State Formation, Nation-Building and State–Society Relations

The process leading to a capable and autonomous state is a long one, and the product of relations between state and society, on the one hand, and the state and the international system, on the other. State formation is fundamentally shaped by critical junctures such as war, social revolution and colonialism (Skocpol Reference Skocpol1979, Mann Reference Mann1986, Reference Mann2012, Tilly Reference Tilly1992, Ertman Reference Ertman, Lange and Rueschemeyer2005, Rueschemeyer Reference Rueschemeyer, Lange and Rueschemeyer2005). In particular, capacity and autonomy arise when state formation results in political centralisation, which enables ruling elites to invest resources in increased state capacity.

A key turning point concerns how pre-existing political authorities responded to the rise of capitalism and the nation-state in Europe in the nineteenth century (Skocpol Reference Skocpol1979, Migdal Reference Migdal1988, Waldner Reference Waldner1999, Wimmer Reference Wimmer2018). The expansion of economic, and consequently political and military, power that accompanied capitalist development presented a fundamental challenge to authorities across the globe. Threatened with the loss of independence, ruling elites sought to centralise power, bureaucratising mediated states, redistributing property rights to initiate capitalist development and mobilising the masses through nationalist appeals (Skocpol Reference Skocpol1979, Anderson Reference Anderson1991, Wimmer Reference Wimmer2018).Footnote ³ In doing so, leaders necessarily clashed with landed elites, on whose shoulders previous mediated states rested. In Prussia and Japan, leaders centralised sufficient power to drive through reforms, enhancing state capacity and initiating industrialisation. In contrast, where the nobility was able to resist modernisation, revolutionary movements in France, Russia and China upended the political system and expanded state powers by building mass inclusion states (Skocpol Reference Skocpol1979, Goodwin Reference Goodwin2001). In many cases, however, the weakness of existing rulers resulted in colonial conquest,

The end of the nineteenth century was also a critical juncture for nation-building. A key distinction can be made between countries in which political leaders had centralised power prior to the arrival of mass politics at the end of the nineteenth century and those that had not (Waldner Reference Waldner1999, Wimmer Reference Wimmer2018). Where centralised political systems existed, states developed the infrastructural power required for public goods provision and promoted linguistic integration (Wimmer Reference Wimmer2018, p. 12). In the long run, state capacity and linguistic homogeneity supported nation-building and minimised the salience of ethnic divisions. In contrast, where centralised political authorities were absent, subsequent states lacked capacity to deliver public goods and were linguistically diverse, contributing to the politicisation of ethnicity and a challenge to national cohesion (Wimmer Reference Wimmer2018). Whereas nation-building supported the development of an autonomous and capable state, ethnic politicisation often undermined state capacity and autonomy, as the state and the resources it controlled become the focus of inter-ethnic competition (Horowitz Reference Horowitz1985, Jesudason Reference Jesudason1989, Wimmer Reference Wimmer2018).

The majority of Africa and Asia fell under European or Japanese colonial rule in the late nineteenth century, leaving contrasting legacies for post-independence states. The Japanese sought to integrate Korea and Taiwan into the metropole, investing in state administration, and agricultural and industrial production, and leaving a legacy of centralised authority and competent bureaucracy that proved vital to subsequent state-led development (Cumings Reference Cumings1984, Waldner Reference Waldner1999, Kohli Reference Kohli2004). In contrast, European colonialists in much of Africa and European-colonised Asia sought to govern and extract resources on the cheap, co-opting or creating ‘traditional authorities’ through indirect rule (Mamdani Reference Mamdani1996). European colonialism also directly contributed to the creation of ethnically heterogeneous and divided states, organising mass migration to meet labour requirements, dividing populations among ‘customary’ authorities and selectively promoting certain groups over others (Horowitz Reference Horowitz1985, Mamdani Reference Mamdani1996, Gomez and Jomo Reference Gomez and Jomo1999, Slater Reference Slater2010, Boone Reference Boone2014).

The wave of independence following the Second World War constituted another pivotal moment, providing some post-independence leaders with new opportunities to re-shape politics. Across East Asia, the Japanese withdrawal left a political vacuum, unleashing revolutionary movements that contributed to political centralisation and state-building through social revolutions or by forcing existing rulers to strengthen the state to ward off the communist threat. Slater (Reference Slater2010) shows that where unmanageable class and ethnic conflict combined, ruling elites were able to centralise political power and build particularly strong states in Malaysia and Singapore. Importantly, this shows that while ethnic diversity has often impeded political centralisation and state-building, where multi-ethnic elites face a severe threat to their collective security, this can provide the motivation to centralise power. Consequently, ethnic heterogeneity is not necessarily an obstacle to political stability or developmental progress (Slater Reference Slater2010, Stewart Reference Stewart2010, Wimmer Reference Wimmer2018). In other instances, the impulse for state-building instead came from within state structures. For example, in the likes of Egypt and Peru, modernising military-bureaucratic elites seized power and conducted land reforms aimed at destroying dominant social classes in ‘revolutions from above’ (Trimberger Reference Trimberger1978).

These multiple, overlapping processes resulted in highly divergent states in terms of their autonomy, capacity and cohesion, and consequently contrasting potential to lead the structural transformation of their economies. South Korea and Taiwan provide examples in which colonial legacies and post-independence politics greatly strengthened the state, providing an essential foundation for economic development. Building on the bureaucratic legacy of Japanese colonialism, elite ideological polarisation and mass repression led to the further centralisation of political power, resulting in capable and autonomous states in South Korea and Taiwan (Vu Reference Vu2010). Rulers in South Korea and Taiwan decisively split from landed elites, opening the way for far reaching land redistribution that wiped out powerful potential opponents.Footnote ⁴ States in Korea and Taiwan were heavily involved in primitive accumulation that supported the development of industrial capitalists through state land reform and control of the financial sector, with the result that at early stages, ‘capitalists were dependent on state largesse’ (Amsden Reference Amsden1992, p. 52, Woo Reference Woo1991). Finally, the state utilised its coercive powers to suppress the peasantry and labour, limiting collective action, extracting an agrarian surplus and limiting wage costs (Amsden Reference Amsden1992, Waldner Reference Waldner1999, Chang Reference Chang2009).

The contrast with Latin American countries such as Argentina, Brazil and Peru is instructive since contrasting histories of state formation resulted in very different states with less beneficial implications for structural transformation.Footnote ⁵ Prior to the 1930s, the landed elite continued to control politics in many Latin American countries.Footnote ⁶ Disruption to global trade during the Great Depression prompted many Latin American countries to pursue state centralisation and industrial diversification through import substitution of basic manufacturing (Kaufman Reference Kaufman, Gereffi and Wyman1990, Ross Schneider Reference Ross Schneider and Woo-Cumings1999). However, while industrial capital and foreign investors gained political influence at this time, the landed elite blocked far-reaching land reform (Thorp Reference Thorp1998, Kay Reference Kay2002, Albertus Reference Albertus2015). The concentration of landholdings meant the continuation of rural poverty and inequality, limiting the size of the domestic market, as well as impeding the transfer of an agrarian surplus to industry. By the 1950s, as the market for basic goods was saturated, governments promoted foreign investment as part of a push for secondary import substitution for capital and foreign exchange-intensive production for high-end consumers (Stallings Reference Stallings, Gereffi and Wyman1990). However, this resulted in modest employment creation, condemning many to the informal economy (Weyland Reference Weyland1996, Thorp Reference Thorp1998, Kay Reference Kay2002). While several countries, of which Brazil is perhaps the leading example, made progress with economic diversification, the political influence of industrial capital – particularly multinational firms – and labour movements prevented governments from disciplining firms and demanding the productivity improvements required for upgrading (Evans Reference Evans1995). Instead, governments pursued a cycle of debt-fuelled import substitution, contributing to the 1980s debt crisis (Weyland Reference Weyland2004).

The literatures on state-led development and African politics are, for the most part, more pessimistic still about the prospects of African states making any positive contribution to agrarian transformation or economic diversification, much less industrial upgrading. According to this body of work, pre-colonial, colonial and post-colonial state formation has undermined state autonomy, capacity and cohesion (Callaghy Reference Callaghy and Ergas1987, Herbst Reference Herbst2000, van de Walle Reference van de Walle2001). Post-independence leaders inherited ethnically diverse countries in which state power depended heavily on neo-customary authorities (Mamdani Reference Mamdani1996). Many post-independence leaders opted to maintain the status quo, rather than threatening the political order with the reforms required to build state capacity and instigate capitalist development (Boone Reference Boone1992). As such, for many, rather than rational-legal bureaucracies of Weber’s ideal type, African states are ‘neo-patrimonial’, combining the façade of the formal institutions of a modern state with the reality of highly personalised authority and informal politics (van de Walle Reference van de Walle2001, Kohli Reference Kohli2004). Unsurprisingly, based on this diagnosis, the prospects for state-led development are not good.

While many African states undoubtedly face major challenges, these analyses tend to downplay the diversity that exists within Africa, as well as the significant progress that has been made in building states capable of playing important developmental roles in certain cases (Allen Reference Allen1995, Mkandawire Reference Mkandawire2001, Boone Reference Boone2003, Whitfield et al. Reference Whitfield, Therkildsen, Buur and Kjær2015). For some critics of this literature, the prevalence of clientelism and informal institutions is as much a reflection of the absence of structural transformation and capitalist development as an indication of anything distinct about African political systems (Khan Reference Khan2010, Whitfield et al. Reference Whitfield, Therkildsen, Buur and Kjær2015). Moreover, though there are certainly examples of state collapse, and economic and political crisis, this is far from uniform across a diverse continent and an extended period since independence. Notably, past research has identified examples of ‘centralised-bureaucratic politics’ in which strong authoritarian presidents centralised control over rent allocation and built relatively strong bureaucracies in response to political crises, providing the basis for political stability and rapid growth in post-independence Côte d’Ivoire, Kenya, Malawi and Tanzania (Allen Reference Allen1995, Mkandawire Reference Mkandawire2001, Kelsall Reference Kelsall2013).Footnote ⁷

Since the turn of the Millennium, renewed interest in the role of the state in development has resulted in growing attention to countries where relatively high-capacity states have overseen periods of rapid economic development. Of particular prominence in these discussions is Ethiopia, which in many respects constitutes a leading candidate for state-led development in Africa. With a long history of a hierarchical state, Ethiopia was the only African authority that was able to centralise power and mobilise sufficient military and diplomatic resources to ward off European colonialism in the nineteenth century, defeating the Italian invasion of 1896 and, indeed, significantly extending its territory alongside the European ‘Scramble for Africa’.Footnote ⁸ Moreover, Ethiopia in the 1970s experienced one of the few social revolutions on the continent, resulting in the eradication of the landed elite and the bureaucratisation of the state. The result was a state with greatly expanded capacity and territorial reach, and autonomy from social forces. Ethiopia shared with many of its African neighbours a legacy of ethno-linguistic diversity that presented a major challenge to nation-building. However, from the perspective of existing theory, the Ethiopian state was comparatively well placed to play a role in state-led development.

Authoritarianism and Developmental Orientation

The history of state formation shapes a state’s potential to promote structural transformation. Nonetheless, these processes on their own offer little insight into why a ruling elite might tackle the politically challenging collective action problems involved in structural transformation, rather than, for example, merely using state capacity to enrich itself. Indeed, Ethiopia’s poor economic performance in the twentieth century, despite what is, in certain respects, a positive legacy of state formation, clearly underscores that other factors are at play.

The approach taken in this book builds on the common simplifying assumption that, above all, leaders want to stay in power, and that their strategy for doing so derives from their perceptions of threat to their rule (Migdal Reference Migdal1988, Geddes Reference Geddes1994). Where ruling elites face a severe collective threat, they are likely to put aside personal interests and focus on their collective political survival. Under certain circumstances, the ruling elite’s response to collective threats will align with the mobilisation of existing state capacity and investment in further state strengthening to promote structural transformation. For many early industrialisers, foreign threats provided the impulse for such elite collective action focused on strengthening the economy and retaining sovereignty (Skocpol Reference Skocpol1979). Meanwhile in late developing East Asia, a combination of external military and internal revolutionary threats led elites to focus on structural transformation (Doner et al. Reference Doner, Ritchie and Slater2005). The consolidation of the nation-state system following the Second World War has, by and large, reduced foreign threats. As such, severe domestic political threats – notably class and ethnic divisions rooted in histories of state formation – have been increasingly important as a motivation for elite collective action (Huntington Reference Huntington1968, Slater Reference Slater2010). In particular, when authoritarian rulers face mass distributive pressures in the context of extreme resource shortages, this can provide the incentive to pursue state-led development in order to expand the resources required to secure mass compliance (Doner et al. Reference Doner, Ritchie and Slater2005, Doner Reference Doner2009).

This argument can be elaborated as follows. Rulers typically seek to consolidate power in the most straightforward way by forming alliances with and distributing resources to local elites who can be relied on to maintain order, for example the landed class in an agrarian economy, or ethnic and communal elites in multi-ethnic societies (Horowitz Reference Horowitz1985, Waldner Reference Waldner1999, Boone Reference Boone2003, Albertus Reference Albertus2015). It is only where such straightforward options are unavailable that leaders seek to secure direct control over the masses. Frequently, such a strategy is the result of splits between the ruling coalition and socio-economic elites due to divergent economic interests or social background, with states seeking to usurp socio-economic elites by establishing direct control over the popular classes (Trimberger Reference Trimberger1978, Boone Reference Boone2003, Albertus Reference Albertus2015, Albertus et al. Reference Albertus, Fenner and Slater2018). In certain situations, however, ruling and socio-economicelites may face sufficiently severe collective threats that they coordinate in the pursuit of mass control and the maintenance of political order (Slater Reference Slater2010).

In either case, this raises the question as to how states can secure mass compliance. Following Etzioni (Reference Etzioni1975) and Slater (Reference Slater2010), rulers draw on symbolic, coercive and distributive state powers.Footnote ⁹ For incumbents, the most desirable situation is one in which leaders can rely on normative appeals and symbols that underpin legitimate authority, reducing the need to employ coercive or distributive powers. Historically, leaders would invoke the divine right of kings and religion to enhance legitimacy, while subsequent regimes have turned to the symbolic powers of nationalism, Marxist-Leninism or anti-communism to legitimate authoritarian rule. Coercion, meanwhile, is central to state power and, from a Weberian perspective, a defining feature of a state. The ability to deploy physical threats and sanctions can be vital means by which regimes suppress opponents. Coercion may take high visibility forms – arresting political organisers or dispersing protests – or low-visibility and low-intensity forms including surveillance, harassment or denial of services that prevent the emergence of mass opposition in the first place (Levitsky and Way Reference Levitsky and Way2010). While symbolic and coercive powers are both important, however, many authoritarian regimes persist for years without widespread legitimacy, while rulers cannot rely on coercion exclusively and indefinitely (Skocpol Reference Skocpol1979, Geddes Reference Geddes1999, Migdal Reference Migdal2001, Albertus et al. Reference Albertus, Fenner and Slater2018). Instead, the state’s distributive powers are essential to secure the mass acquiescence on which authoritarian regimes depend. Indeed, where rulers believe that their political survival depends on broad-based distribution, yet available resources are insufficient to meet these requirements, the ruling elite must act collectively in the pursuit of state-led development to enable mass distribution (Doner et al. Reference Doner, Ritchie and Slater2005, Doner Reference Doner2009).

This basic dynamic of developmental impulse as a response to distributive pressures can be deepened by incorporating the concept of ‘coercive distribution’ (Albertus et al. Reference Albertus, Fenner and Slater2018). From this perspective, certain authoritarian regimes seek mass incorporation through ‘far-reaching distributive policies that deliver basic necessities to the vast majority of the populace’, frequently, though not always, displacing rival socio-economic elites in the process (Albertus et al. Reference Albertus, Fenner and Slater2018, p. 4). In such circumstances, distribution is a top-down, pre-emptive initiative by rulers seeking to entrench their power, rather than a response to interest group demands. Rulers utilise coercive distribution to ‘enmesh’ the masses in relations of dependence, requiring the expansion of state infrastructural power and the re-routing of people’s survival strategies through these state structures.Footnote ¹⁰ Distribution is intentionally broad-based, perhaps even universal, with a view to enmeshing as broad a section of the population as possible.Footnote ¹¹ The result is that with the masses dependent on the authoritarian state for ‘their basic material necessities, their ability to individually defy or collectively mobilize against their rulers is severely compromised’ (Albertus et al. Reference Albertus, Fenner and Slater2018, p. 2).

In an agrarian economy, in which land is the principal productive resource, land redistribution is the clearest example of coercive distribution, with many authoritarian regimes expropriating the landed elite and enmeshing the peasantry through the allocation of usufruct rights on state-owned land (Albertus et al. Reference Albertus, Fenner and Slater2018, Albertus Reference Albertus2021). However, land reform cannot ensure mass compliance indefinitely. For smallholder farmers to use the land productively and translate this into improved livelihoods, they also require agricultural extension services and market access. Furthermore, population growth, diminishing returns in agriculture and urbanisation progressively undermine the redistributive potential of land reform (Platteau Reference Platteau and Dercon2005, Reinert Reference Reinert2007, Boone Reference Boone2014). Coercive distribution therefore requires a process of ‘upkeep’, resulting in continuous pressure for the expansion of distribution – in addition to or instead of land access – routed through state structures in order to maintain mass acquiescence (Albertus et al. Reference Albertus, Fenner and Slater2018).

Where the ruling elite views mass distribution as essential to its own political survival, and the absence of easy revenues – such as those provided by natural resources – limits the regime’s ability to finance large-scale spending, there are strong incentives for leaders to pursue economic development to grow the pie, enabling the distribution that can maintain popular compliance (Doner et al. Reference Doner, Ritchie and Slater2005, Doner Reference Doner2009). Following land reform, this distribution is initially likely to take the form of agricultural extension services and improved agricultural inputs to raise productivity. However, the limits to agricultural production will inevitably necessitate a shift, first, to economic diversification and, subsequently, industrial upgrading as means of providing mass employment opportunities and improved living standards, alongside services such as urban housing, education and healthcare. To effectively support regime maintenance, each of these resources must be distributed through state and party structures, cultivating mass dependence on the regime. From this perspective, then, the commitment of authoritarian rulers to state-led development is inseparable from their efforts to ensure their own political survival by averting mass unrest. This basic causal sequence is outlined in Figure 2.1.

Figure 2.1

The political drivers of state-led development

This causal process can be illustrated with reference to some of the most successful instances of late development in East Asia. South Korea and Taiwan faced both a severe military threat from neighbours in North Korea and mainland China, respectively, and the major internal threat of communist revolution. Enabled by a split from the landed elite and spurred by the threat posed by the peasantry, ruling elites in South Korea and Taiwan conducted sweeping land reform in the post-war years which enabled them to consolidate control over the peasantry (Wade Reference Wade1990, Amsden Reference Amsden1992, You Reference You2015). To that end, leaders built state and party structures that established direct relations with smallholders, and used these to extend additional forms of distribution such as improved inputs and extension services required to raise agricultural productivity but also cultivate peasant dependence (Amsden Reference Amsden1979, Wade Reference Wade1983, Reference Wade1990, Byres Reference Byres, Breman and Mundle1991).Footnote ¹² Similar dynamics underpinned industrialisation, with ruling elites motivated by ‘fear … of a resurgence of political unrest and labor militance’, resulting in the promotion of labour-intensive industry to create employment and state pressure on firms to bargain wage increases for productivity improvements, driving industrial upgrading (Amsden Reference Amsden1992, p. 208, Doner et al. Reference Doner, Ritchie and Slater2005). Moreover, the state invested in ‘productivist’ social policies, providing mass distribution, but also prioritising education and health for their contribution to a well-trained and healthy labour force, and employment-linked pensions used to provide additional resources for state investment (Holliday Reference Holliday2000, Gough Reference Gough, Gough and Wood2004, Kwon Reference Kwon2004). The result is that ‘while popular sectors have been politically subordinated and even brutally repressed … they have rarely been economically ignored’ (Doner et al. Reference Doner, Ritchie and Slater2005, p. 331).

Mass distributive pressures also shaped the Malaysian government’s focus on agriculture and economic diversification. Growing frustration amongst the Malay population at inter-ethnic inequality and a communist insurgency fundamentally shaped the early independence period from 1957. Upheaval was perceived as a collective threat to ruling and socio-economic elites resulting in ‘a race to enmesh the masses’ but in this case ‘not to eradicate feudal elite rivals’ (Albertus et al. Reference Albertus, Fenner and Slater2018, p. 66). The United Malays National Organisation (UNMO) that dominated the ruling coalition sought to expand control of the Malay rural population by investing in infrastructure, irrigation and agricultural inputs (Scott Reference Scott1985, Gomez and Jomo Reference Gomez and Jomo1999). UNMO avoided land redistribution, which would have brought them into conflict with an important source of support amongst large landholders. However, the open land frontier enabled large-scale land allocations to smallholders (Albertus et al. Reference Albertus, Fenner and Slater2018, p. 67). Rising agricultural productivity benefitted both large capitalist farmers and smallholders, consolidating support among capitalist farmers, control over smallholders and achieving rice self-sufficiency to support industrialisation (Scott Reference Scott1985, Wong Reference Wong1987, Henley Reference Henley2012). Similar political pressures underpinned the state’s push for economic diversification from the early 1970s. Labour-intensive manufacturing was prioritised to meet the growing demands of the majority Malay population and smooth inter-ethnic tensions (Jesudason Reference Jesudason1989, Gomez and Jomo Reference Gomez and Jomo1999, Welsh Reference Welsh and Huber2002). The result was one of ‘the world’s most impressive cases of export-led growth through low-cost manufacturing’ (Doner et al. Reference Doner, Ritchie and Slater2005, p. 354).

Once again, the contrast between these East Asian leaders’ commitment to national development and the standard portrayal of African leaders is stark. Rather than the pursuit of state-led development in response to an existential threat, African rulers are presented as lacking in ‘political discipline’ (van de Walle Reference van de Walle2001, p. 14), ‘predatory’ (Evans Reference Evans1995) and motivated by the ‘politics of the belly’ (Bayart Reference Bayart1993). From this perspective, politics has become a winner takes all struggle amongst political elites to control state resources and particularly the foreign aid receipts and external trade that can be used for self-enrichment and ‘systematic clientelism’ (Bratton and van de Walle Reference Bratton and van de Walle1997, p. 65, De Waal Reference De Waal2015). While much of the African politics literature attributes these failings to the venal tendencies of rulers themselves, a number of studies focus on the structural constraints that shape rulers’ behaviour. From this latter perspective, the challenge facing many African countries is not necessarily poor political leadership, but an acute form of the ‘politician’s dilemma’ whereby strategies to address short-term political imperatives undermine developmental impulse (Geddes Reference Geddes1994, Migdal Reference Migdal2001). Post-independence leaders often faced broadly similar challenges to their colonial predecessors in trying to project state authority over national territory. Frequently they arrived at similar political strategies, relying on neo-customary authorities to maintain order, rather than displacing these intermediaries (Mamdani Reference Mamdani1996, Boone Reference Boone2003). Moreover, in their efforts to consolidate control over the state and its resources, leaders have often marginalised weak capitalists and undermined their incentives to invest and upgrade production (Boone Reference Boone1992, Whitfield et al. Reference Whitfield, Therkildsen, Buur and Kjær2015). For the most part, the revolutionary movements and foreign threats that prompted many East Asian leaders to undertake sweeping reforms have been lacking in Africa. Instead, rebel movements in Africa have been launched primarily in response to urban crises and have struggled to build support amongst the peasantry (Mamdani Reference Mamdani1996, Mkandawire Reference Mkandawire2002).

Ethiopia under the EPRDF was therefore relatively unusual in facing a somewhat comparable set of incentives to many East Asian regimes. The TPLF originated in the student movement of the 1970s and spent nearly two decades mobilising the Tigrayan peasantry as part of a Maoist insurrection, forming the EPRDF coalition in the late 1980s and fighting its way to national power in 1991. While the EPRDF forces were militarily dominant by the time it took power, the EPRDF lacked a solid support base outside Tigray. The landed elite had been destroyed by the land-to-the-tiller reforms of the 1970s, while most non-Tigrayan ethnic leaders were either sceptical of or outright resistant to the new Tigrayan-dominated government, with the result that regime consolidation through the co-optation of existing elites was infeasible. As such, the EPRDF sought to build a ‘coalition with the people’ (Vaughan Reference Vaughan2011, p. 619), pursuing a pre-emptive strategy to consolidate a political base amongst the ethnically diverse peasantry, by far the most numerous social group in the country. To do so, the EPRDF built on past land reforms and expanded mass provision of agricultural extension and social services. All these resources were distributed through an extensive network of party-state structures and thereby used to tie the people to the party-state.

From very early on, moreover, the EPRDF recognised the limitations of land and agriculture as a distributive strategy, with rapid population growth inevitably eroding this system of enmeshment. Consequently, the government prioritised industrial development, with a view to expanding employment opportunities, absorbing rural surplus labour and enabling an urban transition without disruption to political order. This project of state-led development was wrapped in a nationalist narrative of an Ethiopian Renaissance as a counter-balance to growing ethno-nationalism. The result was a relatively rare instance in which an African authoritarian ruling party viewed structural transformation as imperative to its political survival. The first part of this book elaborates this argument in detail. Chapter 3 examines Ethiopian state formation and how this contributed to the creation of a relatively autonomous state with some capacity for state-led development, and the revolutionary struggle that brought the TPLF/EPRDF to office. Chapter 4, meanwhile, turns to the political dynamics that shaped the EPRDF while in power and how the intersection of class and ethnicity contributed to elite vulnerability, and thereby the necessity of broad-based development.

Contrasting Explanations of State-Led Development

This theoretical argument builds on existing theories of state-led development, as should be clear from the works cited above, but expands on and diverges from them in certain respects. First, the approach differs from those who place a central emphasis on developmental ideology or political leadership in explaining the orientation of political leaders (Leftwich Reference Leftwich2000, Soifer Reference Soifer2015). A focus on the structural constraints facing rulers is essential to avoid an overly voluntaristic approach that may overstate the benign influence of, for example, Park Chung-hee in Korea and Lee Kuan Yew in Singapore, or, indeed, Ethiopia’s Meles Zenawi or Rwanda’s Paul Kagame. Ultimately, however visionary a leader may be, if the political context in which they operate is not conducive to their vision of development, they will be unable to realise their goals. This analytical focus on structural factors does not preclude acknowledgement of an important role for individual leaders and their political ideas. While elite threat perceptions can provide the incentives for leaders to pursue state-led development, this does not necessarily mean that they will do so. In the context of these structural constraints, leaders can play vitally important roles, setting out a vision for national development and the role of the state in it, or indeed undermining developmental prospects by pursuing strategies that undermine productive incentives. The approach pursued in this book, therefore, seeks to ground analysis of individual leaders within the structural factors that shape their incentives to pursue state-led development.

A focus on mass distributive pressures also contrasts with much of the literature that argues that a key factor enabling state-led development in East Asia was the suppression of the popular classes, enabling governments to focus on long-term investments rather than short-term consumption (Deyo Reference Deyo, Gereffi and Wyman1990, Waldner Reference Waldner1999, Haggard Reference Haggard2018). The implication of some of this work is that all redistribution is necessarily detrimental to structural transformation. But as Doner et al. (Reference Doner, Ritchie and Slater2005) argue, while redistribution can add to labour costs and undermine industrial competitiveness, certain forms of redistribution are perfectly compatible with economic development. For example, land reform can enhance productivity while placing the means of production in the hands of the peasantry (Albertus Reference Albertus2021); investment in education, health and pension schemes were part of a distinctive ‘productivist’ social policy across much of East Asia (Holliday Reference Holliday2000, Gough Reference Gough, Gough and Wood2004, Kwon Reference Kwon2004); housing in Singapore contributed to mass compliance while subsidising labour costs and industrial competitiveness (Trocki Reference Trocki2006); and bargaining wage increases for productivity improvements can contribute to upgrading and rising living standards (Amsden Reference Amsden1992). While it is certainly true that the peasantry and working class were suppressed in many East Asian countries, such coercion is supplemental, rather than contradictory, to the pre-emptive distributive efforts of an authoritarian regime seeking mass compliance.

Another contrasting claim in the literature is that authoritarian regimes in East Asia sought to build performance legitimacy through rapid development in place of the process legitimacy of competitive elections (Castells Reference Castells, Appelbaum and Henderson1992, Leftwich Reference Leftwich2000, Campos and Root Reference Campos and Root2001, p. 3, Mkandawire Reference Mkandawire2001). While this focus on growth and distribution certainly resonates with the theory outlined above, it also raises several questions. Legitimacy is a notoriously difficult concept to pin down or to measure, and it is far from clear that many East Asian regimes or Ethiopia’s EPRDF were necessarily considered legitimate by the majority of their populations. Indeed, many authoritarian regimes persist for extended periods without popular legitimacy (Skocpol Reference Skocpol1979, Geddes Reference Geddes1999). Rather, what authoritarian regimes require is the active support of an influential minority and mass acquiescence. A focus on resource distribution and the way in which this binds the masses to the regime offers a much more tangible explanation for mass acquiescence than a general appeal to legitimacy.

As is no doubt evident from the preceding discussion, Doner, Ritchie and Slater’s concept of ‘systemic vulnerability’ has been particularly influential on the theoretical approach pursued in this book (Doner et al. Reference Doner, Ritchie and Slater2005, Doner Reference Doner2009). Doner et al. (Reference Doner, Ritchie and Slater2005, p. 328) hypothesise that states succeed in industrial upgrading when ruling elites face mass distributive pressures amidst extreme resource constraints, requiring leaders to take action to grow the pie in order to avoid ‘unmanageable mass unrest’. Moderate distributive pressures may cause leaders to invest in agricultural development and economic diversification, as in Indonesia, Malaysia and Thailand. However, it is only when ruling elites in Singapore, South Korea and Taiwan were faced with ‘systemic vulnerability’ – mass distributive pressures, severe resource constraints and an external threat that necessitated military spending – that they successfully tackled economic upgrading. Despite important insights regarding elite threat perceptions and mass distributive pressures, this theory – as outlined in a rather short paper – remains ambiguous in certain respects and is in need of elaboration for present purposes. While placing causal emphasis on ‘popular pressures’, Doner et al. are rather ambiguous about the origins of these pressures, their nature and what forms of distribution are required to address them. First, the existence of mass pressures is taken for granted, rather than situating the need for mass incorporation in the context of elite splits or collective threats. Second, reference to ‘restive popular sectors’ (Doner et al. Reference Doner, Ritchie and Slater2005, p. 327) suggests the importance of popular mobilisation and redistributive demands. However, as others have argued, heavily suppressed popular sectors in South Korea and Taiwan were not particularly vocal in making redistributive demands (Deyo Reference Deyo, Gereffi and Wyman1990, Haggard Reference Haggard1990, Amsden Reference Amsden1992, Albertus et al. Reference Albertus, Fenner and Slater2018). Rather state distribution was primarily ‘a supply-driven campaign for regime consolidation and social control, not a demand-driven attempt to curry popular favour’ (Albertus et al. Reference Albertus, Fenner and Slater2018, p. 55). Third, systemic vulnerability is also somewhat ambiguous regarding the outcome it seeks to explain. The paper is presented as a means of explaining why some countries succeeded in industrial upgrading, while others did not. Yet, case studies on South Korea and Taiwan also present systemic vulnerability as an explanation for land reforms conducted in the early post-war period, long before any attempt at upgrading (Doner et al. Reference Doner, Ritchie and Slater2005, pp. 341–342). Doing so raises the question as to whether this is a theory of industrial upgrading or development as a whole. Consequently, this ambiguity leads to uncertainty as to what resources are required to meet distributive pressures and at what points in the process of structural transformation. To address this gap, the final section of this chapter conceptualises the three main resources that are the focus of analysis in this book, namely the key resources required for the survival strategies of the majority of the population at different stages of structural transformation: access to land, employment and social protection.

Finally, the theoretical approach outlined above shares important commonalities with the growing body of work on ‘political settlements’, particularly in research on African political economy. Building on Mushtaq Khan (Reference Khan2010), authors have adapted Political Settlements theory to explain patterns of industrial policy, structural transformation and a range of developmental outcomes (Whitfield et al. Reference Whitfield, Therkildsen, Buur and Kjær2015, Gray Reference Gray2018, Khan Reference Khan2018, Kelsall et al. Reference Kelsall, Schulz, Ferguson, Hau, Hickey and Levy2022, Hickey and Sen Reference Hickey and Sen2023). This body of work directs attention to the underlying settlement or order that provides political stability and shapes elite incentives for contrasting development strategies. This political settlement, in turn, is based on the balance of power between contending groups within society and, in particular, the degree of cohesion and vulnerability within the ruling elite. Though sharing many points in common, the approach taken in this book diverges in certain respects from much work on political settlements. First, for Khan, and many others, elite vulnerability is considered detrimental to development since vulnerability prompts rulers to distribute rents to secure short-term survival rather than promoting long-run development (Khan Reference Khan2010).Footnote ¹³ This, of course, contrasts with the focus above on mass distributive threats as an important driver of elite collective action. Second, this body of work pays little attention to mass politics, with the main consideration being the extent to which masses can be mobilised by contending elites in their struggle for power. The control of popular classes and the threat they pose to ruling elites is, in contrast, a central concern of the theoretical approach taken here.Footnote ¹⁴ Finally, political settlements theory focuses analytical attention primarily on the patterns of recent and contemporary elite bargaining. While undoubtedly important, the analysis tends to pay much less attention to the long run history of state and nation-building, which are key factors in the analysis in this book.

Globalisation and Late-Late Development

The literature on state-led development suffers from a degree of ‘methodological nationalism’, with most studies emphasising characteristics of the state, industrial policy and the predominately domestic political dynamics that shape these. Yet state-led development has always been shaped by global economic processes, and the global economy has changed in important ways in recent decades that fundamentally re-shape opportunities for structural transformation of the sort undertaken in East Asia.

An important feature of the contemporary global economy is the fragmentation of industrial production into global value chains that make up an integrated global system of production and division of labour (Gereffi Reference Gereffi2005, Kaplinsky Reference Kaplinsky2013). The offshoring and outsourcing of low technology, low-wage activities by multinational firms began in the 1960s, but accelerated since the 1990s as a result of the massive expansion of the global workforce due to: population growth and the integration of former socialist countries into the capitalist economy from the early 1990s; technological changes, particularly in communications, that enable the development of an integrated global system of production; and increased capital mobility resulting from economic liberalisation since the 1970s.

Trade agreements are an important factor shaping the structure of value chains and the pattern of industrial production. South Korea and Taiwan’s initial steps in the manufacture of textiles and apparel were aided by preferential access to US markets, while protectionist quotas and tariff restrictions on imports to Europe and North America subsequently provided incentives for Korean and Taiwanese firms to outsource production to countries with lower labour costs and available quotas, such as Bangladesh, China and Vietnam (Gereffi Reference Gereffi2018). In the last two decades, the United States’ African Growth and Opportunity Act (AGOA) and the European Union’s Everything But Arms agreement have provided preferential quota and tariff-free access to a range of producers based in African and least developed countries, respectively. The result is that many countries have new opportunities to engage in export-oriented light manufacturing.

However, global value chains have important implications for countries taking initial steps in industrialisation. The tendency has been for lead firms in global value chains, based primarily in the advanced economies, to retain control of design, branding and marketing, where most value is concentrated, outsourcing the least profitable, labour-intensive manufacturing activities (Gereffi Reference Gereffi, Gereffi and Korzeniewicz1994, Kaplinsky Reference Kaplinsky2013). Outsourced activities utilise widely available technology, with little potential for innovation, resulting in low-profit margins and high competition based on low-wage jobs that offer little potential to raise living standards. This is particularly the case for low-technology sectors, including apparel, which tend to be the first step into manufacturing for industrialising countries (Gereffi Reference Gereffi2005, Kaplinsky Reference Kaplinsky2013). Moreover, the fragmentation of production into value chains has meant that initial participation in low-technology assembly activities does not necessarily provide a firm foothold to expand into the rest of the value chain or to establish domestic production linkages, as has often been the case in the past (Reinert Reference Reinert2007, Kaplinsky Reference Kaplinsky2013, Whitfield et al. Reference Whitfield, Therkildsen, Buur and Kjær2015).

While most of the value chain literature focuses on manufacturing, similar processes shape some agro-industry, in particular horticulture and floriculture. European, US and, increasingly, also developing country supermarkets and buyers operate as lead firms in buyer-driven value chains, shaping what is to be produced and how (Dolan and Humphrey Reference Dolan and Humphrey2000, Barrientos et al. Reference Barrientos, Dolan and Tallontire2003, Reference Barrientos, Knorringa, Evers, Visser and Opondo2016). Integration into these value chains for producers, ranging from foreign and domestic investors hiring wage labour to smallholders working as contract farmers, can provide market access and new economic opportunities. However, relationships within the value chain are characterised by inequality, favouring lead firms, which demand high quality and low cost. The result is a tendency to favour better off farmers able to invest in meeting exacting production standards and to impose downward pressure on wages and working conditions (Barrientos Reference Barrientos2008).

Linked, in part, to the changing nature of global production systems is the changing role for foreign and domestic firms in structural transformation. For early industrialisers in Europe and North America, and late industrialisers in East Asia, manufacturing was dominated by domestic capitalists, with states actively promoting learning to acquire cutting edge technology, and providing incentives for firms to upgrade production to reach global competitiveness (Chang Reference Chang2004). Prior to the Second World War, domestic firms in late developing economies in East Asia and elsewhere had already acquired some manufacturing expertise on which post-war state-led efforts could build, whether through colonial influence in Korea and Taiwan or émigré firms elsewhere (Eckert Reference Eckert1991, Amsden Reference Amsden2001). In contrast, and almost by definition, late-late developing countries currently attempting to promote industrialisation face the significant challenge that they lack domestic capitalists with manufacturing expertise (Amsden Reference Amsden2001). Nowhere is this more the case than in Africa where colonial regimes prevented the emergence of an indigenous capitalist class and where post-war industrialisation efforts tended to produce heavily protected and inefficient state enterprises (Berman and Leys Reference Berman and Leys1994, Whitfield et al. Reference Whitfield, Therkildsen, Buur and Kjær2015). In a context of late-late development, the productivity frontier for global competition has advanced so far – even in low-value, low-productivity roles that tend to be outsourced – that it is extremely difficult for emerging domestic firms to acquire the technological capabilities required to compete (Whitfield et al. Reference Whitfield, Therkildsen, Buur and Kjær2015, Whitfield and Staritz Reference Whitfield and Staritz2021). Moreover, lead firms in global value chains tend to rely on existing suppliers that can be trusted to meet demanding requirements for quality and flexibility, making it very hard for new firms to break in and secure market access. The result is that rather than nurturing domestic firms to participate in global production, as with the likes of South Korea and Taiwan, later examples of industrialisation have all involved states targeting foreign investors and providing these foreign firms with incentives to relocate part of their production (Mortimore and Vergara Reference Mortimore and Vergara2004).

Foreign Direct Investment (FDI) can play a vitally important role in industrialisation, as an initial source of capital, as well as providing access to cutting edge technology, training, managerial expertise and market access. Indeed, FDI – attracted by low taxes and export processing zones – is credited with playing key roles in economic diversification, employment creation and export promotion across East Asia (Welsh Reference Welsh and Huber2002, Doner Reference Doner2009). Yet, industrialisation through FDI promotion also risks attracting footloose investors that operate as enclaves to take advantage of low labour costs, importing inputs for low-productivity assembly and thereby failing to establish linkages to the rest of the economy. Moreover, attracting globally competitive, capital-intensive production may have muted effects on employment creation and therefore do little to absorb surplus labour (Doner Reference Doner2009, p. 13). Instead, it is by establishing linkages between FDI and domestic firms that FDI has the potential to make contributions to economic upgrading, mass employment creation and rising living standards. The comparative experiences of northeast and southeast Asia are instructive. South Korea and Taiwan strictly controlled capital inflows to promote domestic industry. Where FDI was allowed, it was directed to new priority industrial sectors through joint ventures with domestic firms, while local content requirements promoted transfer of technology and managerial expertise (Wade Reference Wade1990, Chang Reference Chang2004, Hauge Reference Hauge2019). In contrast, Malaysia’s semiconductor industry focused for decades on downstream assembly work for multinational firms, with few linkages to domestic capital (Jesudason Reference Jesudason1989, Gomez and Jomo Reference Gomez and Jomo1999, Welsh Reference Welsh and Huber2002, Doner Reference Doner2009).

The state therefore has a key role, not just in encouraging FDI inflows, but also supporting the development of domestic capitalists; actively promoting linkages between foreign firms and domestic suppliers; focusing on market access, technology transfers and training; and promoting clusters of industries that facilitate linkages (Lall and Narula Reference Lall and Narula2004, Doner Reference Doner2009). As Doner et al. (Reference Doner, Ritchie and Slater2005) argue, the state is more likely to be successful when regimes perceive that success in industrial upgrading, and the mass distributive resources it offers, is essential to regime survival. Yet, it would also be an exaggeration to conclude that the sole difference between Korea, which has succeeded in industrial upgrading, and Malaysia, which has not, is the intensity of the threats faced by those regimes. The changing global economy has also played a role, with the challenge involved in nurturing domestic firms to acquire technological capabilities and integrating with global value chains only increasing over time. The result has been that all ‘the later entrants into globalised systems, from Malaysia to Mexico and Costa Rica, have gone the FDI route’ of targeting lead transnational firms and encouraging them to relocate parts of their production to the host country rather than nurturing domestic capital (Lall and Narula Reference Lall and Narula2004, p. 458).

These changes in the structure of the global economy therefore have important implications for ruling elites pursuing structural transformation as a means of meeting mass distributive pressures. Put simply, it is increasingly hard for states to deliver mass employment and rising living standards through reinforcing processes of agrarian transformation, economic diversification and upgrading. Whereas the generation and extraction of an agrarian surplus was a key factor driving primitive accumulation in the past, global capital mobility now means that agrarian transformation is not essential for industrialisation (Bernstein Reference Bernstein2004, Reference Bernstein, Akram-Lodhi and Kay2009). The potential linkages between agriculture and industry are therefore likely to weaken, with manufacturing dependent on FDI rather than agrarian transformation, while manufacturing is largely export oriented rather than providing agricultural inputs. For Bernstein, the agrarian question of capital – the contribution of agriculture to primitive accumulation – has been resolved through access to global capital. What is left is the agrarian question of labour, namely whether capitalism is capable of ‘generating sufficient, and sufficiently secure, employment to provide a living wage to the great majority’ (Bernstein Reference Bernstein, Akram-Lodhi and Kay2009, p. 251). The changing nature of the global economy presents serious concerns in this regard. Industrialisation through global value chains and FDI promotion frequently results in modest employment creation. Moreover, intense competition in low-productivity sections of value chains results in pressure from lead firms on price and flexibility that translates into low pay, long working hours, precarious working conditions and repression of worker organisations (Anner Reference Anner2020). Consequently, diversification into manufacturing does not necessarily translate into improved social outcomes (Barrientos et al. Reference Barrientos, Gereffi and Rossi2011, Anner Reference Anner2020).

The EPRDF’s attempts to promote industrialisation as a means of creating mass employment and maintaining mass compliance illustrate some of these challenges, as discussed in Chapter 6. Early on the EPRDF identified a pressing need for structural transformation in response to rapid population growth and the inability of agriculture to absorb surplus labour indefinitely. Concerted efforts to promote industrialisation from the early 2000s focused on domestic firms – both private and those owned by party and state – to produce initially for the domestic market and subsequently to move into exports. However, these firms and the state agencies responsible for designing industrial policy incentives struggled to build the technological capabilities required for global competition and to secure market access in global value chains (Whitfield and Staritz Reference Whitfield and Staritz2021). Latterly, in an attempt to accelerate industrial development and generate desperately needed employment, the government changed track to an approach more in line with the structure of the contemporary global economy, building industrial parks in an attempt to attract foreign investors. Despite major achievements in attracting some leading global firms, progress in employment creation was far too limited and too late to address what was by the mid-2010s a growing distributive crisis.

Demographic Transitions and Late-Late Development

The changing nature of the global economy presents challenges for incumbents seeking to use employment-intensive growth to expand the resources available for distribution and thereby maintain political control of the masses. However, late-late industrialising countries concurrently face the challenge of rapid demographic growth that results in particularly intense distributive pressures.

Early industrialisers undertook structural transformation alongside a demographic transition from high fertility and high mortality to low fertility and low mortality. In the first countries to undergo this transition, changes were gradual, with progressive improvements in public health and sanitation leading to gradual reduction of mortality, which subsequently prompted a gradual reduction in fertility. The faster decline in mortality than fertility led to population growth (Dyson Reference Dyson2013, Livi‐Bacci Reference Livi‐Bacci2017). The demographic transition in many countries in the global south has taken a distinct form. Following the Second World War, there has been a marked and more rapid decline in mortality due to technology transfer and improved public health. Fertility rates, in contrast, have declined more slowly, resulting in a far more rapid and large-scale population increase (Dyson Reference Dyson2011, Reference Dyson2013, Livi‐Bacci Reference Livi‐Bacci2017, Benanav Reference Benanav2019). Indeed, developing countries have had peak annual population growth rates more than double those of the advanced economies that underwent the demographic transition earlier (Livi‐Bacci Reference Livi‐Bacci2017, pp. 167–169). The result is that there are currently around three billion people aged under 25 in the world, of whom 89 per cent live in the Global South. This is more than entire world’s population in 1950 (Benanav Reference Benanav2014, p. 105).

While declining mortality is, of course, a great success, the economic and political ramifications of rapid population growth are mixed. A large population offers the potential benefit of a large domestic market that can be an important factor supporting industrialisation through economies of scale. Moreover, declining fertility rates following earlier decline in mortality leads to a temporary increase in the working age population. Many East Asian countries benefitted from this ‘demographic gift’ of low dependency ratios at a key stage of industrialisation (Bloom and Williamson Reference Bloom and Williamson1998). However, the demographic dividend resulting from a large working age population is far from automatic. Even in the first countries to undergo the demographic transition – where more modest population growth coincided with industrialisation – international migration to the ‘new world’ provided an important means of relieving population pressure (Amsden Reference Amsden1992, Livi‐Bacci Reference Livi‐Bacci2017). In contemporary developing countries, population growth is larger and opportunities for international migration considerably more limited than in the past. While some have expressed optimism that African economies may soon benefit from a demographic dividend along East Asian lines (Garcia and Fares Reference Garcia and Fares2008), this would require creation of sufficient jobs to absorb the youth bulge. Thus far, the dominant picture in much of Africa is instead one of ‘catastrophic youth unemployment’, informality and vulnerability, with the result that this young population may be ‘more of a threat than an opportunity’ (Meagher Reference Meagher2016, pp. 483–484, Benanav Reference Benanav2019).

These demographic changes also have important implications for urbanisation. Based on the earlier European experience in which industrialisation took place in parallel with urbanisation, the common assumption has long been that urbanisation was a product of this economic transformation.Footnote ¹⁵ Yet the experiences of Latin America and Africa, where urbanisation has rapidly outpaced industrialisation, have led to the contrasting conclusion that urbanisation is primarily a product of the demographic transition and that economic growth only moderately influences the pace of urbanisation (Dyson Reference Dyson2001, Reference Dyson2013, Fox Reference Fox2012, Gollin et al. Reference Gollin, Jedwab and Vollrath2016, Benanav Reference Benanav2019). The main historical barriers to urban expansion were poor public health that resulted in mortality exceeding fertility rates, and limited supplies of food and fuel. Improvements in public health, agriculture and energy around the industrial revolution enabled large-scale urbanisation first through rural-urban migration and then urban population growth. For early industrialisers, industrialisation at around the same time went some way to creating employment for this growing urban population. The experience of urbanisation in late developing countries has, again, diverged from that of the ‘early urbanisers’ with more rapid and large-scale population growth resulting in far more rapid and large-scale urban expansion that has outpaced industrialisation (Dyson Reference Dyson2013, Benanav Reference Benanav2019, Fox and Goodfellow Reference Fox and Goodfellow2021).

The combination of the growing difficulty of employment creation through industrialisation in the contemporary global economy, and an increasingly populous and urbanised society therefore presents particular challenges for regimes seeking to maintain power through mass distribution, as outlined in Figure 2.2. For late-late developing countries, even where domestic political drivers align with the promotion of state-led development, population growth places intense pressure on available land resources and employment opportunities, while the global economy constrains the ability of states to deliver mass employment through industrialisation. Structural transformation may not therefore automatically follow from elite efforts to pursue state-led development.

Figure 2.2

The distributive challenge of late-late development

The question then arises as to what happens when a regime that has sought to consolidate its power through coercive distribution fails to generate the resources needed to meet these distributive pressures? Under what circumstances will failure to meet mass distributive pressures lead to regime collapse, and when can regimes turn to other measures, such as repression, to maintain power? The literature on authoritarian durability suggests that while distribution is of vital importance, it is not necessarily the case that unmet distributive demands lead to regime change. Where there is a high degree of elite cohesion and authoritarian regimes draw on the full powers of the state, incumbents are likely to be able to resist even powerful opposition movements pressing for their removal for a period. Instead, splits within the ruling elite that erode the elite’s ability to mobilise these powers almost always precede regime collapse (Skocpol Reference Skocpol1979, O’Donnell et al. Reference O’Donnell, Schmitter and Whitehead1986, Haggard and Kaufman Reference Haggard and Kaufman1997, Levitsky and Way Reference Levitsky and Way2010, Slater Reference Slater2010).

A common finding in this literature is that strong political parties are key to authoritarian strength, while powerful parties in control of strong states tend to be the most durable authoritarian regimes (Geddes Reference Geddes1999, Brownlee Reference Brownlee2007, Levitsky and Way Reference Levitsky and Way2010, Slater Reference Slater2010, Lachapelle et al. Reference Lachapelle, Levitsky, Way and Casey2020). Political parties play a particularly important role in elite collective action since they regulate access to power between ambitious individuals, mediate intra-elite conflict and provide elites with the possibility of future advancement and collective security (Brownlee Reference Brownlee2007, p. 39, see also Geddes Reference Geddes1999, Magaloni and Kricheli Reference Magaloni and Kricheli2010). This is particularly the case at the often problematic moment of leadership succession and generational change in leadership, when the potential for elite divisions and factional disputes is at its peak (Huntington Reference Huntington1968). Clear rules for succession and established mechanisms for compensating those not selected can be vital to retaining elite cohesion (Brownlee Reference Brownlee2007). In contrast, where parties are unable to play this role, offering little hope for advancement or collective security, the inevitable result is factionalism and elite conflict. Where this occurs, some elite factions see political advantage in allying with the opposition (Brownlee Reference Brownlee2007). Social movement theorists highlight such elite fractures as a key factor shaping opportunities for mass mobilisation, with elite divisions raising the possibility that some part of the elite might align with protestors and reducing the likelihood that protests could be forcibly suppressed (Tarrow Reference Tarrow, McAdam, McCarthy and Zald1996, Reference Tarrow1998). In the absence of such divisions, where parties are able to maintain elite collective action, opposition will have few opportunities to challenge the regime through either elections or mass protests (Brownlee Reference Brownlee2007).

In Ethiopia, the challenge of rapid population growth leading to rural land shortages and youth landlessness was already apparent by the early 1990s. Agricultural intensification through productivity growth and extensification through the expansion of cultivated area served to ease these pressures for a period, as discussed in Chapter 5, but could not stall this demographic process indefinitely. For the EPRDF government, industrialisation and mass employment creation was viewed as the only solution to this growing distributive crisis. As such, limited progress in industrial development constituted a major problem, manifest in mass landlessness and underemployment for rural youth, and mass un- and underemployment in urban areas. While Ethiopia is, from a demographic perspective, in its demographic ‘window of opportunity’ (Hailemariam Reference Hailemariam, Cheru, Cramer and Oqubay2019, p. 380), politically, this translated into a distributive crisis rather than a demographic gift. Mass protests began in 2014 and 2015, sparked by state expropriation of peasant landholdings to make way for urban expansion, investments and infrastructure projects. However, these protests soon spread, first across Oromo and subsequently across much of the rest of the country, against an authoritarian regime that had failed to deliver. To understand the impact of these protests, however, it is necessary to turn to elite politics. The EPRDF’s ascent to power and maintenance of political order while in office was rooted in a high degree of elite cohesion based on a collective response to political threats. The 2012 death of Prime Minister Meles Zenawi, who had centralised his hold on political power since 2001, left a political vacuum and emerging divisions within the ruling elite. The result was that the mass distributive crisis that came to the fore from 2014 provoked a crisis amongst the ruling elite, with contending factions taking a very different view of the degree to which the protests represented a threat to their political survival or an opportunity for political advancement. This ambiguous response enabled the spread of the protests, and the combination of mass protests and elite conflict ultimately led to the EPRDF’s collapse.

Access to Land as a Distributive Resource

Control over the distribution of land has long been a key means of establishing and maintaining political authority in agrarian societies (Moore, Jr. Reference Moore1967, Huntington Reference Huntington1968, Skocpol Reference Skocpol1979, Mann Reference Mann1986). In mediated states, land grants were invariably the means of securing the compliance of intermediaries who maintained local order in exchange for control over the agrarian surplus. Equally, land reform has been a key means of consolidating state power and forming a centralised bureaucratic state (Migdal Reference Migdal1988, Albertus Reference Albertus2015, Albertus et al. Reference Albertus, Fenner and Slater2018). Land reform was pursued by both revolutionary movements, such as in China and Vietnam, and counter-revolutionary regimes in South Korea, Taiwan and parts of Latin America to eliminate the threat of powerful elites and to consolidate control over the peasantry.

In Africa, land tenure has also been a key means by which the colonial and post-colonial state sought to project authority into rural areas. Land constitutes ‘a politicized resource or political asset, rather than a politically neutral market commodity’ (Boone Reference Boone2014, p. 310). Radical land redistribution has been comparatively rare, limited to the likes of Ethiopia and Tanzania. Nonetheless, access to land and other agricultural inputs has been an important distributive resource across the continent since colonial times, forming what Seekings calls an agrarian welfare regime (Seekings Reference Seekings, Shapiro, Swenson and Donno2008, Reference Seekings2012). State intervention often takes the form of the adaptation and reinvention of ‘neo-customary’ tenure in which the relation between the state and its citizens remains mediated through neo-customary authorities (Platteau Reference Platteau and Dercon2005, Seekings Reference Seekings2012, Boone Reference Boone2014). Moreover, by providing agricultural extension, and controlling product markets through parastatals, colonial and post-colonial governments sought to develop agriculture while also limiting state responsibility for social spending by placing the burden for supporting relatives on family and communities. Through such an agrarian regime, states seek to integrate farmers into market-oriented production, while inhibiting the commodification of land and labour (Seekings Reference Seekings, Shapiro, Swenson and Donno2008, p. 26).

While political influence over land allocation is a potentially powerful tool, land is only one factor of production alongside labour and capital. Assessing the impacts of land distribution must take into account the broader relations of production within the agrarian economy. Moreover, state land distribution is frequently accompanied by provision of agricultural inputs and extension services required to raise productivity on the land, as well as facilitation of market access for produce. That is to say, the land question is inevitably bound up with the agrarian question. Equally, it should be noted that land access and the agricultural division of labour is highly gendered (Razavi Reference Razavi2003, Whitehead and Tsikata Reference Whitehead and Tsikata2003), with the result that the state control is also likely to be highly differentiated by gender.

A distributive strategy based on access to land is also vulnerable to population growth, whether the result of fertility exceeding mortality or through migration into an area. A locality may be able to absorb a growing population initially without a significant reduction in living standards either as a result of high initial land availability or through efforts to raise agricultural productivity. Ultimately, however, agriculture faces diminishing returns, with the result that relevant state or customary authorities will be forced to restrict access to land – by wealth, age, gender or ethnicity – thereby limiting its distributive reach, or face declining landholdings and living standards (Platteau Reference Platteau and Dercon2005, Boone Reference Boone2014). This has important political implications since, as Boone notes,

Access to land for agricultural livelihoods is also coming under increasing pressure from urban expansion. Rapid population growth translates into urban population growth and intense demand for access to urban land for housing, business and infrastructure (Fox and Goodfellow Reference Fox and Goodfellow2021). Meanwhile, the reallocation of land rights from rural to urban users – whether through market processes or state intervention – is often politically charged. As structural transformation and urbanisation proceed, therefore, the value of land as a distributive tool necessarily declines: urban land becomes more valuable than rural land, while access to jobs supplant access to land as the main survival strategy for the majority of the population. The result is that a ruling coalition seeking to secure the compliance of the masses must find alternate distributive mechanisms to maintain popular acquiescence.

Access to Jobs as a Distributive Resource

As argued above, industrial policy is the key tool available to regimes seeking to promote economic diversification that creates jobs and industrial upgrading that enables expansion into higher productivity, higher wage sectors that contribute to rising living standards. While the state’s attempts to mediate access to jobs in an industrialising economy are likely to entail weakening state control over the popular sectors compared to state control of land in an agrarian economy, the ability to shape the creation and allocation of job opportunities is nonetheless a powerful political tool.

The potential and limitations of industrial policy in the context of globalisation is therefore an important distributive and political question. Though states are undoubtedly constrained by global trade rules, economic competition and the power of lead firms, states do retain the ability to influence the development, operation and distribution of benefits within global value chains and thereby the dynamics shaping employment creation and working conditions (Barrientos et al. Reference Barrientos, Gereffi and Rossi2011, Anner Reference Anner2020). State intervention can take a wide range of forms. States may take a direct role in production through state-owned enterprises or help to establish domestic private firms by mobilising large-scale capital, as with the Korean conglomerates (chaebol), or by providing credit and training to Small and Medium Enterprises, as in Taiwan (Cheng Reference Cheng, Gereffi and Wyman1990, Campos and Root Reference Campos and Root2001). Increasingly, however, the trend has been for the state to promote FDI and to find ways of linking foreign investors to domestic suppliers. The literature highlights three main functions of state governance in value chains (Gereffi Reference Gereffi2005, Alford and Phillips Reference Alford and Phillips2018). Facilitation involves assisting firms to navigate value chains, including through the provision of infrastructure, export processing zones and negotiating trade agreements. Regulation is used to place limits and restrictions on firms with a view to preventing negative externalities. Finally the state plays a role in distribution, shaping the distribution of rents within the value chains through a range of policies such as social policies, taxation and labour regulation (Alford and Phillips Reference Alford and Phillips2018, pp. 102–103).

In spite of these tools, the constraints imposed by value chains in both basic manufacturing and agro-industry are considerable. Global population growth has led to a massive increase in the global workforce relative to the needs of capital, resulting in widespread unemployment, underemployment and the erosion of workers’ bargaining power (Benanav Reference Benanav2014). As such, commercial pressures and the power of lead firms tend to result in downward pressure on producers to raise quality while reducing costs, very often resulting in insecure and flexible labour requirements, and poor working conditions (Barrientos Reference Barrientos2008). The resulting jobs may not be such a prized distributive resource to secure mass compliance after all. Moreover, like land access, job prospects in value chains are highly differentiated by gender and age. In particular, a recurrent tendency is for firms in horticulture and light manufacturing to seek a labour force of young women, who are favoured for their ‘nimble fingers’ and the perception that they are more passive in labour relations (Barrientos et al. Reference Barrientos, Dolan and Tallontire2003). In contrast, the less numerous managerial and supervisory posts are more likely to be dominated by men.

Access to Social Protection as a Distributive Resource

The expansion of state social protection has long been associated with capitalist development and the need to respond to market risks by de-commodifying the means of subsistence (Esping-Andersen Reference Esping-Andersen1990, Polanyi Reference Polanyi2001). As such, unemployment insurance, child benefits and pension schemes, amongst others, have been used to provide protection against risk and vulnerability across the lifecycle. Equally, however, states have long been motivated to expand social protection to maintain political order and limit unrest, from the Poor Laws of nineteenth century England to social transfers in contemporary South Africa, Brazil and China (Seekings and Nattrass Reference Seekings and Nattrass2005, Ngok Reference Ngok, Midgley and Piachaud2013, Ravallion Reference Ravallion2015, Saad-Filho Reference Saad-Filho2015).

The dominant form of social protection in Europe, Asia and Latin America has been contributory social insurance linked to labour market participation. States extended segmented social insurance provisions to secure the acquiescence of workers across Latin America, the generosity of which was heavily stratified based on their relative bargaining power (Mesa-Lago Reference Mesa-Lago1978, Weyland Reference Weyland1996). In several East Asian countries, meanwhile, ‘productivist’ social policy was used as an investment in economic development, as well as a distributive resource (White and Goodman Reference White, Goodman, Goodman, Kwon and White1998, Gough Reference Gough, Gough and Wood2004, Kwon Reference Kwon2004). However, the expansion of contributory social insurance is linked to the expansion of the formal labour market. In many developing countries, high levels of informality translate into low social insurance coverage.

The emergence of social transfers as a global policy priority is important here (Von Gliszczynski Reference Von Gliszczynski2015, Hickey and Seekings Reference Hickey, Seekings, Hickey, Lavers, Niño-Zarazúa and Seekings2019). Policy innovations in countries such as Brazil, India, Mexico and South Africa in the 1990s led to the creation of major social transfer programmes delinked from prior financial contributions. Such schemes range from poverty-targeted cash transfers to lifecycle benefits such as pensions and child grants to cash-for-work schemes. The distributive role played by these social transfer programmes has often been to fill the gap left by a lack of access to land or jobs for sections of the population. In South Africa and much of Latin America, lack of access to land as a result of unequal landholdings, and highly segmented labour markets have meant that broadly distributed social transfers to mothers of young children and the elderly have been used to maintain social peace and avoid more difficult redistributive choices (Seekings and Nattrass Reference Seekings and Nattrass2005, Seekings Reference Seekings2012, Saad-Filho Reference Saad-Filho2015). Given the common recourse to agrarian welfare regimes in many African countries, Seekings (Reference Seekings2012) argues that it is only when access to land becomes unsustainable through population pressure that African governments turn to other forms of distribution such as social protection. However, perhaps the key factor shaping such distributive pressures is the inability of urban labour markets to absorb rural surplus labour. State intervention is particularly likely where these pressures culminate in distributive crises that highlight to incumbents the failings of existing forms of distribution and the potential for such a crisis to threaten the regime (Lavers and Hickey Reference Lavers and Hickey2021).

*********************

The remainder of the book applies this theoretical approach to the analysis and understanding of the EPRDF’s attempts to maintain power and its ultimate disintegration. This begins in Chapter 3 by examining the historical process of the formation of the Ethiopian state and the revolution of the 1970s.