‘the logic of neo-patrimonialism’, but which he finds wanting. We are led to conclude that such explanations are unfalsifiable, and, by referring to an ubiquitous and all-embracing concept that seemingly explains even contradictory phenomena, uninformative.

Other scholars follow the wake of Mkandawire’s critique in questioning the utility of the concept of ‘neo-patrimonialism’. Early on, Theobald (1982) and Crook (1989) had noted the catch-all nature of the term, arguing that its application in so many settings required too much conceptual stretching, leading to a loss of analytical utility. Pitcher, Moran and Johnston (2009) follow Theobald in criticizing the term for being all-embracing, as it has been used to denote a set of social relations at various levels; rent-seeking behaviour and personalist patterns of authority; an economic logic that blurs the distinction between public service and private gain; and a regime type. They argue that this type of account mistakenly conflates types of authority with types of regime. In so doing, they ignore the possibility that patrimonial forms of authority can lead to different types of regime. A discussion on the emergence of democracy in a patrimonial context in Botswana illustrates this point. Similar ideas on the malleability of neo-patrimonialism are present in Bach (2012), who distinguishes between ‘regulated’ forms of neo-patrimonialism, where neo-patrimonialism is only present in some areas within the state, and ‘predatory’ forms of neo-patrimonialism, where the entire state apparatus is subject to a neo-patrimonial logic. It is only in Africa that the latter understanding has come to dominate theoretical debates; in other contexts, neo-patrimonialism has not been deemed incompatible with public-regarding policies. We return to ideas on types of neo-patrimonialism when discussing the work of Tim Kelsall.

Mkandawire’s argument for changing the way we think about rents, from whether they are distributed to how they are distributed, is picked up by Douglass North, John Wallis and Barry Weingast’s (2009) ambitious theoretical work attempting to rethink the relationship between politics and economic development. Instead of presuming that rational-legal forms of domination precede the advent of sustained economic growth, these accounts note that historically, the transformation towards a capitalist system started in political settings far removed from liberal democracy, only eventually converging towards the system of governments prevalent in today’s developed countries. Their theory posits that human societies can move between three different ‘states’: foraging orders, constituted by dispersed hunter-gatherer populations; limited access societies (LAOs), where personal relations form the basis of social organization; and open-access orders, in which there are wide areas in society where interactions take place on an impersonal basis, and social order is sustained through political and economic competition. Importantly, the seeds of open access orders are sown in LAOs, as they achieved growth in a context where rent allocation followed political imperatives. This observation paves the way for new ways of thinking where imperfect forms of governance – at least when judged from the dominant ‘good governance’ paradigm – can still lead to economic growth.

4.2 Mushtaq Khan’s Theory of Political Settlements

The implication of North and co-authors’ work is that any account of how developing countries can achieve economic transformation must investigate how rent allocations in personalized settings can be growth-enhancing. This is the line of research followed by Mushtaq Khan, the most influential theorists of the political settlements school. The concept of ‘political settlement’ is introduced in a number of Khan’s writings, of which Khan (2010) contains the most extensive exposition and is generally the most cited. Khan posits that in any society, groups differ in their capacity to engage in and win conflicts. He refers to this capacity as a group’s ‘holding power’, and it is a function not only of wealth and income, but also of “the political ability to organize, the numbers of people that can be mobilized, and perceptions of legitimacy” (Khan 2010, p. 6), as well as differences in groups’ determination and their strategies of mobilization. The distribution of power between competing groups is called the ‘political settlement’. In the absence of a compromise between the various groups, societies are likely to be in a state of violence or anarchy. To reach such a compromise, it is necessary to erect economically and politically sustainable institutions capable of distributing rents[10] consistently with the distribution of power in society (ie. with the political settlement). Conversely, conflict tends to emerge when particular groups perceive these institutions as being unreflective of their true power.

Most developing countries differ from developed countries in not having their distribution of incomes based just on formal institutions and rights, but on a combination of formal and informal institutions, such as patron-client allocative rules. For Khan, this feature of developing countries owes not to the cultural differences alluded to by many writers in the neo-patrimonialism school, but by the structure of economies where the formal, capitalist sector is still small, and the majority of powerful groups do not have to conduct complex business transactions, so they have little interest in pushing for informal rule-enforcement (Khan 2010; Khan, Andreoni and Roy 2016).[11] An institutional arrangement – formal or informal – will be growth-enhancing if the process governing the allocation of rents incentivizes actors to invest in productive activities and in the adoption of new technologies. A particularly important function of rent-allocation for the adoption of new technologies consists of inducing firm owners, managers and supervisors to exert effort in learning-by-doing (Khan 2013b). In this way, industrial policy – which Khan actually refers to as ‘technology policy’ and essentially conceives of as an incremental change in the allocation of rents by the state – should compel them to develop the requisite organizational capabilities for competitiveness.

Khan conceives of the relationship between political settlements and industrial policy in terms of a ‘growth-stability trade-off’. If industrial policy consists of the introduction of novel institutional arrangements, then it is natural to expect these arrangements to face opposition from groups with vested interests in the status quo. Opposition from these groups can lead to a decline in political stability. Khan posits that every society has a minimum required level of stability before there is a breakdown in order. At the same time, institutional arrangements also require a minimum level of economic performance in order to be sustainable. Different institutional arrangements will map onto different points of the growth-stability space, and, in general, more ambitious strategies (which may have greater growth potential) are more likely to face political opposition, leading to a decline in stability. This growth-stability trade-off will differ according to the nature of the political settlement. In some political settlements, growth comes at a lower stability cost than in others, where there is a greater variety of interests opposing pro-growth institutional arrangements.

Although there are many ways of characterizing a political settlement, Khan identifies two central properties of clientelist political settlements (the modal type of political settlement in the developing world) that determines their growth potential. The first of them is the distribution of power in the ruling coalition. The horizontal dispersion of power is the relative power of groups excluded from the ruling coalition, while the vertical distribution of power is the relative power of upper-level faction of the ruling coalition in comparison with lower-level factions. A greater horizontal concentration of power allows a ruling coalition to act with a longer time horizon, since it is less likely to be displaced, while a greater vertical concentration of power increases the ruling coalition’s power to enforce and implement policies.

The second relevant characteristic of a political settlement is the power and technological-entrepreneurial capabilities of capitalists, in combination with their political power. This also forms a 2×2 matrix of combinations of high and low capabilities, and high and low political power. For Khan, a situation in which capitalists’ capabilities are high and their political power is low holds the most potential, since it allows for an easier disciplining of learning rents. Conversely, if capitalists’ political power is low but their capabilities also are, ruling coalitions may have fewer policy options.

Besides the contemporaneous growth implications of a given type of political settlement, the growth-stability trade-off has a dynamic aspect, as its shape is also a function of the strategy being pursued by the ruling coalition. At any point in time, strategies based on confrontation with powerful groups will lead to more adverse growth-stability tradeoffs, while strategies of cooptation might be easier in that respect. However, though the latter may be easier in the short run, they can have negative dynamic impacts, as failure to confront powerful vested interests at the early stages of development may come to haunt development efforts further along the line. In Khan’s representation, this corresponds to a steepening of the curve representing the growth-stability trade-off. This point underscores the idea that the distribution of power in society is not exogenous, but “changes through social mobilizations, political activities creating new political organizations and coalitions, and the process of institutional change itself.” (Khan 2010, p. 37). Thus, the politics of industrialization is immersed in the broader political struggles taking place in a given society. These observations offer us some starting points for thinking about the politics of industrialization in more strategic ways. We return to this point in the conclusion, where we synthesize the evidence on political strategies for industrialization.

4.3 The Political Survival of Ruling Elites

Mushtaq Khan’s theoretical work has inspired a series of publications that build on his original theory and use it to explain the growth performance of particular sectors and countries. Publications on industrial policy in this line of research include work by Whitfield, Therkildsen, Buur and Kjaer (2015); Behuria (2015; 2016); and Gray (2013; 2015 2018), among a number of other working papers.[12] The framework has also been employed to deal with issues such as housing (Croese 2017), education (Abdulai and Hickey 2016; Languille 2016), public service provision (Hirvi and Whitfield 2015), and natural resource governance (Hickey and Izama 2017; Macuane, Buur and Monjane 2017). Here, we focus on the work on industrial policy, in view of this paper’s theme.

The political settlements literature mostly uses Khan’s framework to study more localized issues, as opposed to Khan’s greater concern with theorization and analysis of political settlements at the societal level. For this purpose, Whitfield and Buur (2014) and Whitfield et al. (2015) develop what they call a “political survival of ruling elites approach”. They posit that rulers’ policy choices towards a particular sector, their ability to implement these choices, and the manner in which they interact with business are driven by the imperative of buttressing their position in power. Successful industrial policy within a given sector requires three key conditions. The first is that ruling elites, or at least a faction among them, have an interest in developing the sector, and that they have sufficient power to resist claims from parties whose interests are opposed to development of that sector. This allows them to insulate bureaucrats from pressures. The authors refer to this as the ‘mutual interests’ condition. The second requirement is that bureaucrats have adequate technical expertise to deal with a particular industry, referred to as the ‘pockets of efficiency’ condition. Finally, bureaucrats must have a working relationship with business, although this relationship cannot be too close, to avoid the risk of capture. This is the ´learning for productivity’ condition, which bears a close resemblance to Peter Evans’ concept of ‘embedded autonomy’. The case studies collected in Whitfield et al. (2015) use this framework to study the differential rates of success of various economic sectors in different African countries. Similarly, Behuria (2015) uses the framework to explain why the RPF regime in Rwanda used a military enterprise to develop the pyrethrum sector, while the same regime engaged in pro-market reforms in the mining sector.

Despite the authors’ general adherence to Mushtaq Khan’s conceptual framework, the first component of this approach, the “mutual interests” condition, represents a significant, though underemphasized, difference between the “political survival of ruling elites” approach and Khan’s political settlements framework. Here they insert a predictive element that is absent in Khan’s work, which is generally agnostic on the question of when and why ruling elites (or “ruling coalitions”, the term most often employed by Khan) choose to adopt particular policies. To put it succinctly, we can characterize Khan’s political settlements framework as a theory of policy enforcement, while the “political survival of ruling elites” approach (and similar work by the likes of Behuria, Gray and others) is both a theory of policy genesis and enforcement. Although this difference is never mentioned explicitly, it is telling that in a review article aiming to clearly delineate the scope of political settlements analysis, Khan (2017, p. 9) remarks:

Particular changes in policies or institutions are the outcomes of the agency of governments or other organizations, and are therefore not predictable and certainly do not mechanically reflect the distribution of power in a society (our emphasis). Governments and organizations exercising agency represent particular constituencies, are influenced by their own ideologies, or international pressures and other contingent factors. Sometimes they make huge mistakes in assessing the reaction to their activities and end up worse off. A government does not typically represent the full balance of interests of all of the powerful organizations in a society. A variety of policies and institutions can therefore emerge at different times. The much more important dynamic question is whether the institutions and policies that emerge in this way will survive and be sustainable, and how they will be implemented, given the distribution of power in that society. The political settlements framework provides a way of evaluating the sustainability of the institutions and policies that emerge, but it does not of course predict the precise institutions and policies that actually emerge.

As will be discussed in this paper’s conclusion, theorizing about policy genesis is a much more difficult endeavour than thinking about the conditions for policy enforcement, which brings up a host complex theoretical issues.

4.4 Other Varieties of Political Settlements Analysis

Although the accounts roughly grouped under the “political survival of ruling elites” approach are some of the most prominent intellectual offspring of Mushtaq Khan’s work, political settlements analysis is by no means confined to them. Simplifying a more varied panorama, we can broadly identify two additional literature strands that use some variant of political settlements analysis. Firstly, there is the work of authors associated with ODI’s Africa Power and Politics Programme, such as Tim Kelsall (2011; 2013) and David Booth (2012; Booth and Golooba-Mutebi 2012; Booth and Cammack 2013) which, despite being close to the spirit of political settlements analysis, do not directly employ the concept[13]. Nonetheless, they helped spearhead the movement towards the analysis of the relationship between growth and different forms of formal and informal power in distinctively non-Weberian settings. Their accounts emphasize leaders’ time horizons and the degree to which their control of rents is centralized as potential source of developmental leadership (Kelsall 2011; Booth and Golooba-Mutebi 2012). Additionally, Booth highlights the collective action dimension of the politics of development, as opposed to what he claims is the principal-agent conception dominant among development policy institutions, linked to concepts such as voice, empowerment and accountability (Booth 2012; Booth and Cammack 2013).

The other types of political settlements analysis encompass a wide range of policy-oriented publications that adopt varying conceptions of what a political settlement is. Pospisil and Rocha Menocal (2017), for example, take Di John and Putzel’s (2009) definition of political settlements as “common understanding or agreements, usually among elites, about how resources are distributed and power is exercised”. Parks and Cole (2010, p. 6) take the key elements of a political settlement to be “powerful actors, operating in the pursuit of their interests, leading to the establishment or reshaping of institutions to sustain the political settlement, including formal state institutions and informal arrangements.” This diversity of definition leads Laws (2012) to survey the multiple meanings that have been ascribed to the term “political settlements”, coming up with four clusters of themes:

  1. Political settlements describe state-society relations.
  2. Political settlements are the product of elite negotiations, which shape the character of formal and informal institutions governing political and economic behaviour (the ‘rules of the game’).
  3. Political settlements reflect the overall balance of power in society.
  4. Political settlements are the means through which societies overcome violence and political instability.

It is clear that these definitions of ‘political settlement’ are much more wide-ranging than Khan’s definition of a political settlement as simply the “distribution of organizational power” (Khan 2017, p.2). In fact, similarities in terminology aside, it does not make sense to think of one political settlements framework, but of a variety of them. This allows Bell and Pospisil (2017), for example, to speak of a “formalised political unsettlement”, something which would unconceivable under Khan’s definition of the term.  These conceptual differences seem to stem from an interpretation of the term “political settlement” that is closer to everyday language, as opposed to Khan’s more precise and technical definition. Although many of these publications cite Khan and draw on his work, it is clear that their understanding of what constitutes a “political settlement” is different.[14] We deal with the question of how best to define “political settlements” in the conclusion.

The research adopting these broader definitions of “political settlement” can be divided roughly into two streams. The first seeks to expand the scope of political settlements analysis and has come to deal with a variety of issues of concern to the international development community, such as state fragility, identity and nation-building, political representation, gender inequality, and inclusivity more generally[15]. Here we cannot discuss this research in detail, given our focus on industrial policy, but we note that these accounts, while couched in the language of political settlements and elite pacts, venture into well-trodden social scientific territory, drawing from the voluminous literature that informs this variety of topics.

Rather than applying the political settlements framework to different topics, the second stream of research tries to develop the concept of “political settlements” itself. Based on his survey, Laws (2012, p. 21), proposes a definition of “political settlements” that conceives of them as “on-going and adaptable political processes that include specific one-off events and agreements”, which shape the nature and performance of institutions, and are played at two levels (Khan’s horizontal and vertical dimensions). Laws and Leftwich (2014) further highlight the importance of the international sphere for political settlements. Similarly, Kelsall (2016) inspired by accounts in the new wave of political thinking about development policy such as Levy (2014), puts forward three diagnostic questions meant to guide political settlements analysis:

  1. How inclusive/exclusive is the political settlement? Do the majority of elites accept the settlement, or do most of them remain willing to use violence to achieve their aims?
  2. What motivates elite to accept the political settlement? Is it because they are coordinated around a common purpose, or because they are given a share of spoils?
  3. By what norms is the bureaucracy governed? Is it by impersonal rule-following and meritocracy? Or by patron-client relations and nepotism?

The answers to these questions place a country in a three-dimensional political settlements space, which allows us to assess is its developmental potential, and the potential inroads for development partners.

Other authors venture into more theoretical territory. Leftwich (2010) argues for the importance of human agency and coalition-building for developmental outcomes, as opposed to what he sees as a bias towards structuralist explanations in the literature. This point is echoed by Hickey (2013) and by Hudson and Leftwich (2014), who also stress the role of ideas, ideologies, and contingency in politics. Hickey argues that political settlements analysis should be permeable to a greater variety of theoretical influences, ranging from ‘mainstream’ political science, to institutional economics, to critical theory. Meanwhile, Hudson and Leftwich make the constructivist argue that, contrary to analyses that assume rational, self-interested actors, “interests and ideas are part of the same story, and that self-interest is not the obvious and consistent force it is always assumed to be.” (Hudson and Leftwich 2014, p. 108). They also put forward the sketch of a framework for applying political economy analysis in practice that pays attention to power, ideas, structure and agency. Finally, the work of Booth (2014; Booth et al. 2015) calls for consideration of international factors in political economy analysis, as well as the PDIA (Problem Driven Iterative Adaptation) framework developed by Andrews, Pritchett and Woolcock (2017). While these contributions make a strong case for greater intellectual openness in the political economy research agenda, in the next section we discuss how feasible it might be to incorporate all these considerations when thinking more narrowly about development strategies and policy.


  1. Discussion

5.1 Appraising the Literature – A Conceptual Approach

The history of research on the politics of late industrialization can be characterized by successive cycles of rise, spread, and fall of theoretical frameworks. These frameworks have frequently been centred on a few core concepts carrying much of the framework’s analytical weight, and which serve the function of “providing a new thought-instrument or conceptual tool to help particular people understand and define, and thus begin to deal with, certain problems” (Geuss 2008), in this case economic development. While these concepts are associated with innovations in the study and practice of industrial policy, and as such provide “a new thought-instrument or conceptual tool to help particular people understand and define, and thus begin to deal with, certain problems” (Ibid.), their proliferation has often been accompanied by a loss of terminological clarity, which has hampered the consolidation of a cumulative research programme. In this section, rather than individually assessing the merits and limitations of each strand of research – a task that would require a much more extensive treatment – we adopt a conceptual lens to briefly comment on the evolution of the research strands described above. We argue that terminological choices, even when accompanied by sound scholarly analysis, can have a lasting impact on the diffusion of policy paradigms, and that researchers would be well-advised to pay greater attention to the potential unintended side-effects of their choices.

There is a long tradition in thinking about the role of concept formation in social science research. In political science, its main exponent has been Giovanni Sartori, whose 1970 American Political Science Review article on “Concept Misformation in Comparative Politics” was a seminal contribution to the field.[16] In this and subsequent contributions, Sartori condemns the lack of definitional clarity in much of social science research, which leads to ‘conceptual stretching’ and thus to a loss of connotative precision. He advocates a systematic way of thinking about concept formation, based on the trade-off between a concept’s “intension” (ie. the number of attributes that define it, and thus the meaning it calls forth) and its “extension” (the range of cases to which it can be appropriately applied). The specific combination of intension and extension allows one to place concepts along a “ladder of abstraction”, ranging from richer, more specific concepts, to the most general and inclusive.

This kind of thinking on concept formation is well-illustrated by an examination of ‘the developmental state’, the most prominent concept in the study of the politics of development.[17] Chalmers Johnson originally intended it to capture how the post-war Japanese state provided a distinctive development model, combining elements of both capitalism and socialism. He saw this as a way of overcoming the then-dominant dichotomy between market- and state-led development strategies.[18] As such, Johnson’s concept of the ‘developmental state’ can be considered a ‘low-level category’ in Sartori’s ladder of abstraction. The concept’s subsequent application to other high-growth East Asian political economy required an increase in its generality, as an account of Japanese economic institutions, “in their full complexity” (Johnson 1982, p. 7), could not do justice their historical trajectories. Nonetheless, if we accept Onis’ quote in section 2.1 as a comprehensive list of the properties of developmental states, we might still consider that the concept embodied a useful trade-off between intension and extension, since it adequately captured the main characteristics of these East Asian states.

By the time of the publication of Evans’ landmark book on Embedded Autonomy, one could argue that the concept had become too general. The continuum from developmental to predatory states, even if considered purely as a heuristic device (as in Evans 1987), seems far too simplistic as a model of variations in state characteristics. While Zaire/DRC is an archetypal predatory state, would it make sense to characterize other, more normal, but clearly ‘neo-patrimonial’ states in Africa as being closer developmental states? Or does it make more sense to follow Kohli and think of different dimensions along which states can vary, creating discrete categories for each (eg. cohesive vs. fragmented, capitalist vs. multiclass)? The study of democracy is a useful analogy here. Few would endeavour to compare the degree to which two eminently authoritarian regimes, say Nazi Germany and Maoist China, were more or less democratic. In fact, as argued by Sartori (1970):

“Differences in degree obtain only after having established that two or more objects have the same attributes or properties, ie., belong to the same species. Indeed, it is only within the same class that we are entitled – and indeed required – to ask which object has more or less of an attribute or property.”

Furthermore, as noted by Kohli, one of the key limitations of the concept of ‘developmental state’ is that it in its definition it incorporates what it is supposed to explain, ie. the kinds of states that are better-suited to promoting development. Again, this relates to one of Sartori’s key admonishments to always seek ‘minimal definitions’ that exclude accompanying or varying properties, which should be left as the subject of investigation[19]. In practice, many scholars, including Peter Evans, do engage with these more interesting questions, even while employing the concept of ‘developmental state’. However, the problem appears when, as a result of its own success, the concept gets vulgarized in the hands of less sophisticated authors. Oftentimes, the latter propagate the idea that industrial development can only be promoted by mimicking South Korea, leading to overambitious strategies that can become very dangerous in the very dissimilar conditions faced by most states in the developing world.

As the discussion above shows, the diffusion of the concept of ‘developmental state’ constitutes an exemplary case of ‘conceptual stretching’, whereby a vaguely-defined concept is applied in ambiguous ways (Sartori 1970). This is in large part attributable to a lack of attention to conceptual clarity and consistency, as  the term often comes to be applied according to its lexical definition (ie. as in everyday language), as opposed to the ‘stipulative’ definition (ie. the strict definition specified in a social scientific account, which might not coincide completely with the lexical definition). This often results in a situation of ‘collective ambiguity’, where “each scholar ascribes his own meanings to his key terms” (Sartori 2009b, p. 111). In particular, the notion that a ‘developmental state’ is a state that strives for development is problematic. Besides the problem of ascribing intentionality to a state, as opposed to the government at its helm, these understandings ignore the other characteristics associated with East Asian developmental states, succinctly grouped by Weiss’ (2000) as their “organizational arrangements” and their “institutional links with organized economic actors”. A selective adoption of elements of developmental stateness, perhaps influenced by a ‘high-modernist’ vision (Scott 1998)[20] can then result in disappointing outcomes, at best, and in worst case scenarios lead to counterproductive forms of government intervention.

The concept of ‘neo-patrimonialism’ suffers from a similar kind of conceptual fuzziness. While it has not achieved the same imprint on political imaginations, which is easily attributable to its discouraging overtones, it arguably impacts donors, journalists, and foreign policymakers through its continued influence in academia (Mkandawire 2015). Here we do not delve into the definitional issues surrounding the term, since they have been satisfactorily dealt with elsewhere[21], but instead choose to discuss the concept of ‘political settlements’, which has had a strong impact on development policy and research, but is yet to receive the same degree of scrutiny.

The concept of ‘political settlement’ suffers from many of the same problems as that of ‘developmental state’, as its diffusion led to conceptual stretching and collective ambiguity as undesirable side-effects. However, unlike ‘developmental state’, the term ‘political settlement’ predates its most prominent elaboration in the work of Khan, and can easily be employed in common lexicon.[22] As discussed in texts such as Laws (2012) trying to make sense of a cacophony of definitions, the term ‘political settlements’ lends itself to conflation with peace agreements or elite bargains, which in fact would be its most natural interpretation according in everyday language. In spite of that, those opposing such an interpretation of the term tend to use it as a shorthand for ‘social structure’, with emphasis on the role of elites. The discussion then hinges on the best way that this social structure can be dissected to understand what characteristics make it better-suited to achieving various development goals. However, as noted by Van de Walle (2016, p.170), the message of these accounts often (but by no means always) degenerates into “the banal claim that economic outcomes are largely determined by the interplay of social forces”, and explanations lack “a more explicit description of causal mechanisms, and a way to measure the settlement independently of outcomes”, making it “largely non-falsifiable”.

Khan differs from other authors in giving a precise stipulative definition of ‘the political settlement’ (ie. the distribution of holding power), although it resembles other definitions in being difficult to operationalize in empirical research. Nonetheless, his (version of the) political settlements framework, particularly in its 2017 restatement, has the virtues of being clear, parsimonious, and internally consistent. The separation between the drivers of policy and institutional adoption and the drivers of their enforcement resembles the recent advances in the study of democratization, in that the transition to a certain state of affairs is to be understood separately from its probability of survival (Przeworski et al. 2000). Moreover, the political settlement, much like the ‘economic infrastructure’ in traditional Marxist theory, acts as a centripetal force pulling the de facto operation of institutions and policies closer towards being more reflective of the distribution of holding power in society. However, the framework is not as deterministic as (vulgar) Marxist theory, as it accepts that the political settlement is only one of many influences on the institutional structure. Political settlements generate a tendency towards an equilibrium that, like equilibria in many economic models, can rarely be observed in practice, since the speed of convergence towards the equilibrium is usually lower than the speed to which the determinants of that equilibrium change.

Perhaps the foremost merit of Khan’s framework when compared to other competing frameworks is its intentionally limited scope. Although earlier texts might at times suggest otherwise, he does not claim his framework fully explains differences in developmental outcomes, but only puts it forward as one of many possible ways of slicing up social reality to explain different development outcomes. Key in this respect is the abandonment of claims on when ‘ruling elites’ or any other group with decision-making power will adopt certain policies or act in certain ways, which helps sidestep problematic structure-agency issues (Leftwich 2010; Hickey 2013; Hudson and Leftwich 2014).[23] In fact, the application of certain policy ideas to unpropitious contexts is a common reason for failure in politics[24]. In contrast, research agendas that aim to develop all-encompassing theories of human action might set their ambitions too high, or they might become too complex to be applied in any meaningful form in political praxis or policy analysis. For instance, few would challenge the notion that power and ideas matter in politics, but that does not necessarily imply that we need to get bogged down in theories of power, or to develop an explanation for how ideas matter in politics, in order to understand how formal institutions and policies get transformed by being introduced into a given context. Such practices may easily incur into conceptual stretching, as a given conceptual framework extends into other fields of inquiry and core terms may lose their analytical distinctiveness. At the same time, research in these other fields starts adopting the vocabulary of a given framework, mistakenly branding it as a whole new area of research, and hindering the dialogue with previous research on the same topic.

The counterpart of the proliferation of concepts such as ‘developmental states’, ‘neo-patrimonialism’, and ‘political settlements’ is the evanescent nature of the concepts developed in otherwise impactful scholarly publications, such as Kang’s (2002) “mutual hostages” or Kohli’s (2004) “cohesive-capitalist states”.  When they get traction, these conceptual innovations often “stick, escape our control and become part of reality itself”, leading to a “change (in) the “real” situation with which one is confronted so that other, unforeseen problems emerge” (Geuss 2008). At the same time, the tendency towards conceptual stretching and concepts’ undisciplined application, which frequently accompany their diffusion, risks becoming “a hindrance to understanding”(Hirschman 1970), especially in works that merely seek to fit empirical cases into their preferred theoretical framework. The reasons behind the popularization of some terms and theoretical frameworks over others are usually historically-specific and, as discussed by Geuss, can be ascribed to a mixture of analytical or cognitive deficiencies, their attractiveness to a sufficient number of people, and contingent historical events. Nonetheless, in shaping the way academics, politicians, and policymakers think about development strategies, these industrial policy paradigms are certainly impactful.

However, when they do not take hold, concepts remain meaningful within individual publications, but fail to contribute to a systematization of social scientific knowledge. This leads to a tension between the formulation of nuanced and sophisticated empirical case studies and aspirations of a cumulative research agenda. Irrespective of whether one finds these aspirations realistic, it still is important to understand how the different contributions to the literature on the politics of industrial policy fit together, and what lessons can be distilled from an overview of different research strands. We feel that such an exercise is yet to be satisfactorily conducted, so we attempt to do it in the next section.


  1. The Role of Organizations in Industrial Policy

Differences in terminology, research methodology, or scope of analysis can often obscure the substantial overlap that exists between the various contributions to the literature on the politics of industrial policy. That is not to say that a consensus exists; nonetheless, the survey of the academic literature conducted in this paper suggests that there do exist some shared insights on the nature of the political economy challenges that need to be dealt with by development strategists. Here we argue that one of the keys for understanding the challenges of industrial policy and for finding strategies for overcoming them is understanding the role of organizations in a country’s political economy.

6.1 The Politics and Management of Organizations

The purpose of industrial policy is to improve firms’ capabilities (Dercon et al. 2018). As highlighted by most of the literature on the political economy of industrial policy – most clearly in the “political survival of ruling elites” approach (Whitfield et al. 2015) – policies to bring this outcome about require interactions between political elites, state bureaucrats and firms. The degree of involvement of each might of course vary from one case to another, and one could even argue that the form, degree of interaction and relative strength of these three types of organization broadly delineates the development model followed by any particular country.[25]

This is not meant to imply that all political economy analysis can be reduced to study of organizations. Individuals come together and interact in many other types of groups, including factions, networks, families, patronage ties, or loosely-organized social movements, to name a few. These all matter for understanding the distribution of power in society. Still, when dealing with economic policy, organizations must inescapably be the object of study, since most economic activity is conducted by economic organizations (firms); public policies are implemented by state organizations (public bureaucracies); and the formulation of economic strategy is the responsibility of political organizations, even though political power is often exercised by less institutionalized groupings of individuals. It follows that, when thinking about industrial policy, three types of organizations are the main players: firms, public bureaucracies, and political organizations.

6.2 Political and Technical Characteristics of Organizations

Each organization can be described in terms of its political (or coalitional) and its technical characteristics. This is well-captured by Schneider’s (1991) discussion of the distinction between politicos, técnicos, and political técnicos in the bureaucracy of authoritarian Brazil (See section 2.3). As heads of public organizations, these individuals were able to exert both political and technical functions. By ‘political’ we refer to all activities related to the interplay of interests between organizations and between individuals leading those organizations, including the formation of coalitions with actors belonging to the political and economic spheres, and all other forms of bargaining and jockeying for position normally associated with politics, often with negative overtones. Technical functions are those directly related to an organization’s capacity of delivering on its objectives, and include the quality of management practices, the cohesion and durability of organizational structures (Huntington’s 1968 proverbial ‘institutionalization’), and the skill and expertise of those working for it.[26]

This distinction between organizations’ political and technical dimensions, though seldom highlighted, appears implicitly in many other contributions to the literature, such as:

  • Crook’s (1989) distinction between Houphouet-Boigny’s patrimonial strategies of domination and the persistence of relatively strong organizational norms in the Ivoirian bureaucracy. This allowed it to record much higher economic growth rates than other African countries, despite following a broadly similar development strategy;
  • Kohli’s (2004) discussion of how the transition from Syngman Rhee’s to General Park’s regime allowed the latent qualities of the South Korean bureaucracy – itself largely a product of the country’s long tradition of bureaucratic rule – to flourish and be harnessed for developmental objectives;
  • Khan’s (2010) typology of capitalists’ technological-entrepreneurial capabilities and their political power;
  • The distinction between “mutual interests” and “pockets of efficiency” in the “political survival of survival of ruling elites” approach (Whitfield et al. 2015).

Besides the political connections of firm owners, politicians, and top bureaucrats, the political role of organizations is also manifested through more sociological channels. The original literature on the developmental state highlighted the social networks and patterns of socialization that enabled East Asian states to achieve the proverbial ‘embedded autonomy’. Similarly, Schneider (1991; 1993) shows how career incentives and patterns of bureaucratic circulation matter for bureaucratic outcomes. From a different perspective, ‘instrumental’ Marxist theorists of the state such as Miliband (1969) have argued that social ties between individuals at the helm of the state and elites from other spheres enable even a seemingly class-neutral entity such as the liberal-democratic state to favour some class interests over others. Although the concept of ‘political dimensions’ of organizations being proposed here is related to the concept of ‘holding power’, it includes not only the relative strength of organizations in a hypothetical conflict situation, but also the web of interests that ties members of these organizations together (or pull them apart), resulting in patterns of cooperation and conflict.

But despite seeming fairly evident, the distinction between technical and political aspects is often lost in many accounts that overemphasize the importance of politics to the detriment of organizational strength. The end result are voluntaristic frameworks that assume political will is a sufficient condition for successful industrial policy, disconsidering its often substantial technical requirements. In ignoring the technical and organizational requirements of industrial policy, there is a risk of incurring in “premature load bearing” (Andrews, Pritchett and Woolcock 2017) whereby the assignment of a task that is beyond an organization’s capabilities results in a loss of capability.

Among the examples of accounts distinguishing between organizations’ political and technical dimensions given above, there was no mention of political organizations such as legislatures, electoral organizations, or political parties. At first glance, one might be tempted to consider them to have a purely political dimension, almost by definition (ie. mattering only insofar as they are vehicles political bargaining, the projection of power, the formation of political ties etc). However, at least since the work of Samuel Huntington (1968), the degree of institutionalization of political institutions has been identified as essential to governments’ ability to maintain social order, exercise their power, and thereby effect desired policies. Political parties are perhaps the key organizations in the political realm due to their role in aggregating political preferences, elaborating public policies, and structuring political competition. With regards to thinking about economic performance, Bizzarro et al. (2018) find a positive statistical relationship between the strength of parties’ internal organization (defined as their degree of unity, centralization, stability, and organizational complexity, as well as their ties to long-standing constituencies) and growth in a panel of countries. They posit that these results could be driven by a number of advantages enjoyed by strong parties, including:

  • Better decision-making processes, both internally and in response to popular demands;
  • Greater accountability between party leaders and party members;
  • Longer time-horizons in policymaking;
  • An enhanced capacity to solve coordination problems.

These in turn lead to:

  • Reduced incentives to engage in predatory economic policies;
  • A greater priority given to productivity-enhancing public goods and services benefitting the wider population;
  • A higher capacity of reaching authoritative decisions on contested matters of public policy;
  • Greater power of implementation once policies have been decided.

Besides their association with higher growth rates, it is straightforward to see how these characteristics could lead to better industrial policy outcomes. Importantly, these factors are posited to act in similar ways in both authoritarian and democratic settings, attesting to the general importance of strong organizations, irrespective of regime type.

6.3 Comparing Asian and African Corruption

Many accounts have pointed out the fact that although corruption is often blamed for the disappointing postcolonial economic performance of African countries, corruption was (and is) also widespread in Asian countries seen as exemplars of developmental success. However, such a view can be deceptive because it obscures important differences in the nature of corruption across the two regions. Generally speaking – and keeping in mind the perils of generalizing across large and diverse world regions – while corruption in East and Southeast Asia has mostly taken the form of cronyism, countries in Sub-Saharan Africa have been afflicted by prebendalism (Joseph 1987; Van de Walle 2007). Cronyism involves the enmeshment of the public and private interests in the top layers of organizations in ways that are often detrimental to the public interest. Meanwhile, prebendalism involves the circumventing of the organization’s mission for one’s private benefit. While prebendalism at the top of organizational hierarchies can resemble cronyism, once an organization is pervaded by prebendalism at multiple levels, its capacity to function in any meaningful way can be seriously put into question. In terms of the distinction being illustrated here, it makes sense to think of cronyism as a form of corruption pertaining to the political dimension of organizations, while prebendalism hinders their technical operation.

Studies have attested to the comparative weakness of African organisations across different realms, including firms (Lemos and Scur 2015) and political parties (Bizzarro et al. 2018). While a similar comparative survey on the strength of internal organization – as opposed to capability or performance – has not been conducted within African bureaucracies[27], existing indicators of institutional quality, together with qualitative and anecdotal evidence, lend reason to believe that they are also weak in that dimension. A similar kind of weakness also seems to afflict African business associations (Brautigam, Rakner and Taylor 2002), with all attendant side consequences in terms of their capacity to negotiate with the state for productivity-enhancing policies[28]. The prevalence of prebendalism is clear in Kohli’s (2004) study of Nigeria, where he discusses how these extremely dysfunctional forms of corruption were widespread across the institutional landscape, including political, bureaucratic, business, and even military organisations.  Of course, this is not to say that prebendalism is ubiquitous, nor that it is an exclusively African characteristic, or that the shortfall in organizational strength in African countries can be attributed solely to it. In fact, lack of technical and managerial experience is probably the main culprit. Nonetheless, it is important to keep in mind these differing forms of corruption when dismissing corruption as an explanatory variable.

The contrast between African and Asian types of corruption is perhaps most strikingly described in Peter Lewis’ (2007) comparative study of Nigeria and Indonesia, itself not a beacon of good governance. He credits Indonesia’s superior economic performance partly to the greater capacity of the organizations inhabiing its political economy, which allowed elites to reconcile their interests and coalesce around a coherent set of political and economic strategies. But the differences extend beyond the organizational realm, as political dynamics also played a fundamental role. Lewis notes that although both Indonesia and Nigeria had more than their fair share of corruption, corrupt activities were more regulated and organized in Indonesia. The leadership curbed excessive rent-seeking and never allowed corruption to undermine macroeconomic stability. Meanwhile, “Nigeria’s contentious, divided elites and unstable politics have fostered an anarchic realm of rent distribution” (Lewis 2007, p. 7), where corruption was a major driver of severe macroeconomic problems. This latter observation also harks back to the texts discussed in section 3.2, which study the way in which political coalitions have been built in postcolonial Africa, finding that they have rested mostly on material bases, as opposed to more programmatic commitments (even if self-advantageous, as in the case of Indonesia) . This points to the detrimental way in which political competition has generally operated in the continent, and to the need to build strong political organizations capable of fostering group cohesion.

Nigeria is renowned for being a country with particularly dysfunctional politics, and we cannot unreflectively generalize conclusions reached in these studies to other settings in the continent. Nevertheless, insofar as those familiar with the political context of other African countries can recognize some similarities, these accounts are useful in identifying common syndromes, even if they may be present elsewhere in more diluted forms. The differences between the Indonesian and Nigerian experiences highlighted by Lewis, together with the above considerations on cronyism and prebendalism, also demonstrate the heuristic value of an analytical distinction between the political and technical dimensions of organizations. As we will see below, to fully understand the meaning of this distinction for thinking about industrial policy, we need to explore the dynamic interactions between politics and organizational strength.


  1. Conclusion – Political Strategies for Industrial Development

7.1 Reconfiguring Interests and Forming Coalitions

Industrial policy interventions are characterized by a ‘pecking order’, whereby state bureaucratic organizations need to put in place policies that induce firms to exert effort in learning, and it must support them in this process (Khan 2013b). In turn, the task of directing the capabilities of these bureaucratic organizations falls upon politicians and political organizations. As discussed at length in the work of Mushtaq Khan, the problem is that organizations of various types are likely to resist policies which they perceive as contrary to their interests, and that they will be successful in their resistance if they are sufficiently powerful. Therefore, industrial policies must be designed to realign the interests of powerful stakeholders and organizations in ways that incentivize them to invest in learning and productive activities.

However, the practice of industrial policy is not simply a matter of finding a happy coincidence of interests, in which collaboration always makes all parties better-off.[29] Using the language of New Institutional Economics, Khan (2010) frames this problem in terms of the costs of transitioning from one institutional configuration to another, and discusses ways of navigating the ‘growth-stability trade-off’ when faced with coalitions that oppose the institutional changes.[30] In view of this, a development strategy must be based on a sound coalition of politically-powerful (or powerful enough) organizations that stand a chance of dealing with these form of resistance. This can involve a mixture of setting adequate incentives, co-optation, and confrontation of opposing coalitions. Such confrontations must be thought out strategically, as “in some cases, early compromises with powerful groups can build up more serious problems later.” (Khan 2010, p. 37) We return to this question below, when we join up the various components of development strategies, and discuss some commonly-employed tactics.

7.2 Strengthening Organizations

Besides concerns with incentive compatibility and coalitional demands, development strategies also require close attention to the strengthening of organizations. If we accept the view that stronger organizations are better-suited to delivering successful industrial policy, and that developing countries are characterized by generalized organizational weakness – though subject to great variability between and within organizations (Leonard 2010; Rasul, Rogger and Williams 2017) – then leaders need to consider how they will build capabilities in the relevant organizations. Usually, this means improving the capability of public organizations, an issue that cannot be explored at length here, but which is certainly essential.[31] Nonetheless, insofar as we are dealing with the politics of capacity-building, the main takeaway is that capacity-building requires steady political support from those in positions of authority and protection from forces with an interest in disrupting reform efforts and maintaining the status quo.[32]

Improvements firm capabilities are the subject of much of economic research, and there is no need to deal with them here.[33] Building on the discussion in section 6.2, it might be more interesting to think about capacity-building in political organizations. While solid political organizations are not a sine qua non of successful industrial policy, as authority can be exerted in informal ways, in the long run they help ensure that adequate policies are sustained for a reasonably long period of time, and that they are adapted as the economy evolves.[34] Moreover, as noted by Van de Walle (2007), the lack of adequate organizational resources is one of the reasons why postcolonial African leaders typically made recourse to the distribution of state resources to ensure political stability. Booth et al. (2015, p. iv) echo the general importance of organizational resources by claiming that “the key to leadership transitions that do not interrupt economic growth is the presence of one or other of two sorts of strong institution: a governing party with a tradition of consensual decision-making or, in the special case of Thailand, a state bureaucracy that can insulate policy from changes in political leadership.”[35]

The role of political organizations in maintaining a unified structure of authority and ensuring governability is also dealt with by the literature on authoritarian regimes (eg. Gandhi 2008; Svolik 2012). Benjamin Smith (2005), and Steven Levitsky and Lucan Way (2012; 2013) note that ruling parties forged during military conflict, which required the strengthening of the party’s organizational structures, are more likely to remain in power. Crucially, Levitsky and Way emphasize the role of non-material resources that ensure the maintenance of parties’ unity, such as “strong identities, solidarity ties, and the discipline generated by violent origins” (Levitsky and Way 2012, p. 870). One would suspect these same features to be useful for policy implementation by ensuring politicians’ cohesion around a defined set of objectives, as well as the power to enforce policies over the opposition of lower-level factions and other social forces. It is no coincidence that some of the contemporary African states most widely recognized for their ability to bring about developmental outcomes are ruled by political parties forged over the course of armed struggle (Jones et al. 2013).

This is not to commend single-party, authoritarian regimes; strong parties are present in both dictatorships and democracies, and they are more likely to be able to implement desired policies in both kinds of regimes (Bizzarro et al. 2018). Neither do we intend to sing the praises of war. Instead, we wish to put forward the point that leaders concerned with putting in place ambitious development strategies need to find ways of building the strength of their political parties, as well as other political institutions, a task that can be accomplished through skilled democratic leadership.

7.3 Political Resources

As mentioned above, strong political parties are an important resource for leaders to both buttress their power and implement economic policies. However, the potential resources at a reformist leader’s (or leadership’s) disposal are not restricted to their control over political organizations. Before discussing this point, it is important to clarify what we mean by ‘political resources’. We use the term to refer to all factors that increase the probability that a leader is able to implement his or her programme, and that decrease the stability costs of implementing it. Although much of the literature discusses the imperatives of coalition-building, and although the ability to build coalitions is important for implementing economic programmes, the two ideas are related but not coterminous. For instance, the existence of relations of authority over certain groups, or the command of military power, can be considered political resources, since they may improve the ability to implement a given programme and decrease the stability costs in doing so. However, most people would balk at calling the use of these resources ‘coalition-building’. This is not to say that coalition-building is not important. As we discuss below, the ability to build pro-growth coalitions is an essential component of a development strategy. Still, we choose to frame the problem in a more encompassing, and perhaps more cumbersome way, so as to be able to incorporate a wider array of political strategies.[36]

So how exactly are we to conceive of these political resources? Here, it may be useful to consider how other authors have thought about the issue. Examples of factors that have been included under this, or similar rubrics, include:

  • Presidents’ “executive toolbox”, including agenda power, budgetary prerogatives, cabinet management, partisan powers, and informal institutions. (Chaisty, Cheeseman and Power 2014)
  • “shrewd policies, imaginative political strategies, entrepreneurship, working to generate consensus, and political accommodation” (Hudson and Leftwich 2013, p. 80)
  • “organizational capabilities, the capacity of (the) leadership to mobilize and enthuse, and their skill in identifying and rewarding the right people through formal or informal networks.” (Khan 2017, p.5)
  • “Mobilization of political resources involves…the broader process of gaining support or acquiescence of social groups for those policies…Successful formulation and implementation of development policy requires both technical skills in the state apparatus and persuasive ability and legitimacy of the government.” (our emphasis) (Sikkink 1991, p. 207)

One may readily note that the factors such as leaders’ imaginations, their capacity to enthuse, or their persuasive ability are very hard to codify in a standardized set of practices. In fact, it is their intangibility which makes political strategies so context-dependent. This does not mean that these strategies cannot be analysed (it is the aim of this paper after all) but that academic attempts to generalize explanations for when leaders will choose particular strategies, or for the role of ideas in generating political outcomes, must be thought through very carefully. And while this is difficult enough in academia, it is even harder to incorporate them in a policy framework or in the work of international development agencies. That is why we opt to leave the description of these political resources intentionally vague, as locally-embedded actors will be the most adept in choosing the appropriate political strategy. Our role here is instead to raise the issues that they are likely to confront, and to offer a more structured way of thinking about the politics of industrial policy.

7.4 Putting it all Together – the Developmental Leader’s Problem

Having discussed the role of interest reconfiguration, organizational strengthening and political resources in development strategies, we are now ready to combine them in a unified way. Leaders can use their political resources to both harness existing political, bureaucratic and productive organizations to improve firm capabilities, and to create political space for the development of new capabilities among new and existing bureaucratic and political organizations.[37] Moreover, a part of these leadership resources can also be ‘reinvested’ in the enlargement of political resources, which to some extent can be synergistic with the reconfiguration of interests and organizational strengthening. For instance, by forming a pro-growth coalition, or strengthening a political party, a ruler can both improve economic performance and increase their probability of remaining in power. More succinctly, we can formulate the challenge as follows:

A sequence of decisions over time on how to allocate political resources across (a) interest reconfiguration and (b) organizational strengthening to improve firm’s capabilities, subject to political opposition and coalitional demands.

One can note the similarities between the way we frame these political strategies and the Bellman Equation in dynamic optimization problems.[38] In terms of the components of a Bellman Equation, we can think of the increase in firm capabilities as the objective function (ie. the variable we would like to optimize); the configuration of interests and the distribution of organizational power are the state variables (ie. the aspects of the current situation that matter at any point in time); and the allocation of political resources is the control variable (ie. the ones over which a choice can be made). Of course, politics is not mathematics, so the metaphor can only travel so far, but it serves to illustrate the fact that choices have to be made, the potential constraints these choices will be subject to, and the need to formulate a strategy from a dynamic perspective. Moreover, by being relatively catholic in describing the nature of these resources, interests, opposing political forces, and coalitional demands, it can help make sense of the variety of development strategies that can be observed in practice, while hopefully providing some structure to policymakers’ strategizing.

Clearly, development strategies will vary according to the nature of a country’s political economy. The elements that we flag here can help make sense of why these strategies might have varied in the past, and to what effect. An interesting point that can be raised is how the pre-existing distribution of organizational capabilities affects the development model pursued by a given country. For instance, in revolutionary regimes emerging out of revolution and civil conflict, such as China, Vietnam, Mexico, Ethiopia or Rwanda, the ruling party has been the chief architect and executor of the development strategy. In regimes forged by the Cold War, such as South Korea, and Brazil from 1964 onwards, it was the military that took a leading role in development. Meanwhile, France is an example of a country where at times unstable politics, as during the Fourth Republic (1946-58), was compensated by the presence of a strong state manned by a highly capable administrative elite (Loriaux 1999). Finally, in a country with both unstable politics and a weak state such as Bangladesh, the private sector has taken the leading role in development (Khan 2013a).

So what are the concrete implications of this way of thinking for the formulation of development strategies, particularly in contemporary Africa? As discussed in the historical examples above, the feasibility of a particular development strategy will depend on a country’s political economy profile, including the configuration of interests and political forces and the distribution of organizational capabilities. Clearly, stronger organizational capabilities and a greater amount of leadership resources will enable a more ambitious development strategy than what is possible in organizationally weak and politically fragmented polities. But we need to be more specific as to how these political economy profiles map onto optimal development strategies.[39] Below we make some preliminary considerations:

  • In African countries suffering from generalized organizational weakness, where political resources also tend to be limited, in the short run it might be easier to focus on interest reconfiguration as a way of resolving incentive and collective action problems.[40] This is likely to involve forms of political bargaining, coalition management, and rent allocation. In these scenarios, organizational strengthening, while always welcome if possible, is likely to play a minor role until the leadership is able to muster sufficient political resources.
  • In some countries, most organizations are weak, but political resources are relatively plentiful and there are strong political organizations. In these cases, there might be room for combining short-run interest reconfiguration, thus generating some economic growth, with medium- to long-run strengthening of state organizations. This is presumably the strategy currently being pursued by Africa’s ‘illiberal state-builders’.
  • Elsewhere, private firms might be the strongest organizations, but in Africa these are often in the hands of foreigners or politically weak minorities (Ramachandra, Gelb and Shah 2009; Gelb, Meyer and Ramachandran 2014). In these cases, a feasible development strategy might follow the example of South-East Asian countries such as Malaysia or Indonesia, where foreign firms and Chinese-owned businesses received state support, or at the very least ‘benign neglect’, in exchange for acquiescing to rent redistribution and jobs provision to ethnic or racial majorities (Lubeck 1992; Bowie 1993; Ritchie 2005; Lewis 2007) .
  • In African countries where coalition-building is still primarily based on the distribution of material resources, “the ring-fencing of policy fields accepted as being of special national interest” (Booth 2012, p.x) may be a good idea to ensure policies are not modified by opposing forces. Peter Lewis (2007, p. 278) discusses how in Indonesia this was done by introducing “ex ante lock-in measures” to “ensure asset holders against arbitrary expropriation from unforeseen policy shifts”, and delegating economic policy “to an effective technocratic team” thus “signalling the government’s intentions to private economic actors.”
  • In war-torn countries or those with ‘failed states’, the challenge is to forge coalitions and build institutions that can preserve peace, while opportunities of engaging in industrial policy might be limited.[41]

In practice, there will be a broader range of possible political economy configurations, which in turn will vary in more granular ways than described here. But while mapping the distribution of organizational power and the constellation of interests present at a given point in time might be helpful, we believe that the need for such kinds of information collection has been overemphasized in the literature. Firstly, trying to measure variables such as the social distribution of power poses some inherent difficulties. Moreover, the time and resources expended in such an exercise are better employed elsewhere, especially since by the time such an assessment has been completed, the context might have changed in important ways.

This is not to argue that information is superfluous, or that more information is not always better. But it is likely that a high-level political actor already has an understanding of the broad contours of the political environment in which they have to operate. Where they do not have it, perhaps when trying to operate at the provincial level, or in a policy area with which they are unfamiliar, it certainly is a good idea to seek more information. But, generally speaking, we believe that keeping in mind the ideas presented in this paper and the framing of the problem outlined in this section can already go a long way in helping to structure the policy and political challenges likely to be faced.

Further research and synthesis work would be helpful in bringing to light the different techniques that in the past have been to navigate the developmental leader’s problem. The collective findings of this research can then be assembled to create a sort of “industrializer’s toolbox” (to paraphrase Chaisty et al.’s (2012). But in the absence of such an exercise, we believe that our formulation of the developmental leader’s problem, together with the collective insights of decades of research in the political economy of industrial policy, can help provide some needed structures to the strategies of leaders with transformational aspirations.


  1. References

Abdulai, A.-G., & Hickey, S. (2016). The politics of development under competitive clientelism: Insights from Ghana’s education sector. African Affairs, 115(458), 44-72.

Acemoglu, D., Johnson, S., & Robinson, J. A. (2001). The Colonial Origins of Comparative Development: An Empirical Investigation. American Economic Review, 91(5), 1369-1401.

Alesina, A., Easterly, W., & Matuszeski, J. (2011). Artificial States. Journal of the European Economic Association, 9(2), 246-277.

Allen, C. (1995). Understanding African Politics. Review of African Political Economy, 22(65), 301-320.

Andrews, M., Pritchett, L., & Woolcock, M. (2017). Building State Capability: Evidence, Analysis, Action. Oxford: Oxford University Press.

Ashraf, Q., & Galor, O. (2013). The ‘Out of Africa’ Hypothesis, Human Genetic Diversity, and Comparative Economic Development. American Economic Review, 103(1), 1-46.

Bach, D. C. (2012). Patrimonialism and neopatrimonialism: comparative receptions and transcriptions. In D. C. Bach, & M. Gazibo (Eds.), Neopatrimonialism in Africa and Beyond (pp. 25-45). Oxford: Routledge.

Bach, D. C., & Gazibo, M. (Eds.). (2012). Neopatrimonialism in Africa and Beyond. London: Routledge.

Barber, M. (2015). How to Run a Government: So that Citizens Benefit and Taxpayers Don’t Go Crazy. London: Penguin Books.

Bayart, J.-F. (1993). The State in Africa: The Politics of the Belly. London: Longman.

Bayart, J.-F. (2000). Africa in the World: A History of Extraversion. African Affairs, 99(395), 217-267.

Behuria, P. (2015). Between party capitalism and market reforms – understanding sector differences in Rwanda. Journal of Modern African Studies, 53(3), 415-450.

Behuria, P. (2016). Countering threats, stabilising politics and selling hope: examining the Agaciro concept as a response to a critical juncture in Rwanda. Journal of Eastern African Studies, 10(3), 434-451.

Behuria, P., Buur, L., & Gray, H. (2017). Research Note: Studying Political Settlements in Africa. African Affairs, 116(464), 508-525.

Bell, C., & Pospisil, J. (2017). Navigating Inclusion in Transitions from Conflict: The Formalised Political Unsettlement. Journal of International Development, 29(5), 576-593.

Berman, B. J. (1998). Ethnicity, Patronage and the African State: The Politics of Uncivil Nationalism. African Affairs, 97(388), 305-341.

Bizzarro, F., Gerring, J., Knutsen, C. H., Hicken, A., Bernhard, M., Skaaning, S.-E., . . . Lindberg, S. I. (2018). Party Strength and Economic Growth. World Politics, 70(2), 275-320.

Bloom, N., & Van Reenen, J. (2007). Measuring and Explaining Management Practices across Firms and Countries. Quarterly Journal of Economics, 72(4), 1351-1408.

Bloom, N., Lemos, R., Sadun, R., Scur, D., & Van Reenen, J. (2014). The New Empirical Economics of Management. Journal of the European Economics Association, 12(4), 835-876.

Boone, C. (1993). Commerce in Côte d’Ivoire: Ivoirianisation without Ivoirian Traders. Journal of Modern African Studies, 31(1), 67-92.

Boone, C. (1994). States and ruling classes in postcolonial Africa: the enduring contradictions of power. In J. S. Migdal, A. Kohli, & V. Shue (Eds.), State power and social forces: Domination and transformation in the Third World (pp. 108-140). Cambridge: Cambridge University Press.

Booth, D. (2012). Development as a collective action problem: Addressing the real challenges of African governance. London: ODI Africa Power and Politics Programme.

Booth, D. (2014). Towards a developmental state concept that is fit for purpose (in the 21st century, in Africa). Paper presented at the African Studies Association Annual Meeting, Indianapolis 20-22 November.

Booth, D., & Cammack, D. (2013). Governance for Development in Africa: Solving Collective Action Problems. London: Zed Books.

Booth, D., & Golooba-Mutebi, F. (2012). Developmental Patrimonialism? The Case of Rwanda. African Affairs, 111(444), 379-403.

Booth, D., Dietz, T., Golooba-Mutebi, F., Fuady, A. H., Henley, D., Kelsall, T., . . . van Donge, J. K. (2015). Developmental Regimes in Africa: initiating and sustaining developmental regimes in Africa. Synthesis Report.

Bowie, A. (1993). Crossing the Industrial Divide: State, Society, and the Politics of Economic Transformation in Malaysia. New York: Columbia University Press.

Brautigam, D., Rakner, L., & Taylor, S. (2002). Business Associations and Growth Coalitions in Sub-Saharan Africa. Journal of Modern African Studies, 40(4), 519-547.

Centeno, M. A., & Ferraro, A. E. (Eds.). (2013). State and Nation Making in Latin America and Spain: Republics of the Possible. Cambridge: Cambridge University Press.

Chabal, P. (1999). Africa Works: Disorder as Political Instrument. Oxford: James Currey.

Chaisty, P., Cheeseman, N., & Power, T. (2014). Rethinking the ‘presidentialism debate’: conceptualizing coalitional politics in cross-regional perspective. Democratization, 21(1), 72-94.

Collier, D., & Gerring, J. (Eds.). (2009). Concepts and Method in Social Science: The Tradition of Giovanni Sartori. New York: Routledge.

Commission on State Fragility, Growth and Development. (2018). Escaping the Fragility Trap. International Growth Centre and Blavatnik School of Government, University of Oxford.

Cooper, F. (2002). Africa since 1940: The past of the present. New York: Cambridge University Press.

Crespi, G., Fernández-Arias, E., & Stein, E. (2014). Rethinking Productive Development: Sound Policies and Institutions for Economic Transformation. New York: Palgrave Macmillan.

Croese, S. (2016). State-Led Housing Delivery as an Instrument of Developmental Patrimonialism: The Case of Post-War Angola. African Affairs, 116(462), 80-100.

Crook, R. C. (1989). Patrimonialism, Administrative Effectiveness and Economic Development in Cote d’Ivoire. African Affairs, 205-228.

Cumings, B. (1984). The origins and development of the Northeast Asian political economy: industrial sectors, product cycles, and political consequences. International Organization, 38(1), 1-40.

Dercon, S., Lippolis, N., & Peel, S. (2018). How do the Goods a Country Produces Affect its Growth? Background Paper for Programme on Rethinking African Paths to Industrial Development. Oxford: Blavatnik School of Government and Centre for the Study of African Economies, University of Oxford.

Di John, J., & Putzel, J. (2009). Political Settlements: Issue Paper. Governance and Social Development Resource Centre, GSDRC Issues Paper, Birmingham.

Doner, R. F. (1992). Limits of State Strength: Toward an Institutionalist View of Economic Development. World Politics, 44, 398-431.

Doner, R. F., & Schneider, B. R. (2000). Business Associations and Economic Development: Why Some Associations Contribute More Than Others. Business and Politics, 2(3), 261-288.

Doner, R. F., Ritchie, B. K., & Slater, D. (2005). Systemic Vulnerability and the Origins of Developmental States: Northeast and Southeast Asia in Comparative Perspective. International Organization, 59, 327-361.

Evans, P. B. (1989). Predatory, Developmental, and Other Apparatuses: A Comparative Political Economy Perspective on the Third World State. Sociological Forum, 4(4), 561-587.

Evans, P. B. (1995). Embedded Autonomy: States and Industrial Transformation. Princeton: Princeton University Press.

Evans, P. B. (1996). Introduction: Development strategies across the public-private divide. World Development, 24(6), 1033-1037.

Evans, P. B. (1997). State Structures, Government-Business Relations, and Economic Transformation. In Business and the State in Developing Countries (pp. 63-87). Ithaca and London: Cornell University Press.

Fernández-Arias, E., Sabel, C., Stein, E. H., & Trejos, A. (2016). Two to Tango: Public-Private Collaboration for Productive Development Policies. Washington, DC: Inter-American Development Bank.

Gandhi, J. (2008). Political Institutions under Dictatorship. New York: Cambridge University Press.

Geddes, B. (1990). Building State Autonomy in Brazil, 1930-1964. Comparative Politics, 22(2), 217-236.

Geddes, B. (1994). Politician’s Dilemma: Building State Capacity in Latin America. Berkeley: University of California Press.

Gelb, A., Meyer, C. J., & Ramachandran, V. (2014). Development as Diffusion: Manufacturing Productivity and Sub-Saharan Africa’s Missing Middle. Center for Global Development Working Paper 357.

Gerring, J., & Barresi, P. A. (2009). Culture: joining minimal definitions and ideal types. In D. Collier, & J. Gerring (Eds.), Concepts and Method in Social Science: The Tradition of Giovanni Sartori (pp. 241-268). New York: Routledge.

Geuss, R. (2008). Philosophy and Real Politics. Princeton: Princeton University Press.

Goertz, G. (2009). Point of departure: intension and extension. In D. Collier, & J. Gerring (Eds.), Concepts and Method in Social Science: The Tradition of Giovanni Sartori (pp. 181-202). New York: Routledge.

Gray, H. (2013). Industrial policy and the political settlement in Tanzania: aspects of continuity and change since independence. Review of African Political Economy, 40(136), 185-201.

Gray, H. (2015). The Political Economy of Grand Corruption in Tanzania. African Affairs, 114(456), 382-403.

Gray, H. (2018). Turbulence and Order in Economic Development: Economic Transformation in Tanzania and Vietnam. Oxford: Oxford University Press.

Haggard, S. (1990). Pathways from the Periphery: The Politics of Growth in the Newly Industrializing Countries. Ithaca, NY: Cornell University Press.

Haggard, S., Maxfield, S., & Schneider, B. R. (1997). Theories of Business and Business-State Relations. In Business and the State in Developing Countries (pp. 36-60). Ithaca and London: Cornell University Press.

Herbst, J. I. (2014). States and Power in Africa: Comparative Lessons in Authority and Control (2nd Edition ed.). Princeton: Princeton University Press.

Hickey, S. (2013). Thinking about the politics of development: towards a relational approach. ESID Working Paper No. 1.

Hickey, S., & Izama, A. (2017). The poilitics of governing oil in Uganda: Going against the grain? African Affairs, 116(463), 163-185.

Himbara, D. (1993). Myths and Realities of Kenyan Capitalism. Journal of Modern African Studies, 31(1), 93-107.

Hirschman, A. O. (1970). The Search for Paradigms as a Hindrance to Understanding. World Politics, 22(3), 329-343.

Hirvi, M., & Whitfield, L. (2015). Public-service provision in clientelist political settlements: Lessons from Ghana’s urban water sector. Development Policy Review, 33(2), 135-158.

Hudson, D., & Leftwich, A. (2014, June). From Political Economy to Political Analysis. Research Paper 25 – Developmental Leadership Program: Policy and Practice for Developmental Leaders, Elites and Coalitions.

Huntington, S. P. (1968). Political Order in Changing Societies. New Haven: Yale University Press.

Hyden, G. (1980). Beyond Ujamaa in Tanzania: Underdevelopment and an Uncaptured Peasantry. London: Heinemann.

Jackson, R. H., & Rosberg, C. G. (1984). Personal Rule: Theory and Practice in Africa. Comparative Politics, 16(4), 421-442.

Jerven, M. (2015). Africa: Why Economists Get it Wrong. London: Zed Books.

Johnson, C. (1982). MITI and the Japanese Miracle: The Growth of Industrial Policy, 1925-1975. Stanford: Stanford University Press.

Jones, W., Soares de Oliveira, R., & Verhoeven, H. (2013). Africa’s Illiberal State-Builders. Oxford Refugee Studies Centre Working Paper .

Joseph, R. (1987). Democracy and Prebendal Politics in Nigeria: The Rise and Fall of the Second Republic. Cambridge: Cambridge University Press.

Kelsall, T. (2011). Rethinking the Relationship between Neo-patrimonialism and Economic Development in Africa. IDS Bulletin, 42(2), 76-87.

Kelsall, T. (2013). Business, Politics, and the State in Africa: Challenging the Orthodoxies on Growth and Transformation. London: Zed Books.

Kelsall, T. (2016, January). Thinking and working with political settlements. ODI Briefing.

Khan, M. H. (2000). Rents, Efficiency and Growth. In M. Khan, & K. S. Jomo (Eds.), Rents, Rent-Seeking and Economic Development: Theory and Evidence in Asia (pp. 21-69). Cambridge: Cambridge University Press.

Khan, M. H. (2010). Political Settlements and the Governance of Growth-Enhancing Institutions.

Khan, M. H. (2013a). Bangladesh: Economic Growth in a Vulnerable Limited Access Order. In D. W. North, S. Webb, & B. Weingast (Eds.), In the Shadow of Violence: Politics, Economics and the Problems of Development (pp. 24-69). Cambridge: Cambridge University Press.

Khan, M. H. (2013b). Political Settlements and the Design of Technology Policy. In J. Stiglitz, J. Y. Lin, & E. Patel (Eds.), The Industrial Policy Revolution II. Africa in the 21st Century (pp. 243-280). London: Palgrave.

Khan, M. H. (2017). Introduction: Political Settlements and the Analysis of Institutions. African Affairs, Online Issue on Political Settlements, 1-20.

Khan, M. H., Andreoni, A., & Roy, P. (2016). Anti-Corruption in Adverse Contexts: A Strategic Approach.

Kjaer, A. M. (2015). Political Settlements and Productive Sector Policies: Understanding Sector Differences in Uganda. World Development, 68, 230-241.

Kohli, A. (2004). State-Directed Development: Political Power and Industrialization in the Global Periphery. Cambridge: Cambridge University Press.

Languille, S. (2016). The scramble for textbooks in Tanzania. African Affairs, 115(458), 73-96.

Laws, E. (2012). Political Settlements, Elite Pacts, and Governments of National Unity. AusAID Background Paper 10 – Developmental Leadership Program.

Laws, E., & Leftwich, A. (2014). Political Settlements. University of Birmingham, International Development Department. Developmental Leadership Program.

Leftwich, A. (1995). Bringing Politics Back In: Towards a Model of the Developmental State. Journal of Development Studies, 31(3), 400-427.

Leftwich, A. (2010). Beyond Institutions: Rethinking the Role of Leaders, Elites and Coalitions in the Institutional Formation of Developmental States and Strategies. Forum for Development Studies, 37(1), 93-111.

Lemos, R., & Scur, D. (2015). A snapshot of mid-sized firms in Africa, Asia and Latin America. Mimeo.

Leonard, D. K. (2010). ‘Pockets’ of Effective Agencies in Weak Governance States: Where are they Likely and Why does it Matter? Public Administration and Development, 30, 91-101.

Levitsky, S. R., & Way, L. A. (2012). Beyond Patronage: Violent Struggle, Ruling Party Cohesion, and Authoritarian Durability. Perspectives on Politics, 10(4), 869-889.

Levitsky, S., & Way, L. (2013). The Durability of Revolutionary Regimes. Journal of Democracy, 24(3), 5-17.

Levy, B. (2014). Working with the Grain: Integrating Governance and Growth in Development Strategies. New York: Oxford University Press.

Lewis, P. (2007). Growing Apart: Oil, Politics, and Economic Change in Indonesia and Nigeria. Ann Arbor: University of Michigan Press.

Lippolis, N., & Peel, S. (2018a). Firms, Productivity Dispersion and Reallocation. Background Paper for Programme on Rethinking African Paths to Industrial Development. Oxford: Blavatnik School of Government and Centre for the Study of African Economies, University of Oxford.

Lippolis, N., & Peel, S. (2018b). Issues in Industrial Policy Design. Background Paper for Programme on Rethinking African Paths to Industrial Development. Oxford: Blavatnik School of Government and Centre for the Study of African Economies, University of Oxford.

Loriaux, M. (1999). The French Developmental State as Myth and Moral Ambition. In M. Woo-Cumings (Ed.), The Developmental State (pp. 235-275). Ithaca: Cornell University Press.

Lubeck, P. (1992). Malaysian Industrialization, Ethnic Divisions, and the NIC Model: The Limits to Replication. In J. Henderson, & R. Appelbaum (Eds.), States and Development in the Pacific Rim. Newbury Park: Sage.

Macuane, J. J., Buur, L., & Monjane, C. M. (2017). Power, conflict and natural resources: The Mozambican crisis revisited. African Affairs, 1-24.

Maxfield, S., & Schneider, B. R. (1997). Business, the State, and Economic Performance in Developing Countries. In S. Maxfield, & B. R. Schneider (Eds.), Business and the State in Developing Countries (pp. 3-35). Ithaca and London: Cornell University Press.

Michalopoulos, S., & Papaioannou, E. (2013). Pre-colonial Ethnic Institutions and Contemporary African Development. Econometrica, 81(1), 113-152.

Michalopoulos, S., & Papaioannou, E. (2014). National Institutions and Subnational Development in Africa. Quarterly Journal of Economics, 129(1), 151-213.

Michalopoulos, S., & Papaioannou, E. (2016). The Long-Run Effects of the Scramble for Africa. American Economic Review, 106(7), 1802-1848.

Miliband, R. (1969). The State in Capitalist Society. London: Weidenfeld and Nicholson.

Mkandawire, T. (2001). Thinking about developmental states in Africa. Cambridge Journal of Economics, 25, 289-313.

Mkandawire, T. (2015). Neopatrimonialism and the Political Economy of Economic Performance in Africa: Critical Reflections. World Politics, 67(3), 563-612.

Moon, C.-I., & Prasad, R. (1994). Beyond the Developmental State: Networks, Politics, and Institutions. Governance: An International Journal of Policy and Administration, 7(4), 360-386.

North, D. C., Wallis, J. J., & Weingast, B. R. (2006). A Conceptual Framework for Interpreting Recorded Human History. NBER Working Paper No. 12759.

North, D. C., Wallis, J. J., & Weingast, B. R. (2009). Violence and Social Orders: A Conceptual Framework for Interpreting Recorded Human History. Cambridge: Cambridge University Press.

Nugent, P. (2004). Africa Since Independence: A Comparative History. Basingstoke: Palgrave Macmillan.

Nunn, N. (2008). The Long-term Effects of Africa’s Slave Trade. Quarterly Journal of Economics, 123(1), 139-176.

Nunn, N., & Wantchekon, L. (2011). The Slave Trade and the Origins of Mistrust in Africa. American Economic Review, 101(7), 3221-3252.

Onis, Z. (1991). The Logic of the Developmental State. Comparative Politics, 24(1), 109-126.

Oqubay, A. (2015). Made in Africa: Industrial Policy in Ethiopia. Oxford: Oxford University Press.

Ostrom, E. (2005). Understanding Institutional Diversity. Princeton: Princeton University Press.

Parks, T., & Cole, W. (2010). Political Settlements: Implications for International Development Policy and Practice. Occasional Paper No. 2, The Asia Foundation.

Pitcher, A., Moran, M. H., & Johnston, M. (2009). Rethinking Patrimonialism and Neopatrimonialism in Africa. African Studies Review, 52(1), 125-156.

Pospisil, J., & Rocha Menocal, A. (2017). Why Political Settlements Matter: Navigating Inclusion in Processes of Institutional Transformation. Journal of International Development, 29, 551-558.

Pritchett, L., Sen, K., & Werker, E. (Eds.). (2018). Deals and Development: The Political Dynamics of Growth Episodes. Oxford: Oxford University Press.

Przeworski, A., Alvarez, M. E., Cheibub, J. A., & Limongi, F. (2000). Democracy and Development: Political Institutions and Well-being in the World. 1950-1990. Cambridge: Cambridge University Press.

Ramachandran, V., Gelb, A., & Shah, M. K. (2009). Africa’s Private Sector: What’s Wrong with the Business Environment and What to Do About It. Washington, DC: Center for Global Development.

Rasul, I., & Rogger, D. (2018). Management of Bureaucrats and Public Service Delivery: Evidence from the Nigerian Civil Service. Economic Journal, 128, 413-446.

Rasul, I., Rogger, D., & Williams, M. J. (2017). Management and Bureaucratic Effectiveness: A Scientific Replication. Mimeo.

Ritchie, B. K. (2005). Coalitional Politics, Economic Reform, and Technological Upgrading in Malaysia. World Development, 33(5), 745-761.

Rocha Menocal, A. (2015). Inclusive political settlements: evidence, gaps, and challenges of institutional transformation. University of Birmingham, International Development Department. Birmingham: Developmental Leadership Program.

Rodan, G. (1989). The Political Economy of Singapore’s Industrialization. Basingstoke: Palgrave Macmillan.

Sartori, G. (1970). Concept Misformation in Comparative Politics. American Political Science Review, 64(4), 1033-1053.

Sartori, G. (2009). Guidelines for Concept Analysis. In D. Collier, & J. Gerring (Eds.), Concepts and Method in Social Science: The Tradition of Giovanni Sartori (pp. 97-150). New York: Routledge.

Sartori, G. (2009). The Tower of Babel. In D. Collier, & J. Gerring (Eds.), Concepts and Method in Social Science: The Tradition of Giovanni Sartori (pp. 61-96). New York: Routledge.

Schneider, B. R. (1991). Politics within the State: Elite Bureaucrats and Industrial Policy in Authoritarian Brazil. Pittsburgh: University of Pennsylvania Press.

Schneider, B. R. (1993). The Career Connection: A Comparative Analysis of Bureaucratic Preferences and Insulation. Comparative Politics, 25(3), 331-350.

Schneider, B. R. (1998, October). Elusive Synergy: Business-Government Relations and Development. Comparative Politics, 101-122.

Schneider, B. R. (1999). The Desarrollista State in Brazil and Mexico. In M. Woo-Cumings (Ed.), The Developmental State (pp. 276-305). Ithaca and London: Cornell University Press.

Schneider, B. R. (2015). Designing Industrial Policy in Latin America: Business-State Relations and the New Developmentalism. New York: Palgrave Macmillan.

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

Sikkink, K. (1991). Ideas and Institutions: Developmentalism in Brazil and Argentina. Ithaca: Cornell University Press.

Skocpol, T. (1985). Bringing the state back in: strategies of analysis in current research. In P. Evans, D. Rueschemeyer, & T. Skocpol, Bringing the State Back In. Cambridge: Cambridge University Press.

Smith, B. (2005). Life of the Party: The Origins of Regime Breakdown and Persistence under Single-Party Rule. World Politics, 57(3), 421-451.

Stubbs, R. (2009). What ever happened to the East Asian Developmental State? The unfolding debate. The Pacific Review, 22(1), 1-22.

Svolik, M. (2012). The Politics of Authoritarian Rule. New York: Cambridge University Press.

Theobald, R. (1982). Patrimonialism. World Politics, 34(4), 548-559.

Tilly, C. (1992). Coercion, Capital, and European States, AD 990-1992. Cambridge, MA: Blackwell.

Van de Walle, N. (2001). African Economies and the Politics of Permanent Crisis, 1979-1999. New York: Cambridge University Press.

Van de Walle, N. (2007). Meet the new boss, same as the old boss? The evolution of political clientelism in Africa. In H. Kitschelt, & S. I. Wilkinson (Eds.), Patrons, Clients and Policies: Patterns of Democratic Accountability and Political Competition (pp. 50-67). Cambridge: Cambridge University Press.

Van de Walle, N. (2016). Conclusion: democracy fatigue and the ghost of modernization theory. In T. Hagmann, & F. Reyntjens (Eds.), Aid and authoritarianism in Africa: Development without democracy (pp. 161-175). London: Zed Books.

Wade, R. (1990). Governing the Market: Economic Theory and the Role of Government in East Asian Industrialization. Princeton: Princeton University Press.

Wade, R. (1992). East Asia’s Economic Success: Conflicting Perspectives, Partial Insights, Shaky Evidence. World Politics, 44(2), 270-320.

Weber, M. (1991). The Nature of Social Action. In W. Runciman (Ed.), Weber: Selections in Translation. Cambridge: Cambridge University Press.

Weber, M. (2009). Bureaucracy. In H. Gerth, & C. Wright Mills (Eds.), From Max Weber: Essays in Sociology. Oxford: Routledge.

Weiss, L. (2000). Developmental states in transition: adapting, dismantling, innovating, not ‘normalizing’. The Pacific Review, 13(1), 21-55.

Whitfield, L., & Buur, L. (2014). The politics of industrial policy: ruling elites and their alliances. Third World Quarterly, 126-144.

Whitfield, L., Therkildsen, O., Buur, L., & Kjaer, A. M. (2015). The Politics of African Industrial Policy: A Comparative Perspective. New York: Cambridge University Press.

Woo-Cumings, M. (Ed.). (1999). The Developmental State. Ithaca: Cornell University Press.

Young, C. (1994). The African Colonial State in Comparative Perspective. New Haven and London: Yale University Press.


[1] Although ‘political economy’ is a term with many contending definitions, in this piece we will use it to refer to the field of study concerned with understanding how power relations – or ‘politics’ broadly defined – affect the distribution of resources in a given society.

[2] We don’t engage in older debates such as that on ‘dependency theory’ because those theories have lost much support, as the East Asian ‘tiger economies’ have shown that manufacturing-led development is possible in the ‘global periphery’.

[3] By ‘industrialization’, we refer to the process whereby productive factors are reallocated towards more productive activities and firms learn to produce higher quality goods. The manufacturing sector has traditionally been a locus of such processes, but ‘industrialization’ does not need to refer exclusively to manufacturing. See Dercon, Lippolis and Peel (2018) and Lippolis and Peel (2018a) for more in-depth treatments of this issue.

[4] For an elaboration of the distinction between ideal types and other kinds of concept, see Goertz (2009).

[5] To use Sartori’s (1970) vocabulary, he decreases the concept’s intension, increasing its extension and moving it up in the ladder of abstraction towards a global concept. As a result, the way in which Evans employs the concept differs from earlier accounts.

[6] Kohli mentions the United States as a possible example of a fragmented-capitalist state, while Sweden might be considered a coherent-multiclass state.

[7] Similarly to the concept of ‘developmental state’, ‘neo-patrimonialism’ has also been employed in different ways, and the precise meaning of the term has inspired a theoretical debate. See Bach and Gazibo (2012) for an in-depth exploration of neo-patrimonialism in Africa and other regions.

[8] See Van de Walle (2001) for a prominent account of how neo-patrimonialism has played out in the economic sphere in postcolonial Africa.

[9] See, for example, Himbara (1993) on Kenya and Boone (1993) on Cote d’Ivoire.

[10] Khan uses the textbook definition of an economic rent as a payment to a factor of production that is higher than the minimum required to attract that factor to a particular industry. He only deviates from neoclassical theory in highlighting the distinction between attracting a factor to a particular industry and inducing it to produce. See Khan (2000).

[11] For a more in-depth treatment of the relationship between an economy’s structure and the kind of policy pushed for by firms, see Pritchett, Sen and Werker (2018).

[12] For a more comprehensive review of recent research employing the political settlements approach, see Behuria, Buur and Gray (2017).

[13] These same authors would come to employ it in later texts such as Booth et al. (2015) and Kelsall (2016), which we discuss below.

[14] “What matters in defining a sustainable political economy at the macro level is not an explicit, or even an implicit, agreement or pact between elites, but a stable distribution of power across organizations (emphasis in the original).” (Khan 2017, p. 18)

[15] See, for example, Rocha Menocal (2015), as well as the contributions to the special issue on political settlements in the Journal of International Development, 29(5) (2017).

[16] The discussion in this section is based primarily on Sartori’s (1970) paper, as well as the contributions collected in Collier and Gerring (2009).

[17] See Stubbs (2009) for a more in-depth, historical account of the evolution of the concept of ‘developmental state’.

[18] In chapter 1 of the book, Johnson contrasts his approach to the earlier literature on Japan: “My study does not follow these earlier works in advocating the adoption of Japanese institutions outside of Japan. It does, however, try to lay out in their full complexity (our emphasis) some of the main Japanese institutions in the economic field so that those who are interested in adopting them will have an idea of what they are buying in terms of the Japanese system’s consequences – intended, unintended, and even unwanted.” (Johnson 1982, p. 7).

[19] On this point, see Sartori (2009) and Gerring and Barresi (2009).

[20] See Jones et al. (2013).

[21] See the works cited in section, especially Mkandawire (2001; 2015) and Pitcher et al. (2009).

[22] Schneider (1999) notes that the terms ‘desarrollista’/’desenvolvimentista’, which loosely translate from Spanish and Portuguese as ‘developmental’, were already applied to Latin American states by scholars in the late 1960’s. However the concept of “developmental state” only appeared with force among English-language scholars after the publication of Chalmers Johnson’s treatise.

[23] Citing works on the origins of developmental leadership, Lewis (2007, p. 8) notes that “Structural factors alone cannot explain economic outcomes, making it essential to account for the incentives and strategies of rulers. While developmental leadership is often serendipitous and difficult to anticipate, rulers with a strong economic agenda arise in specific circumstances. Leaders commonly embark on a developmental course in response to economic crisis, domestic instability, or particular types of security threats. Political strategies toward the economy are influenced by the nature of challenges as well as the available coalitions among elites.” (Our emphasis).

[24] See, for example, Sikkink’s (1991) account of the failure of Arturo Frondizi’s government (1958-1962) in implementing a developmentalist programme in Argentina.

[25] See the discussion in section 7.4.

[26] This description of the technical dimension of organizations resounds with the literature in economics on the quality of management practices (Bloom and Van Reenen 2007; Bloom et al. 2014), which has recently been extended to deal with public bureaucracies (Rasul and Rogger 2018).

[27] Rasul and Rogger (2018) and Rasul, Rogger and Williams (2017) do perform this kind of assessment, but restrict themselves to Nigeria and Ghana, respectively.

[28] See section 2.3.

[29] For Arkebe Oqubay, the ‘discovery process’ approach “reduces the process to an unrealistic fable whose moral is that everyone should be nice and listen to each other, the typical charm of liberal assumptions.” (Oqubay 2015, p. 254).

[30] See section 4.2.

[31] For a prominent recent contribution, see Andrews et al. (2017). For a different approach, see Barber (2015).

[32] This might not always be as straightforward as simply getting the support of the top minister, as “authorizing structures” can be more complex than the ideal type hierarchical-bureaucratic structure (Andrews et al. 2017). For the purposes of this paper, there is no need to discuss at length the precise way political support for institutional reforms is to be given. See Andrews et al. (2017) for a more in-depth discussion of how to “manage the authorizing environment” for capability-building, keeping in mind that they mostly take a donor’s perspective, rather than the national leadership’s perspective we take in this paper.

[33] See Dercon et al. (2018) and Lippolis and Peel (2018) for a discussion on the role and dynamics of firm capabilities in industrial development.

[34] See Lippolis and Peel (2018b) for a discussion on the role of institutions in thinking about industrial policy.

[35] Note their recognition of how different settings might differ in their configuration of organizational strength across political, bureaucratic, and business spheres. We return to this theme in the final sub-section.

[36] Tellingly, Mushtaq Khan, one of the foremost theorists of the role of power in industrial policy, does not discuss coalition-building at length. Instead, he frames the problem of articulating development strategies from the perspective of “the ruling coalition”, a term that resembles our reference to leaders and leadership.

[37] As put by Lewis (2007, p. 282): “the challenge is to favour entrepreneurs while assuaging distributional tensions”. Note that favouring entrepreneurs, in and of itself, is not sufficient; they must be favoured in ways that encourage them to improve their productive capabilities.

[38] The Bellman equation is a mathematical optimization method used for maximizing a given variable over a discrete time horizon by breaking down the problem into a series of simpler subproblems.

[39] For a stimulating discussion on the relationship between initial political economy conditions and the design of development strategies, see Levy (2014).

[40] Similar concepts are put forward in Peter Evans’ (1996) idea of “soft technologies of organizational design”, or Brian Levy’s (2014) idea of “multistakeholder strategies”, as well as the literature on the design of incentive-compatible industrial policies in Latin America (Crespi et al. 2014; Schneider 2015; Fernández-Arias et al. 2016).

[41] For a comprehensive, policy-oriented account of how to deal with state fragility, see Commission on State Fragility, Growth and Development (2018).