Power, conflict and natural resources: The Mozambican crisis revisited

Power, conflict and natural resources: The Mozambican crisis revisited Abstract This article explores how the present socio-economic crisis in Mozambique is linked to the prospects of natural resource windfalls for the country. Drawing on the political settlement approach, it explores how the distribution of power both within and outside the ruling elite is structured and consequently how the underlying political processes have been shaped by the expectations of natural resource windfalls. The article argues that the present socio-economic crisis in Mozambique is not due to national resource assets in themselves. Instead, the political and economic downturn in Mozambique should be understood as a manifestation of how the political settlement has been organized and rent mobilization controlled by the ruling elite. To understand how the prospect of rents from natural resource sectors have influenced the political settlement, we have argued that one has to look at the dynamics of power both within and outside the ruling elite and the incentives they create for elites to use the control of power and access to economic benefits to achieve narrow and short-term gains rather than inclusive and longer-term goals. Armed violence between Frelimo Government forces and fighters loyal to the Renamo leader Afonso Dhlakama resumed in 2012, twenty years after the General Peace Agreement.1 The violence was immediately linked to the discovery of natural resources and the imminent signing of lucrative liquefied natural gas (LNG) contracts.2 A ceasefire allowed elections to take place in October 2014. After on and off talks between Renamo and Frelimo, mediated by an international team, over Renamo’s key demands for the appointment of provincial governors in the resource-rich provinces where it claimed to have won a majority in the national elections, the low-intensity civil war escalated, with government forces systematically destroying Renamo bases. It later transpired that the intensification of the war efforts by the Frelimo government was made possible as the result of undeclared loans worth more than 2 billion USD, which were uncovered in 2013 and 2016 and involved three enterprises linked to the security services.3 In February 2017, the government’s external debt was estimated at 137 percent of its gross domestic product (GDP),4 forcing the government to acknowledge that it would not even be able to service the debt up to 2021.5 The International Monetary Fund (IMF) considered the taking out of secret loans to be a breach of trust. As a result, donor agencies suspended all loans with the Frelimo government immediately, undermining the state budget and causing one of the biggest economic crises in the history of Mozambique. When the Frelimo government and Renamo agreed in 2017 to extend a fragile truce and intensify negotiations, Exxon Mobil finally agreed to buy a stake in the Italian ENI gas fields in Mozambique, releasing up to 350 million USD in capital gains taxes, a third of what it would have paid two years earlier. As the peace accord looked set to last, international donors said they would resume financing the government after an international firm carried out an audit of the secret loans. How did the Frelimo government end up in such a mess? Just a few years earlier it was the darling of the international donor community, and the country was seen as a booming investment destination, referred to as an emerging African lion with consistently high growth rates. Existing explanations of the crisis in Mozambique point to the resource curse phenomenon. Warnings were issued that the Dutch disease may manifest itself, since the exploitation of natural resources invariably results in exchange rate appreciation, which in turn would crowd out and hamper the development of an already low productivity agricultural sector, on which roughly 70 percent of the Mozambique population depends.6 It was also argued that Mozambique would be vulnerable to resource curse effects due to its poor institutional environment and weak ability to enforce formal regulations.7 Others have suggested that, due to the autocratic tendencies of the Frelimo ruling party and the widespread corruption within the ruling elite and state bureaucracy, the discovery and exploitation of natural resources would not only aggravate the greed of these circles, but also fatally affect the country’s political stability.8 In addition, others argued that the strong overlap between the geographic areas that are rich in resources and those where Renamo is strongest in the country would make the latter vulnerable to violent conflicts.9 While these perspectives in the field of ‘expectations of resource booms’10 are based on predictions that the country would suffer from its resource abundance in the future, others are less sanguine and assume that the curse that has crippled other resource-rich countries is now crippling Mozambique.11 It has been claimed that the resumption of armed conflict in Mozambique is a direct consequence of the discovery of natural resources, a link that is viewed as resulting from the failures of the country’s political system, one of ‘winner takes all’.12 This is said to have fueled the conflict because rents are captured by the Frelimo ruling elite,13 thus excluding the Renamo leadership, which is left with very limited if any access to state resources.14 Mozambique’s exploitation of its natural resources has started and will increase over the next decade, but revenues and large windfalls have hitherto not come in such quantities that they have triggered one or a combination of the various defining features ‘often lumped together as the resource curse’.15 The important work by Jedrzej George and colleagues nonetheless suggests that ‘shared aspirations and expectations alone may make for material political and economic outcomes’, even when countries have not experienced the expected resource booms.16 Mozambique largely fits this predicament, with more than 23 billion tonnes in coal reserves and more than 100 trillion cubic feet of gas, placing it among the world’s ten largest coal producers in the near future17 and the holder of the third largest gas reserves in Africa.18 The exploration of these reserves alone triggered increased foreign investments that have moved natural resource exploitation to the forefront of the political agenda, resulting in high expectations of continued and sustainable economic growth that would reduce poverty and aid dependence.19 Inflated expectations were sparked by the speeches of high-ranking political leaders, such as former President Armando Guebuza, stating that the gas and coal bonanza would create local jobs and the country would acquire a new source of revenues, allowing the Frelimo government to transform the economy and invest in the social sector.20 It was also suggested that by 2025, if not before, Mozambique would become a middle-income country.21 However, despite the increasing investments in natural resources, the country still experiences what has been termed ‘growth without change’,22 with resettlement problems,23 low levels of contributions to GDP,24 and poor linkages between megaprojects and other sectors of the economy.25 In this article, we explore how natural resources are enmeshed within the crisis in Mozambique through a detailed analysis of political and economic conditions prior to the boom in natural resource investments, which largely have been overlooked.26 As it is the case for other countries experiencing resource windfalls, explanations of the Mozambican crisis tend either to be lumped together as having been triggered by natural resources and explained in personalized terms as a consequence of the predatory behavior of the political elites and their rent-seeking strategies,27 or related to poor institutions.28 We argue instead that the present crisis is not a particular feature of the natural resource endowment per se, but instead of how expectations of a future resource boom with huge revenue windfalls are related to how the political settlement has been organized and rent mobilization controlled. We argue that the shift towards a natural resources-driven economy and the expectations of windfalls are intensifying the already established features of the distribution of power in society, bringing the settlement close to a breakdown. A key foundation of the political settlement in post-independence Mozambique has been the ideology of ‘national unity’ fostered by the ruling party Frelimo, which informs patronage and clientelism. The shift from a more inclusive and decentralized political settlement after the General Peace Agreement, under President Joaquim Chissano, to a more exclusive and centralized one under President Armando Guebuza, triggered and eventually exacerbated internal and external elite conflicts over power and resources. Amidst an increasing competition between different factions within the ruling coalition, the country was plummeted into one of its biggest crises in history. The analysis is based on extensive knowledge of and experience in researching the political economy of Mozambique over the last fifteen years. We draw on primary sources generated during fieldwork for the Elites Production and Poverty programme (2008–2012) and the Hierarchies of Rights programme (2015–2018), many of which cannot be directly quoted due to the prevailing political climate in Mozambique. We also use a broad reading of the secondary literature to conduct a macro-political settlement analysis. The combined result, we argue, is a new interpretation of the political economy of Mozambique that differs from many of the standard narratives on Mozambique in the African studies literature. The next section presents the key features of the political settlement framework, discusses its key assumptions, and its use to analyse the political economy of natural resources in Mozambique over time. The following sections describe the political settlement during the era of President Joaquim Chissano, explaining how Mozambique went from a broad coalition to internal conflict, and during the period of President Armando Guebuza, explaining how the management of control and rents was centralized and monopolized. The final section offers an analysis of current developments during Filipe Nyusi’s presidency from 2015 to the present, during which the crisis matured. Political settlement analysis and ‘foundational ideas’ The economic structures of developing countries like Mozambique do not create strong incentives to use solely the formal institutional state apparatus to manage the economy and its resources. Developing countries are drawn towards the use of patron–client networks to distribute assets and economic opportunities because they are in the process of capitalist transformation. In such countries only a relatively tiny proportion of the productive population depends on the market for its survival and the reproduction of social relations.29 As productivity is generally low, the formal economy based on taxation cannot, in any sustainable manner, support the distribution of power that has evolved in predominantly capitalist societies. Mushtaq Khan’s concept of political settlement attempts to capture, as a shorthand, the set of institutions and power relations that characterize the social order in a particular country.30 Our use of political settlements draws from the work of Lindsay Whitfield, Lars Buur, and colleagues, which elaborates and extends Khan’s theoretical framework.31 We focus on three structural dimensions of a political settlement in developing countries that are important for understanding Mozambique’s economic and political crisis and how it relates to the prospects of natural resource windfalls. The vertical dimension refers to the organizational structure of patron–client networks and the factions within those networks that are involved in forming a ruling coalition. This dimension is related to two aspects of the ruling coalition that are often confused: first, the distribution of power among elite factions within the ruling elite, where relative exclusion can take place; and second, the distribution of power among higher-level and lower-level factions within the ruling coalition in general. While intertwined in practice, both aspects influence how political leaders respond to natural resource opportunities. The horizontal dimension involves the relative strength of excluded political factions, that is, the organizations or factions outside of the ruling coalition. The distribution of power between the ruling coalition and those factions excluded from it is often easier to explore than what goes on within a ruling coalition. How the relationship between the included and excluded factions is organized based on incorporation, co-option, or cohesion usually has clear historical traces. Similarly, the legitimacy and organizational ability of included and excluded factions are historically grounded. Nevertheless, their relative strength cannot be taken for granted, as will become clear from the Mozambican case. Excluded groups can actually be strong, but can also be effectively repressed, at least temporarily, by the ruling coalition. The third structural dimension of a political settlement is how political regimes are financed, including the question of rents and rent-seeking, as no regime can survive and reproduce itself without resources. In African countries, where the technological capabilities and relative power of capitalist entrepreneurs are generally weak and where most of the resources that can be used to reproduce power emerge from or are organized through the state, the importance of active rent creation is central. For obvious reasons, extractive natural resources such as minerals and oil are attractive to ruling elites. Developing new productive industries and sectors takes much longer and requires more work than granting licenses for exploitation, negotiating capital gains taxes and mobilizing finance from transnational companies. Thus, those firms and economic sectors that provide major sources of state revenues, foreign exchange and rents that allow the ruling elite to reproduce power quickly are important. Ideas and political settlements also interact, and Tom Lavers and Sam Hickey argue this is not always clear in political settlements theory, as they see Khan’s theoretical framework as predominantly interest-based.32 In our understanding of how political settlements are organized and legitimized, we include an analysis of the set of shared and contested ideas underpinning a particular settlement. This is important because structural constraints and the balance of power between contending factions ‘rarely, if ever, determine particular policy responses’.33 We focus specifically on the foundational idea of ‘national unity’ embodied in the ruling Frelimo party, as it continues to inform how patronage and clientelism are organized.34 However, we argue that it does not make much sense to retain the separation of Frelimo’s historically based foundational ideas like ‘national unity’ from the perspective of an interest-based political settlement analysis. Having emerged in the 1960s from a coalition of independence movements, Frelimo celebrated its fiftieth anniversary in 2012. It has been in power since independence in 1975 and has changed from being the Marxist-Leninist vanguard of a one-party state, proclaimed in 1977 but repudiated in 1990, to become the dominant party under the liberal multiparty constitution. In the party’s and ruling elites’ own understanding, Frelimo is the shaper, even the dono do país (owner of the country), of Mozambique’s contemporary history.35 It was initially a vanguard party with a few men at its helm guiding the masses and using the state as their instrument.36 When Frelimo broadened its membership base in the 1980s, the ‘admission of members’ was a strict process where only the select few, those who were considered trustworthy or pessoas de confiança, were admitted.37 Such individuals were ‘entirely dedicated to the cause of the party, the country, the people, and socialism, and who, living exclusively by their labor, both accept and direct their ability and energies to the realization of the party’s statutes and programmes’.38 After 2005, this was reasserted under President Armando Guebuza, who realigned state institutions more with the party after a period of relaxation under President Joaquim Chissano. In the analysis that follows, we use these three structural dimensions and show how they have become intertwined with the idea of ‘national unity’ in order to analyse Mozambique’s recent turn to a natural resource economy and the impact this has had on the country’s recent history. It was during the Chissano and Guebuza periods (1994–2004) that the roots of the present crisis were laid and intensified. The Chissano era: from a broad coalition to internal contestation Chissano’s term of office emerged from the tortuous process of economic and political reforms initiated before the death of Mozambique’s first president, Samora Machel, in 1986, and the country’s adherence to the Bretton Woods Institutions in 1984 following the demise of the socialist model of development. Economic reforms were initiated during 1987–1990 with the adoption of a structural adjustment programme, and political reforms culminated in the enactment of a new multiparty constitution in 1990. In 1992 the government signed a General Peace Agreement with Renamo, which ended the sixteen-year civil war initiated in 1976 (a year after the country’s independence) and paved the way for the 1994 general elections. Joaquim Chissano’s two-term multiparty era started with the first multiparty elections in 1994 and ended in 2004, when he relinquished power in favor of Armando Guebuza, due to internal party pressures. During the Chissano era, peace was consolidated after the devastating civil war, and the country experienced double-digit economic growth and relative political stability, apart from Renamo’s complaints that the elections of 1994 and 1999 were rigged. Chissano’s ruling coalition accommodated the interests of many different factions and individuals both within and beyond the ruling party, although not extensively, as it mostly included the Renamo leader Afonso Dhlakama as part of the peace-building process and a few of Renamo’s generals. The Chissano era has been characterized as one of political liberalization, with less emphasis on the state-party link. Religious groups, ethnic-based lobbying and different economic interests were allowed to flourish, combined with a controlled attempt at decentralizing power after 1996. The more hard-line factions within Frelimo contested this political liberalization, as they saw it as a challenge to the foundational value of ‘national unity’ organized around Frelimo’s continued dominance. Within the ruling coalition, the economic reforms involved a relatively controlled process of primitive accumulation that led to the emergence of a domestic business class closely linked to the Frelimo ruling coalition.39 The privatization process was an important element in creating a national business class linked to the party. Carlos Castel-Branco argues that the process of accumulation by the local elites was led by the state in the post-war period and had two distinct stages. In the first stage, the state empowered local elites, de-capitalized and without business skills and experience, through favorable privatizations of public assets and enterprises. In the second stage, it selected from the previously empowered local elites those whose loyalty was more certain when it came to forming a national oligarchy, which over the last decade has been linked through the Frelimo state to international capital and investments in natural resources.40 Despite one of the most ambitious privatization programmes in Africa, public enterprises continue to be important in financing the political settlement in Mozambique, formally by providing top managerial jobs to key members of the Frelimo coalition, and informally by paying for party activities.41 Chissano’s coalition-building approach was technocratic, consisting of an extensive economic, political and administrative reform agenda and the appointment of technicians with little political influence over government positions. This approach broke with that of previous regimes, which had given precedence to politics over technical matters, and drew heavy criticism from internal party factions linked to Frelimo’s ideological socialist hardliners, liberation fighters and older members of the party. The reform programme included strengthening industrial sectors that could provide returns in terms of rents and political support, such as the labor-intensive and rural-based sugar industry.42 It also involved a ten-year public-sector reform programme launched in 2001, funded by the World Bank and donors, which focused on the administrative apparatus of the state, salary reform, and the introduction of a meritocratic and managerial public service. Some of these measures, such as the fight against corruption, conflicted with the nature of the coalition-building approach. Thus, Chissano’s period was also known for the emergence of corruption in the public sector, as epitomized in the phrase, ‘the goat eats where it is tied’, and for its laissez-faire (deixa andar)43 attitude to public service. The Chissano era became known as corrupt in part because of its departure from Frelimo’s strict ideology of Marxist-Leninism and party discipline, but also because of the broad-based process of accumulation that the regime permitted in order to accommodate social groups and factions that could threaten the ruling Frelimo party electorally, and the fear of losing control and power to Renamo. Another way of distributing power and rents was through the process of decentralization. This process was initiated in 1994 but was reversed after the general elections of 1994 and the unexpected electoral successes of Renamo in the name of maintaining ‘national unity’. The Frelimo-controlled parliament approved instead a controversial law that opted for gradualism in the creation of municipalities, 33 being set up in 1997, and 43 in 2007 and 53 in 2013. After boycotting the first municipal elections, Renamo won five municipalities in 2003, thus introducing the first experience of alternation of power in the country. Decentralization was also important for Frelimo local elites, who from the beginning of the movement’s foundation in 1962 claimed to exert power at the local level but were tamed by the strong discourse of ‘national unity’ and centralized party control. Under Chissano, a deconcentration law was approved in 2003 granting more administrative powers to the provincial and district governments. Regarding the horizontal dimensions of the political settlement during the Chissano era, for Renamo accommodation meant the appointment of district administrators proposed by the party, pecuniary benefits for its leadership and the inclusion of some of its officials in the military command, as provided for in the General Peace Agreement signed in 1992. The post-civil war institutional arrangement also included power-sharing between Frelimo and Renamo, as reflected in the power of parliamentary parties to appoint representatives to the National Electoral Commission, the Constitutional Council, the Council of State and the Security and Defence Council. Parliamentary parties were also entitled to have state subsidies funded from the public budget, calculated according to their parliamentary representation. These arrangements provided some level of participation and access to state resources for Renamo cadres under an arrangement sometimes called ‘Frenamo’.44 However, the process of resource distribution to Renamo was clientelistic in nature, bolstering the authoritarian and centralizing leadership style of its leader Afonso Dhlakama and generally benefitting Frelimo. The resources made available to Renamo and Dklakama through the peace agreement and the practice of parliamentary inclusion did nothing to contribute to the institutionalization of Renamo, nor did it benefit a wider range of the movement’s membership. Despite its repression of Renamo, competitive elections proved to be one of the main causes of fear for the ruling party coalition. In the first post-war democratic elections in 1994, Chissano took 53 percent of the votes, while Dhlakama received around 33 percent. Renamo won in the central and northern provinces, but due to the winner-takes-all system, Chissano could maintain control. This was repeated in the 1999 general elections, when Chissano again won the presidency, but with a very slim margin of 4 percent over the Renamo candidate. However, there have been strong suggestions of electoral fraud and even that the elections for both president and parliament were rigged, indicating frail political support for Frelimo among the population, particularly in the central and northern regions where Renamo won comfortably.45 A substantial part of Renamo’s guerrilla forces has not been integrated into the regular army, allegedly because it was agreed in the General Peace Agreement that Renamo could keep a certain number of its forces under its control to protect its leaders. This residual force came to play a role in the political crisis that erupted in 2012, following the approval by Frelimo of a contested electoral law and Renamo’s allegations that its ex-guerrilla fighters were being compulsorily retired from the army and kept from high-ranking positions within the security establishment.46 Despite the relatively inclusive nature of Chissano’s government, Renamo was systematically excluded from access to resources such as jobs in public enterprises, benefits from privatizations and access to opportunities for linkages with megaprojects and other foreign direct investment projects. The main source of accommodation for Renamo was through the narrow opportunities for participation in Parliament and its ability to propose appointees. Renamo was facing a process of weakening, which accelerated after 2004 when Guebuza took control. Moreover, below the thin veneer of tolerance and inclusiveness under Chissano, the harassment of opposition party members did not cease, especially locally, and access to high-ranking public positions and economic opportunities was still substantially dependent on the ruling party’s blessing. At the beginning of the democratic era in 1994, Chissano had assumed that the controlled but limited inclusion of Renamo would reduce the movement’s influence over the years. A decade later, Guebuza would consider this slow-death strategy too slow, given the need to accelerate control over the territory in light of the prospect of opening up natural resource exploitation in former Renamo areas. One dramatic consequence of Guebuza’s strategy would paradoxically trigger a ‘second coming’ for Renamo. Turning to the last dimension of the political settlement, the financing of the regime took place within the context of a reduced national tax base. Development aid accounted for more than 50 percent of the state budget, and the Chissano government’s ‘controlled’ economic and political reforms provided a strong argument that allowed the country to benefit from the Highly Indebted Poor Countries Initiative, leading to a substantial reduction of its external debt. Public procurement and access to natural resource licenses (land, mineral resources) and to public office and employment, all based on the demands of party loyalty, provided the ruling elite with rents with which to buttress the regime. However, this system of rent distribution, although relatively wide in terms of intra-elite inclusion, was centralized and based on the direct control of the state apparatus, which provided considerable power to the president and his technocracy. This contributed to sidelining the lower-level factions within the Frelimo party, mostly in the provinces. Chissano’s reliance on the technocracy came at the expense of reducing the party’s influence over decisions on rents and rent accumulation. Moreover, the high level of dependence on development aid implied the existence of an accountability architecture that forced the government to be more accountable to donors than to national constituencies, society or the party.47 This was heavily criticized inside the party, and, combined with its narrow victory in the 1999 general elections, paved the way for pressures for a new leadership that would privilege party politics over technocracy, and national (party) accountability over accountability to external actors, i.e. preserving Frelimo members’ privileges. It was under Chissano’s regime that foreign direct investment inflows started with large-scale investments, such as the Mozambique aluminum project (Mozal) and the production of natural gas in Pande and Temane by Sasol, which included building a pipeline from Temane in southern Mozambique to Secunda in South Africa. It was also under Chissano that negotiations to resume coal mining were initiated, with Riversdale buying one of the first concessions very cheaply. It was later sold on to Rio Tinto without the state having any legislation in place to tax the capital gains. However, these efforts provided narrow and dispersed opportunities for rents. Mozal, for example, had a project to create linkages with local companies that only benefitted a few members of the political elite.48 Some of this group also seems to have benefitted from the building of the Sasol pipeline, but exactly how is still unclear. During the Chissano era, decision-making power was not only shifted from the party to the state technocracy, which was the gatekeeper for access to rents. The state technocrats also benefitted from their proximity to President Chissano, whence the increasing levels of corruption. It was in this context that Armando Emilio Guebuza was elected secretary-general of the party in 2002 and confirmed as Frelimo’s candidate for the 2004 general elections. Guebuza adopted the fight against corruption of the Chissano era as a platform for his political campaign in the 2004 elections, which he won by a large margin. The Guebuza era: from internal contestation to monopolization In his bid to consolidate the leadership of the party and be elected president, Guebuza revitalized the party structures (party cells) from the grassroots up to the highest levels of the state and parastatal companies. He also promoted the provincial party secretaries as local political leaders, restored the notion of Frelimo as the ruler of the country, and revived the foundational idea of national unity organized in and through the Frelimo party. Party membership as a condition for access to public office was promoted at all levels of society, with an increasing number of party cells in public entities. Mobilization also increased, and in 2012 the general-secretary of Frelimo claimed that the party had more than four million members, although it has never attained such a figure in any election, despite the statutory obligation for members to vote in all elections. This strengthening of the party was important for the balance of power both within and outside the ruling coalition. Within the ruling coalition, Guebuza positioned himself in opposition to the Chissano era’s inefficient public service, with its widespread corruption, and he promised to boost national capitalist development. This did not necessarily mean that all factions within the ruling party coalition would benefit, since the available economic opportunities were distributed among the narrow coalition of those closest to Guebuza, who controlled the economy and rents with a tight fist, in contrast to the Chissano era.49 Despite the revitalization of the party down to the local level, the distribution of power was highly unequal between the top and bottom of the party structure. Within the party, control was effective but not simple, since there were conflicts between higher- and lower-level factions. Frelimo has historically used influential political figures, among them liberation struggle veterans, to exert control over the local-level membership, and even state structures. During elections for both party and public positions, central brigade leaders were responsible to the Frelimo top leadership for ensuring that the candidates preferred by the party leadership at the central level were selected (and elected) in local elections. This created discontent locally, with people staying away from elections. The imbalance of power was also reflected in the way power was distributed within the state and the process of decentralization. Municipalities run by the opposition complained that the central government was blocking their bids to take over the functions provided for in the legislation. The budgetary allocation across government layers also showed the asymmetry of power, with up to 50.3 percent of recurrent expenditure located in the central government between 2010 and 2014, and over 80 percent of capital expenditure, including expenditure for projects implemented at the local level. This has implications for the resources available for the decision-making of local party factions as well as opposition-run municipals, and consequently for their relative power to influence politics. Concerning the horizontal dimension, which is where the relative strength of the excluded factions is located, Frelimo’s and Guebuza’s main political adversaries, Dhlakama and Renamo, were defeated in the 2004 general elections by a margin of more than 30 percent. This margin allowed Guebuza to dismiss their importance as political interlocutors, to cease the limited flow of resources initiated during the Chissano era, and to initiate the sidelining of Renamo’s former guerrillas in the armed forces. Frelimo’s victory in the 2009 elections, which gave it more than two-thirds of the parliamentary seats and more than 70 percent of the vote for Guebuza, consolidated his success as a powerful president and party leader. However, this electoral success was misleading, since it was based on a low voter turnout and increased control over those voting. The turnout for national elections went from 87.9 percent in 1994 to 69.5 percent in 1999 to a low of 36.42 percent in 2004, before rising a little to 44.4 percent in 2009.50 Access to state resources and the capacity to win all parliamentary and presidential elections since coming to power and thus to provide positions and opportunities for organized members consolidated Guebuza’s control not only over the party but also society. However, this did not go unchallenged. Outside the party, Renamo resorted to war after its claims to keep its members in the armed forces and under its command were no longer accepted by the Guebuza government, and after disagreement over a new electoral law approved unilaterally by the Frelimo caucus in Parliament in 2012. In 2014, an agreement between Renamo’s leader and Guebuza allowed Renamo to participate in the 2014 elections, in which its presidential candidate, Afonso Dhlakama, and the party contested the results and had considerable success. Renamo claimed that they had won in six provinces, though more likely in four, in which they demanded to appoint their governors or else they would rule by force. Frelimo and the government refused to respond positively to Renamo’s demands, arguing that it would undermine ‘national unity’. After failed negotiations and alleged attempts by the national army to assassinate Dhlakama, the country was plunged into a military crisis affecting mostly the central region provinces. Regarding the third structural dimension of the political settlement, the financing of the regime, under Guebuza this still relied on donor funding, but there was a greater incentive to achieve autonomy. However, this autonomy was not directed towards following a clear development agenda, as the international development partners wanted. As noted, the logic of the Frelimo state was now one of a bureaucracy more concerned with control than effectiveness.51 Even party members were expected to show strict loyalty to the party, which insisted that those holding state positions were above all accountable to the party. This was not only a political matter, since it also had implications for the economy. The focus on short-term economic gains prevented the government from investing in a consistent strategy to create or strengthen the productive sectors. The country lacked a coherent development strategy apart from poverty reduction action plans, whose scope was narrower. Only in July 2014 did the government adopt the National Development Strategy (NDS) 2015–2035, whose main objective is to improve the living standards of the population through the structural transformation of the economy and the expansion and diversification of the production base: industrialization and agriculture being at the core of the programme.52 The NDS acknowledges that the national economy generally and the rural economy in particular depend heavily on the exploitation and use of natural resources. While the NDS signals that natural resources will be key sources of rents and economic accumulation in the future, their contribution to the economy thus far has been low. According to official data, the contribution of the extractive industry to GDP is increasing at very impressive rates, from 2.2 to 5.1 in the 2011 to 2015 period, but its share is still very low.53 The growth in the extractive sector results from a combination of global and national factors. Changes in the global economy during the 2000s made it highly profitable for foreign companies to develop Mozambique’s natural resource endowment, so there was a clear pull factor. But for the Guebuza ruling elite there was also the push factor of breaking free from aid dependence when criticisms of elections became linked to aid conditions, as when Guebuza was re-elected in 2009. After the general election, in 2010 this culminated in a group of donors forcing through a ‘donor boycott’ involving the suspension of general budgetary support until a set of targets for reforming the biased electoral and parliamentary system had been agreed.54 The combination of revenue sources initially gave the Frelimo party and the Guebuza administration ample room to breathe: natural resource licenses, capital gains taxes (in force since 2013), continued high levels of aid disbursement despite donor criticism, new loans from both the World Bank and various emerging powers like India, China and Brazil, and sovereign bond flotations. The negotiations over major natural resource investments were generally clouded in secrecy, with limited or zero state knowledge or parliamentary participation and oversight.55 Another way to reduce donor dependence and influence was through the selective promotion of pockets of efficiency in key sectors such as state revenues, where special incentives were created to strengthen tax administration. Over the last ten years, this has been responsible for a remarkable increase in revenue collection to reduce dependence on foreign aid, with implications for the government’s increasing ability to resist donor leverage. Thus, between 2005 and 2015 internal revenues increased from 57.8 percent to 75 percent (and state revenues from 56.7 percent to 71 percent), while foreign aid grants decreased from 28.4 percent to 9 percent of total budget revenues.56 The National Institute for Petroleum is another example of creating potential pockets of efficiency for rent management related to the new extractive sectors. But excellence was not necessarily translated into following the rules during the Guebuza era. The sale of a stake in the ENI gas concession in 2013 is a good example of how capable pockets of efficiency can exist formally but only be used selectively. When ENI announced the sale of part of its stake in the Rovuma Basin to the China National Petroleum Corporation, it was, as the regulation suggests, subject to the approval of the Mozambican government. Protracted discussions with the government and ultimately President Guebuza were centered directly on determining the amount of capital gains tax to be paid on the $4.21 billion sale. The Mozambican parliament had by then approved a capital gains tax of 32 percent, but the president had not formally signed the law into effect.57 By negotiating directly with the president, ENI managed to reduce the tax to $400 million, or a rate of 9.5 percent instead of 32 percent. This followed similar decisions on capital gains tax that saw a far lower rate of revenues entering the state’s coffers than the capital gains tax stipulated,58 with continued speculation that the Guebuza regime was using the executive’s discretionary power to accumulate and fund the party and personal ambitions. Therefore, the beginning of the exploitation of natural resource projects, mostly coal and the huge gas discoveries in the Rovuma basin, provided the necessary elements for the expansion of formal and informal rents. Although the contribution of gas and mining to state revenues remained low, two elements in particular held out the prospects of high rents. The first was the optimistic projection that the country would be a major LNG exporter, with a considerable level of fiscal revenues entering the state’s coffers over the next decade, amounting to US$ 500 billion by 2045,59 although these figures are contested.60 The second is that capital gains taxes started to flow after some companies began selling their assets in Mozambique. The process started during Chissano's mandate with the purchase of Riversdale’s coal mines by Rio Tinto in 2011, which until then had not been taxed. The government discovered that these transactions could be taxed, introduced a legislative revision in 2013, and collected revenues in subsequent transactions of natural resource assets totaling US$ 1.3 billion. However, although Mozambique has been compliant with the Extractive Industries Transparency Initiative since 2012, it is not clear how these revenues were used. Growing internal revenues and the availability of resource rents during Guebuza’s second term were also combined with an increase in non-concessional loans (internal and external) to offset the reduction in development aid. Reduced aid inflows stemmed from the global financial crisis and changing donors’ policies, among them the weakening enthusiasm for direct budget support, emerging ideas of aid for trade and a focus on business development. The natural resources boom stimulated the inflow of foreign direct investments in the country and the interests of the emerging economies in the areas of mining, gas and agriculture, with sizeable investments from China, Brazil and India. However, foreign direct investments were concentrated in some areas and were capital- rather than labour-intensive. As has been observed, thirteen megaprojects accounted for 58 percent of the private investment approved between 1990 and 2012 (of US$ 20.2 billion), which employed 20,000 people.61 In summary, under Guebuza’s mandate rent distribution continued, mediated by the control of the state apparatus as the gatekeeper for public procurement, public-private partnerships and contracts with public enterprises.62 An additional element brought about by the national resources boom was access to foreign direct investment geared towards the extractive industries, with low linkages to the local economy, no relevance for the development of productive sectors and low employment rates.63 Moreover, the few linkages established depended on previous experience and institutional capacity, which was only available to those companies that were part of the state’s supply chain or other megaprojects. As a result, a vicious cycle emerged regarding those who benefited from the extractive economy, namely those with political and institutional links that favoured access to such projects.64 It was during this period of feverish expectations of LNG revenues that Guebuza had to relinquish power, having fulfilled two terms in office. The emergence of Nuysi: from bad to worse? Guebuza did not want to give up power just when the vast natural resource endowment finally seemed secure. His power was consolidated at the tenth party congress in 2012, despite increasing dissatisfaction within the ruling party, mostly because of his centralized control over rent opportunities and the prospective opportunities emerging from the state and the new extractive economy. Channeling rents or access to rent opportunities to his closest allies, Guebuza succeeded in securing the election of members loyal to him in the party’s decision-making structures: the Central Committee and the powerful Political Commission. But the combination of internal and external dissatisfaction with Guebuza’s monopolization of the prospective resources led to increasing opposition within the party that heavily influenced the process of choosing Frelimo’s new presidential candidate and ultimately the president of the party. Guebuza had first tried to appoint a loyal successor, but failed to do so because his opponents, mostly the liberation struggle veterans and more liberal members, rallied to support the candidacies of Filipe Nyusi and the former prime minister and Chissano protégé Luisa Diogo. The latter had been aligned to Guebuza as defence minister, and it was during his time as minister that the conflict with Renamo had escalated and a series of secret loans for defence purposes were taken out.65 Nyusi was supported by longstanding military figures of the gas-rich province of Cabo Delgado, chief among them the former Minister of Defence, Joaquim Chipande. Nyusi defeated Luisa Diogo in the internal elections, received the party’s nomination and was elected President of the Republic in the 2014 general elections. In March 2015, Guebuza’s opponents succeeded in electing Nyusi, as president of the party, in line with the Frelimo tradition of combining the presidency of both country and party, using the argument that this avoided fragmenting the political power underpinning the quest for ‘national unity’ in and through the Frelimo party. But Nyusi does not control the Central Committee or the Political Commission, making it difficult for him to assert his formal power. This might change in the eleventh congress of Frelimo being held in September 2017. Increased foreign direct investments and the prospects for natural resource revenues also stimulated the increase in public indebtedness, including so-called secret and illegal debts, amounting to more than $2 billion, which funded dubious investments in the tuna fishing company EMATUM, PROINDICUS for maritime security and Mozambique Assets Management for oil and gas logistics between 2012 and 2014.66 Most of these loans, it was eventually revealed, were channeled into purchasing military and surveillance equipment, and it is suspected that some of them were used for the personal benefit of members of the ruling elite. Between 2009 and 2014 the government also arranged another undisclosed loan of $221 million for the Ministry of Interior, whose objectives are still unknown,67 besides accruing a massive internal state debt because bills for recurrent costs, construction, medicine, ammunition and weaponry over the last years were not honored. This public indebtedness was only possible due to the guarantees of the availability of natural resource rents in what was expected to be the near future, possibly post-2025. This level of public indebtedness results in part from the relative success of the country in reducing its dependence on donors and their leverage over its policy decisions. It also reflected the schisms within the ruling elite and its centralization of power, in the sense that only a reduced number of individuals close to the president and inside the Frelimo Party, the government and parliament knew what was going on. The government even bypassed parliament in the process, which is supposed to grant authorization for loans beyond the levels approved in the state budget. However, donors and international financial institutions such as the World Bank and the IMF are still important sources of state revenues, and they influence investors’ decisions. The discovery of these concealed debts angered the Bank and Fund, who decided to suspend their support to the state budget until an investigation into and audit of these loans were carried out. This plunged the country into a serious budgetary and foreign exchange crisis that affected the prospects of economic growth, with implications for state revenues. Without resources from its international partners, amounting to around $500 million annually, the government was forced to revise the approved budget for 2016 and introduce austerity measures. Despite government claims that social policies would not be affected, the revision of the budget resulted in cuts in social expenditures and increases in defence and security spending.68 The government was initially reluctant to allow a forensic audit as demanded by the IMF and other development partners, which would reveal the real beneficiaries of the concealed loans and end with them being taken to court if necessary to satisfy the demands of the country’s development partners. Eventually the government agreed to an independent audit being carried out by the international company Kroll Inc., which, however, faced difficulties in assessing information, delaying the delivery of results from February to May 2017.69 The secret loans were taken out during the last years of Guebuza’s term, with Nyusi as an insider. However, it involved a narrower group of the ruling elite, creating an oligarchy, with Guebuza at its forefront. Power struggles within the ruling coalition, along with uncertainties regarding the securing of property rights in the future, began to challenge the concentration of power within Frelimo. This is evident under the Nyusi government, with persistent appeals of Frelimo senior members to Guebuza to allow more space for the new president to deal with the impending problems the country faces. It was only when the regime had its back against the wall and was facing the full effects of a financial meltdown that there was a breakthrough in the civil war. International mediators, demanded by Renamo, initially failed to negotiate a proposal for a high-level meeting between Nyusi and Dhlakama with an agenda that included decentralization and the integration of Renamo members into the army, as well as their disarming. As these efforts failed and it became clear that Mozambique would default on all its secret loan payments from the end of 2016, the regime stood on the brink of defaulting altogether, as the state coffers were empty, and the IMF demanded the withholding of all international aid. In late December 2016, Nyusi and Dhlakama agreed on a truce for the festive season, later extended indefinitely, with promises by both parties to reach a peace agreement and end the conflict. Frelimo was on the brink of breaking up, but as the ruling party has done every time it has faced possible defeat, its leadership and the different factions stuck together, since they know that their survival depends on the unity of the party and its control over the state and society. This became clear in early 2017, when Frelimo probably could have challenged the two banks that initially granted the key secret loans with relative ease in the London courts. The two banks, Credit Suisse and the Russian bank VTB, had clearly failed to do due diligence, thus making the loans ‘illegitimate debt’ or a type of ‘odious debt’, as lenders have a fiduciary responsibility to borrowers regarding ‘loan pushing’.70 However, following this route would have required the Frelimo government to make public all information about who took out the loans and benefitted from them. This would have implicated the leaderships around Nyusi and Guebuza and possibly have split the party. In January 2017, the Mozambican administrative tribunal declared that the secret debts were both unconstitutional and contrary to budgetary laws, making the secret loans illegal. However, since April 2017 the loans have been included retrospectively in the 2015 State General Accounts report, since they were ‘proven to be in the public interest’.71 Through this move the state accepted the guarantees and will pay to relinquish individual responsibility, making legal what was formerly illegal. The paradox is that the secret and illegal debts, which were based on the prospect of natural resource windfalls, became legal overnight through their inclusion in the state budget, being passed on to the people of Mozambique in the name of maintaining party unity. The state guarantee is therefore more than just a guarantee of debt, it is also what allowed the civil war to end, as the war-mongers among the ruling Frelimo elite were protected from financial liabilities. The move also allowed the continued drive towards maintaining ‘national unity’ organized in and around the continued dominance of the Frelimo party, not only as the sole legitimate government of the country, but also as the continued guarantor of peace, at least in the short term. Conclusion So far expectations of future rents is the only significant factor stemming from the natural resources boom, but natural resource rents per se do not explain the current situation in which the ruling elites tend to favor more short-term gains, instead of more developmental policies, like the diversification of the economy and the building of strong productive sectors. To understand how the prospect of rents from natural resource sectors have influenced the political settlement, we have argued that one has to look at the dynamics of power both within and outside the ruling elite and the incentives they create for elites to use the control of power and access to economic benefits to achieve narrow and short-term gains rather than inclusive and longer-term goals. In this article, we have sought to contribute to the discussion about Mozambique’s crisis by showing how the expectations related to the natural resource boom that contributed to the crisis are enmeshed in complex political processes. Theoretically, the article draws on the political settlement approach, which offers a comprehensive framework for understanding how the distribution of power is organized and potentially changes over time. We argue that the prospect of natural resource rents exacerbated already existing power struggles and distribution of resources within and outside the ruling elite, informed by the continuation of the ideology of ‘national unity’, which in the last twenty years changed and evolved from a moderately inclusive political settlement in the Chissano’s era to an exclusive one in the Guebuza era. This exacerbated internal and external conflicts and competition over power culminating under Nyusi's presidency with the Frelimo party struggling to stay united for political survival. Internal ruling elite struggles over access to potential short term resource rents have limited Nyusi's capacity to reach an agreement with Renamo and to tackle the root causes of the political crisis with Renamo. The argument in this article, while clearly drawing on previous scholarship on political settlements as well as discussions of the resource curse phenomenon, makes a number of original contributions. First, where Jedrzej Frynas, Geoffrey Wood and Timothy Hinks add expectations of resources to the list of the effects of the resource curse, they do not relate this to how expectations of future natural resources become intertwined with the organization of and possible changes to the evolution of political settlements or of the political order.72 This article directly links the question of expectations related to resource windfalls to the evolution of the distribution of power over time. We argue that, as expectations increased, it put severe stress on the ability to maintain stability in the organization of power, as different factions juggle for access to prospective resources. Second, whereas much of the existing literature in African studies on political settlements is rooted in sectoral analysis, at its core the political settlements approach is a macro-level form of study. This article provides a macro-analysis of the distribution of power and what constitutes such power in societies, linking this analysis to expectations of natural resource rents. Finally, this article specifically engages with a broader discussion of how ideas and ideology intersect with a political settlements analysis. We argue that, in the case of Mozambique, historically based foundational ideas like ‘national unity’, linked to the continued dominance of the Frelimo party as providing the sole legitimate government of the country, cannot be separated from a more interest-based political settlement analysis, as the two are intertwined in practice. Footnotes 1. The 1992 General Peace Agreement ended a sixteen-year insurgency by Renamo against the Frelimo government. 2. Adrian Frey, ‘War in peace: The return of civil war in Mozambique’, Club of Mozambique, 3 May 2016, <http://clubofmozambique.com/news/opinion-war-in-peace-the-return-of-civil-war-in-mozambique/> (11 October 2016). 3. For the main facts of the debt crisis, see Sapo Notícias, ‘Cronologia: Da Ematum à incorporação das dívidas nas contas de Moçambique’, 18 April 2017, <http://noticias.sapo.mz/info/artigo/1502039.html> (30 April 2017). 4. Adrian Frey, ‘Standard & poor’s: Mozambique’s debt to fall from 137% to 118% by 2020’, Club of Mozambique, 12 February 2017 <http://clubofmozambique.com/news/standard-poors-mozambiques-debt-fall-137-118-2020/> (18 April 2017). 5. Frances Coppola, ‘Mozambique is about to default on its ‘tuna bond’, Forbes Investing/Foreign Affairs, 17 January 2017, <https://www.forbes.com/sites/francescoppola/2017/01/17/mozambique-is-about-to-default-on-its-tuna-bond-again/#585fc8e1431a> (1 May 2015). 6. David Rosenfield, ‘The coal mining sector in Mozambique: A simple mode of predicting government revenue’ (Conference Paper No. 9, Institute of Social and Economic Studies, September 2012). 7. Cayley Green and Lisa Otto, ‘Resource abundance in Mozambique: Avoiding conflict, ensuring prosperity’ (Occasional Paper, South African Institute of International Affairs, February 2014). 8. Boniface Mabanza, ‘Mozambique at a crossroads: The natural resources trap’ (Kirchliche Arbeitsstelle Südliches Afrika, 2013) <http://www.woek.de/web/cms/upload/pdf/kasa/publikationen/Mabanza_2013_Mozambique_at_the_crossroad.pdf> (8 October 2016). 9. Green and Otto, ‘Resource abundance in Mozambique’. 10. Jedrzej George Frynas, Geoffrey Wood and Timothy Hinks, ‘The resource curse without natural resources: Expectations of resource booms and their impact’, African Affairs 116, 463 (2017), pp. 233–260. 11. Peter Fabricious, ‘The resource curse comes to Mozambique: Mozambique is teetering on the brink of a major sovereign debt default, which has economic, developmental and security implications’, Institute for Security Studies, 9 June 2016, <https://www.issafrica.org/iss-today/the-resource-curse-comes-to-mozambique> (16 August 2016). 12. James Gordon, ‘Curse or cure? The political implications of Mozambique’s natural resources’ (Norwegian Centre for Conflict Resolution Report, February 2015), p. 2. 13. Ibid. 14. Celso Marcos Monjane, ‘Rethinking the politics of distribution in Africa: The case of Mozambique’ (Newsletter of the African Political Conference Group, Volume 11, 2015), pp. 5–6. 15. Answers to the question of whether or not the resource curse exists are highly polarized. See Christa N. Brunnschweiler and Erwin H. Bulte, ‘The resource curse revisited and revised: A tale of paradoxes and red herrings’, Journal of Environmental Economics and Management 55 (2008), pp. 248–264; Rod Alence, ‘Where did Africa’s resource curse go?’ (Southern Africa Resource Watch Working Paper, Johannesburg, 2012). 16. Frynas, Wood and Hinks, ‘The resource curse without natural resources’, p. 233. 17. Thomas Selemane, ‘Mozambique political process bulletin’ (Issue 53, Center for Public Integrity and The Association of European Parliamentarians with Africa, 15 February 2013), p. 1. 18. African Development Bank, Organization for Economic Co-operation and Development, United Nations Development Programme, Economic Commission for Africa, ‘African economic outlook 2013: Structural transformation and natural resources’ (OECD Publishing, 2013). 19. United Nations, ‘Natural resource management and extractive industries in Mozambique: A UN Mozambique study’, 2013 <mz.one.un.org/por/content/download/11060/99059/file/Study_Eng.pdf> (21 August 2016). 20. Damião Trape, ‘Recursos naturais são a força motriz da industrialização-Guebuza’, Sapo Notícias, 25 July 2013, <http://noticias.sapo.mz/aim/artigo/843225072013183531.html> (10 October 2016). 21. Southern Africa Resource Center, ‘Managing revenues and optimizing the benefits of coal and gas resources in Mozambique’ (Policy issues note, SARC, November 2013), p. 11. 22. Benedito Cunguara, Gorka Fagilde, James Garrett, Rafael Uaiene and Derek Headey, ‘Growth without change? A case study of economic transformation in Mozambique’, Journal of African Development 14, 2 (2012), p. 105–130. 23. Carlos N. Castel-Branco, ‘Economia extractiva e desafios de industrialização em Mozambique’, in Luis de Brito, Carlos Nuno Castel-Branco, Sérgio Chichava and António Francisco (eds), Economia extractiva e desafios de industrialização em Mozambique (Instituto de Estudos Sociais e Económicos, Maputo, 2010); João Mosca and Tomás Selemane, ‘El dorado Tete: Os mega projectos de mineração’ (Centro de Integridade Pública, Maputo, 2011). 24. Human Rights Watch, ‘What is a house without food? Mozambique’s coal mining boom and resettlement’ (USA, 2013). 25. Lars Buur and Celso M. Monjane, ‘Elite capture and the development of natural resource linkages in Mozambique’, in Melanie Pichler, Cornelia Staritz, Karin Küblböck, Christina Plank, Werner Raza, and Fernando Ruiz Peyré (eds), Fairness and justice in natural resource politics (Routledge, London, 2017), pp. 200–217. 26. The notable exceptions in the general resource-related literature have been Bonnie Campbell, ‘Revisiting the reform process of Africa mining regimes’, Canadian Journal of Development Studies 30 (2010), pp. 197–217; Jonathan Di John, ‘Is there really a resource curse? A critical survey of theory and evidence’, Global Governance 17 (2011), pp. 167–184; and Michael W. Hansen, Lars Buur, Anne Mette Kjær and Ole Therkildsen ‘The economics and politics of local content in African extractives: Lessons from Tanzania, Uganda and Mozambique’, Forum for Development Studies 43, 2 (2016), pp. 221–228. 27. For a critique of this observation, see Di John, ‘Is there really a resource curse?'. 28. Institutional economists suggest that institutions mediate the relationship between natural resources and development outcomes; see, for example, Jeffrey A. Frankel, ‘The natural resource curse: A survey’ (Working Papers no. 15836, National Bureau of Economic Research, 2010); Christa N. Brunnschweiler, ‘Cursing the blessings? Natural resource abundance, institutions, and economic growth’, World Development 36, 3 (2008), pp. 399–419; and Naazneen H. Barma, Kai Kaiser T. Minh Le, Lorena Vinuela, Rents to riches? The political economy of natural resource-led development (World Bank, Washington, 2012). 29. Lindsay Whitfield, Ole Therkildsen, Lars Buur, and Anne Mette Kjær, The politics of African industrial policy: A comparative perspective (Cambridge University Press, New York, NY, 2015). 30. Mushtaq H. Khan, ‘Political settlements and the governance of growth-enhancing institutions’ (Paper prepared for the UK Department for International Development, London, 2010). 31. Lindsay Whitfield and Lars Buur, ‘The politics of industrial policy: Ruling elites and their alliances’, Third World Quarterly 35, 1 (2014), pp. 126–144; Whitfield et al., The politics of African industrial policy. 32. Tom Lavers and Sam Hickey, ‘Conceptualising the politics of social protection expansion in low-income countries: The intersection of transnational ideas and domestic politics’, International Journal of Social Welfare 25 (2016), pp. 388–398. 33. Tom Lavers, ‘Understanding elite commitment to social protection: Rwanda’s Vision 2020 Umurenge Programme’ (UNU-WIDER Working Paper 2016/093, August 2016), p. 2. 34. Tony Hodges and Roberto Tibana, A economia política do orçamento em Moçambique (Principia, Lisboa, 2005); Luís de Brito, ‘Instituições políticas e unidade nacional’, in Luís de Brito, Carlos Nuno Castel-Branco, Sérgio Chichava, Salvador Forquilha, António Francisco (eds), Desafios para Moçambique 2016 (Instituto de Estudos Sociais e Económicos, Maputo, 2016), pp. 23–31. 35. Bernhard Weimer, José Jaime Macuane and Lars Buur, ‘A economia do political settlement em Moçambique: Contexto e implicações da descentralização’, in Bernhard Weimer (ed.), Moçambique, descentralizar o centralismo: Economia política, recursos e resultados (Instituto de Estudos Sociais e Económicos, Maputo, 2012), pp. 31–75. 36. Frelimo, ‘Central Committee Report to the Third Congress of Frelimo’ (Mozambique, Angola and Guine Information Centre, London, 1978), p. 34. 37. Ibid, p. 39. 38. Ibid, p. 38. 39. On the privatization process, see Anne M. Pitcher, ‘Sobreviver à transição: O legado das antigas empresas coloniais em Moçambique’, Análise Social XXXVIII (2003), pp. 793–820; Anne Pitcher, Transforming Mozambique: The politics of privatization 1975–2000 (Cambridge University Press, Cambridge, 2002). 40. Carlos N. Castel-Branco, ‘Estado e a capitalização do capitalismo doméstico em Moçambique’ (Boletim Ideias 73, Instituto de Estudos Sociais e Económicos, Maputo, 10 de Junho 2015). 41. See Anne Pitcher, Party politics and economic reform in Africa’s democracies (Cambridge University Press, Cambridge, 2012). 42. Lars Buur, Carlota Mondlane and Obede Baloi, ‘Strategic privatisation: Rehabilitating the Mozambican sugar industry’, Review of African Political Economy 38, 128 (2011), pp. 235–256; Lars Buur, Carlota Mondlane and Obede Baloi, ‘The white gold: The role of government and state in rehabilitating the sugar industry in Mozambique’, Journal of Development Studies 48, 3 (2012), pp. 349–362. 43. Literally ‘let go’. 44. Magazine Independente, ‘FRENAMO em divórcio incendiário!’, 2 November 2009, <http://macua.blogs.com/moambique_para_todos/2009/11/frenamo-em-div%C3%B3rcioincendi%C3%A1rio.html> (29 April 2017). 45. Carrie Manning, ‘Mozambique’s slide into one-party rule’, Journal of Democracy 21, 2 (2010), p. 152. For different interpretations of the elections see David Pottie, ‘The Mozambique elections: Renamo demands a recount’ (Southern Africa Report, Volume 15, No. 3, May 2000); João M. Cabrita, Mozambique: The tortuous road to democracy (Palgrave-St. Martin’s Press, Basingstoke, 2000); Eric Morier-Genoud, ‘Mozambique since 1989: Shaping democracy after socialism’, in Abdul Raufu Mustapha and Lindsay Whitfield (eds), Turning points in African democracy (James Currey, Oxford, 2009), pp.153–166. 46. Victor Igreja, ‘Amnesty law, political struggles for legitimacy and violence in Mozambique’, International Journal of Transitional Justice 9, 2 (2015), pp. 239–258. 47. See Carlos Nuno Castel-Branco, ‘Aid dependency and development: A question of ownership? A critical view’ (Working Paper No. 01/2008, Institute of Social and Economic Studies, 2008). 48. Buur and Monjane, ‘Elite capture and the development of natural resource linkages in Mozambique’. 49. Todd Chapman, ‘Former foreign minister on internal politics’, 27 February 2009, Wikileaks/Public Library of US Diplomacy, <https://www.wikileaks.org/plusd/cables/09MAPUTO320_a.html> (14 February 2017); Todd Chapman, ‘Mozambican businessman talks of corruption at the highest levels of government’, 28 January 2010, Wikileaks/Public Library of US Diplomacy, <https://wikileaks.org/plusd/cables/10MAPUTO86_a.html> (14 February 2017). 50. See Weimer, Macuane and Buur, ‘A economia do political settlement em Moçambique’. 51. Tony Vaux, Amandio Mavela, Joao Pereira, and Jennifer Stuttle, ‘Strategic conflict assessment Mozambique’ (Commissioned by DFID Mozambique, April 2006); Jason Sumich, ‘The party and the state: Frelimo and social stratification in post-socialist Mozambique’, Development and Change 41, 4 (2010), pp. 679–698. 52. República de Moçambique, ‘Estratégia nacional de desenvolvimento 2015–35’ (Governo de Mozambique, Maputo, Julho 2014). 53. Data from the National Statistics Institute. 54. Authors’ personal notes, 2010. 55. Eneas Comiche, ‘Prestação de contas e transparência na gestão de recursos naturais ponto de vista do parlamento Moçambicano’ (Paper Presented at Nordic-Mozambican conference on inclusive growth, Maputo, 2012). 56. República de Moçambique, ‘Proposta do orçamento do Estado para 2016’ (Governo de Moçambique, Maputo, 2015); Ministério de Planificação e Desenvolvimento and Ministério das Finanças, ‘Cenário fiscal de médio prazo 2007–2009’ (Governo de Moçambique, Maputo, 2006); Ministério de Planificação e Desenvolvimento and Ministério das Finanças, ‘Cenário fiscal de médio prazo 2014–2016’ (Governo de Moçambique, Maputo, 2013). 57. Deloitte, ‘The Deloitte guide to oil and gas in east Africa: Where potential lies’, 2013, <http://www2.deloitte.com/content/dam/Deloitte/global/Documents/Energy-and-Resources/dttl-er-deloitte-guide-oil-gas-east-africa%20–08082013.pdf> (23 June 2017). 58. See Center for Public Integrity, ‘Questions raised about the $400 million payment from ENI’, Newsletter, 18 August 2013, <http://www.cip.org.mz/cipdoc/261_CIP_PressRelease_August_02_en.pdf> (10 October 2013). 59. International Monetary Fund, ‘Republic of Mozambique: Selected issues’ (International Monetary Fund, Washington, 2016). 60. See Center for Public Integrity and Resources for Development Consulting, ‘Rovuma basin economic model’ (Power point presentation, Maputo, 2015). 61. Carlos N. Castel-Branco, Nelsa Massingue, and Carlos Muianga, ‘Padrões de investimento privado e tendências especulativas na economia moçambicana’ (Boletim Ideias 75, Instituto de Estudos Sociais e Económicos, Maputo, 23 Junho 2015). 62. Borges Nhamire, ‘Como as PPP são usadas para beneficiar as elites e prejudicar o Estado: O caso da concessão para fornecimento de chapas de matrículas envolvendo as famíias Guebuza e Machel’ (Investigation report, Centro de Integridade Pública, Maputo, 2015); Borges Nhamire and João Mosca, ‘Electricidade de Moçambique: Mau serviço, não transparente e politizada’ (Investigation report, Centro de Integridade Pública, Maputo, 2014). 63. Castel-Branco, Massingue and Muianga, ‘Padrões de investimento privado e tendências especulativas na economia moçambicana’. 64. Castel-Branco, ‘Estado e a capitalização do capitalismo doméstico em Moçambique’. 65. See Assembly of the Republic of Mozambique, ‘Parliamentary inquiry committee on the situation of the public debt’, Final report, 30 November 2016. 66. Ibid. 67. Ibid. 68. João Mosca and António Francisco, ‘Orçamento: Rectificativo ou ratificativo?’, Savana 1178 (5 August 2016), pp. 14–15. 69. Kroll Inc. was commissioned by the government with funding from Sweden. An edited version of the report executive summary was made public on 24 June 2017, by the Attorney General Office. 70. See Joseph Hanlon and Tim Jones, ‘Debt and development’, in Paul Haslam, Jessica Schafer and Pierre Beaudet (eds), Introduction to international development: Approaches, actors, issues, and practices (Oxford University Press, Ontario, 2017). 71. Prime Minister Carlos Agostinho do Rosário quoted in Joseph Hanlon, ‘Has government said it won’t repay secret debt’, Mozambique News Reports & Clippings, 17 April 2017, <http://www.open.ac.uk/technology/mozambique/sites/www.open.ac.uk.technology.mozambique/files/files/Mozambique_367-17Apr17_Did-govt-say-no-pay_BdM-Consumption-falling.pdf> (23 April 2017). 72. Frynas, Wood and Hinks, ‘The resource curse without natural resources’. © The Author 2017. Published by Oxford University Press on behalf of Royal African Society. All rights reserved This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png African Affairs Oxford University Press

Power, conflict and natural resources: The Mozambican crisis revisited

African Affairs , Volume Advance Article (468) – Jul 31, 2017

Loading next page...
 
/lp/ou_press/power-conflict-and-natural-resources-the-mozambican-crisis-revisited-nouRX91bNW
Publisher
Oxford University Press
Copyright
© The Author 2017. Published by Oxford University Press on behalf of Royal African Society. All rights reserved
ISSN
0001-9909
eISSN
1468-2621
D.O.I.
10.1093/afraf/adx029
Publisher site
See Article on Publisher Site

Abstract

Abstract This article explores how the present socio-economic crisis in Mozambique is linked to the prospects of natural resource windfalls for the country. Drawing on the political settlement approach, it explores how the distribution of power both within and outside the ruling elite is structured and consequently how the underlying political processes have been shaped by the expectations of natural resource windfalls. The article argues that the present socio-economic crisis in Mozambique is not due to national resource assets in themselves. Instead, the political and economic downturn in Mozambique should be understood as a manifestation of how the political settlement has been organized and rent mobilization controlled by the ruling elite. To understand how the prospect of rents from natural resource sectors have influenced the political settlement, we have argued that one has to look at the dynamics of power both within and outside the ruling elite and the incentives they create for elites to use the control of power and access to economic benefits to achieve narrow and short-term gains rather than inclusive and longer-term goals. Armed violence between Frelimo Government forces and fighters loyal to the Renamo leader Afonso Dhlakama resumed in 2012, twenty years after the General Peace Agreement.1 The violence was immediately linked to the discovery of natural resources and the imminent signing of lucrative liquefied natural gas (LNG) contracts.2 A ceasefire allowed elections to take place in October 2014. After on and off talks between Renamo and Frelimo, mediated by an international team, over Renamo’s key demands for the appointment of provincial governors in the resource-rich provinces where it claimed to have won a majority in the national elections, the low-intensity civil war escalated, with government forces systematically destroying Renamo bases. It later transpired that the intensification of the war efforts by the Frelimo government was made possible as the result of undeclared loans worth more than 2 billion USD, which were uncovered in 2013 and 2016 and involved three enterprises linked to the security services.3 In February 2017, the government’s external debt was estimated at 137 percent of its gross domestic product (GDP),4 forcing the government to acknowledge that it would not even be able to service the debt up to 2021.5 The International Monetary Fund (IMF) considered the taking out of secret loans to be a breach of trust. As a result, donor agencies suspended all loans with the Frelimo government immediately, undermining the state budget and causing one of the biggest economic crises in the history of Mozambique. When the Frelimo government and Renamo agreed in 2017 to extend a fragile truce and intensify negotiations, Exxon Mobil finally agreed to buy a stake in the Italian ENI gas fields in Mozambique, releasing up to 350 million USD in capital gains taxes, a third of what it would have paid two years earlier. As the peace accord looked set to last, international donors said they would resume financing the government after an international firm carried out an audit of the secret loans. How did the Frelimo government end up in such a mess? Just a few years earlier it was the darling of the international donor community, and the country was seen as a booming investment destination, referred to as an emerging African lion with consistently high growth rates. Existing explanations of the crisis in Mozambique point to the resource curse phenomenon. Warnings were issued that the Dutch disease may manifest itself, since the exploitation of natural resources invariably results in exchange rate appreciation, which in turn would crowd out and hamper the development of an already low productivity agricultural sector, on which roughly 70 percent of the Mozambique population depends.6 It was also argued that Mozambique would be vulnerable to resource curse effects due to its poor institutional environment and weak ability to enforce formal regulations.7 Others have suggested that, due to the autocratic tendencies of the Frelimo ruling party and the widespread corruption within the ruling elite and state bureaucracy, the discovery and exploitation of natural resources would not only aggravate the greed of these circles, but also fatally affect the country’s political stability.8 In addition, others argued that the strong overlap between the geographic areas that are rich in resources and those where Renamo is strongest in the country would make the latter vulnerable to violent conflicts.9 While these perspectives in the field of ‘expectations of resource booms’10 are based on predictions that the country would suffer from its resource abundance in the future, others are less sanguine and assume that the curse that has crippled other resource-rich countries is now crippling Mozambique.11 It has been claimed that the resumption of armed conflict in Mozambique is a direct consequence of the discovery of natural resources, a link that is viewed as resulting from the failures of the country’s political system, one of ‘winner takes all’.12 This is said to have fueled the conflict because rents are captured by the Frelimo ruling elite,13 thus excluding the Renamo leadership, which is left with very limited if any access to state resources.14 Mozambique’s exploitation of its natural resources has started and will increase over the next decade, but revenues and large windfalls have hitherto not come in such quantities that they have triggered one or a combination of the various defining features ‘often lumped together as the resource curse’.15 The important work by Jedrzej George and colleagues nonetheless suggests that ‘shared aspirations and expectations alone may make for material political and economic outcomes’, even when countries have not experienced the expected resource booms.16 Mozambique largely fits this predicament, with more than 23 billion tonnes in coal reserves and more than 100 trillion cubic feet of gas, placing it among the world’s ten largest coal producers in the near future17 and the holder of the third largest gas reserves in Africa.18 The exploration of these reserves alone triggered increased foreign investments that have moved natural resource exploitation to the forefront of the political agenda, resulting in high expectations of continued and sustainable economic growth that would reduce poverty and aid dependence.19 Inflated expectations were sparked by the speeches of high-ranking political leaders, such as former President Armando Guebuza, stating that the gas and coal bonanza would create local jobs and the country would acquire a new source of revenues, allowing the Frelimo government to transform the economy and invest in the social sector.20 It was also suggested that by 2025, if not before, Mozambique would become a middle-income country.21 However, despite the increasing investments in natural resources, the country still experiences what has been termed ‘growth without change’,22 with resettlement problems,23 low levels of contributions to GDP,24 and poor linkages between megaprojects and other sectors of the economy.25 In this article, we explore how natural resources are enmeshed within the crisis in Mozambique through a detailed analysis of political and economic conditions prior to the boom in natural resource investments, which largely have been overlooked.26 As it is the case for other countries experiencing resource windfalls, explanations of the Mozambican crisis tend either to be lumped together as having been triggered by natural resources and explained in personalized terms as a consequence of the predatory behavior of the political elites and their rent-seeking strategies,27 or related to poor institutions.28 We argue instead that the present crisis is not a particular feature of the natural resource endowment per se, but instead of how expectations of a future resource boom with huge revenue windfalls are related to how the political settlement has been organized and rent mobilization controlled. We argue that the shift towards a natural resources-driven economy and the expectations of windfalls are intensifying the already established features of the distribution of power in society, bringing the settlement close to a breakdown. A key foundation of the political settlement in post-independence Mozambique has been the ideology of ‘national unity’ fostered by the ruling party Frelimo, which informs patronage and clientelism. The shift from a more inclusive and decentralized political settlement after the General Peace Agreement, under President Joaquim Chissano, to a more exclusive and centralized one under President Armando Guebuza, triggered and eventually exacerbated internal and external elite conflicts over power and resources. Amidst an increasing competition between different factions within the ruling coalition, the country was plummeted into one of its biggest crises in history. The analysis is based on extensive knowledge of and experience in researching the political economy of Mozambique over the last fifteen years. We draw on primary sources generated during fieldwork for the Elites Production and Poverty programme (2008–2012) and the Hierarchies of Rights programme (2015–2018), many of which cannot be directly quoted due to the prevailing political climate in Mozambique. We also use a broad reading of the secondary literature to conduct a macro-political settlement analysis. The combined result, we argue, is a new interpretation of the political economy of Mozambique that differs from many of the standard narratives on Mozambique in the African studies literature. The next section presents the key features of the political settlement framework, discusses its key assumptions, and its use to analyse the political economy of natural resources in Mozambique over time. The following sections describe the political settlement during the era of President Joaquim Chissano, explaining how Mozambique went from a broad coalition to internal conflict, and during the period of President Armando Guebuza, explaining how the management of control and rents was centralized and monopolized. The final section offers an analysis of current developments during Filipe Nyusi’s presidency from 2015 to the present, during which the crisis matured. Political settlement analysis and ‘foundational ideas’ The economic structures of developing countries like Mozambique do not create strong incentives to use solely the formal institutional state apparatus to manage the economy and its resources. Developing countries are drawn towards the use of patron–client networks to distribute assets and economic opportunities because they are in the process of capitalist transformation. In such countries only a relatively tiny proportion of the productive population depends on the market for its survival and the reproduction of social relations.29 As productivity is generally low, the formal economy based on taxation cannot, in any sustainable manner, support the distribution of power that has evolved in predominantly capitalist societies. Mushtaq Khan’s concept of political settlement attempts to capture, as a shorthand, the set of institutions and power relations that characterize the social order in a particular country.30 Our use of political settlements draws from the work of Lindsay Whitfield, Lars Buur, and colleagues, which elaborates and extends Khan’s theoretical framework.31 We focus on three structural dimensions of a political settlement in developing countries that are important for understanding Mozambique’s economic and political crisis and how it relates to the prospects of natural resource windfalls. The vertical dimension refers to the organizational structure of patron–client networks and the factions within those networks that are involved in forming a ruling coalition. This dimension is related to two aspects of the ruling coalition that are often confused: first, the distribution of power among elite factions within the ruling elite, where relative exclusion can take place; and second, the distribution of power among higher-level and lower-level factions within the ruling coalition in general. While intertwined in practice, both aspects influence how political leaders respond to natural resource opportunities. The horizontal dimension involves the relative strength of excluded political factions, that is, the organizations or factions outside of the ruling coalition. The distribution of power between the ruling coalition and those factions excluded from it is often easier to explore than what goes on within a ruling coalition. How the relationship between the included and excluded factions is organized based on incorporation, co-option, or cohesion usually has clear historical traces. Similarly, the legitimacy and organizational ability of included and excluded factions are historically grounded. Nevertheless, their relative strength cannot be taken for granted, as will become clear from the Mozambican case. Excluded groups can actually be strong, but can also be effectively repressed, at least temporarily, by the ruling coalition. The third structural dimension of a political settlement is how political regimes are financed, including the question of rents and rent-seeking, as no regime can survive and reproduce itself without resources. In African countries, where the technological capabilities and relative power of capitalist entrepreneurs are generally weak and where most of the resources that can be used to reproduce power emerge from or are organized through the state, the importance of active rent creation is central. For obvious reasons, extractive natural resources such as minerals and oil are attractive to ruling elites. Developing new productive industries and sectors takes much longer and requires more work than granting licenses for exploitation, negotiating capital gains taxes and mobilizing finance from transnational companies. Thus, those firms and economic sectors that provide major sources of state revenues, foreign exchange and rents that allow the ruling elite to reproduce power quickly are important. Ideas and political settlements also interact, and Tom Lavers and Sam Hickey argue this is not always clear in political settlements theory, as they see Khan’s theoretical framework as predominantly interest-based.32 In our understanding of how political settlements are organized and legitimized, we include an analysis of the set of shared and contested ideas underpinning a particular settlement. This is important because structural constraints and the balance of power between contending factions ‘rarely, if ever, determine particular policy responses’.33 We focus specifically on the foundational idea of ‘national unity’ embodied in the ruling Frelimo party, as it continues to inform how patronage and clientelism are organized.34 However, we argue that it does not make much sense to retain the separation of Frelimo’s historically based foundational ideas like ‘national unity’ from the perspective of an interest-based political settlement analysis. Having emerged in the 1960s from a coalition of independence movements, Frelimo celebrated its fiftieth anniversary in 2012. It has been in power since independence in 1975 and has changed from being the Marxist-Leninist vanguard of a one-party state, proclaimed in 1977 but repudiated in 1990, to become the dominant party under the liberal multiparty constitution. In the party’s and ruling elites’ own understanding, Frelimo is the shaper, even the dono do país (owner of the country), of Mozambique’s contemporary history.35 It was initially a vanguard party with a few men at its helm guiding the masses and using the state as their instrument.36 When Frelimo broadened its membership base in the 1980s, the ‘admission of members’ was a strict process where only the select few, those who were considered trustworthy or pessoas de confiança, were admitted.37 Such individuals were ‘entirely dedicated to the cause of the party, the country, the people, and socialism, and who, living exclusively by their labor, both accept and direct their ability and energies to the realization of the party’s statutes and programmes’.38 After 2005, this was reasserted under President Armando Guebuza, who realigned state institutions more with the party after a period of relaxation under President Joaquim Chissano. In the analysis that follows, we use these three structural dimensions and show how they have become intertwined with the idea of ‘national unity’ in order to analyse Mozambique’s recent turn to a natural resource economy and the impact this has had on the country’s recent history. It was during the Chissano and Guebuza periods (1994–2004) that the roots of the present crisis were laid and intensified. The Chissano era: from a broad coalition to internal contestation Chissano’s term of office emerged from the tortuous process of economic and political reforms initiated before the death of Mozambique’s first president, Samora Machel, in 1986, and the country’s adherence to the Bretton Woods Institutions in 1984 following the demise of the socialist model of development. Economic reforms were initiated during 1987–1990 with the adoption of a structural adjustment programme, and political reforms culminated in the enactment of a new multiparty constitution in 1990. In 1992 the government signed a General Peace Agreement with Renamo, which ended the sixteen-year civil war initiated in 1976 (a year after the country’s independence) and paved the way for the 1994 general elections. Joaquim Chissano’s two-term multiparty era started with the first multiparty elections in 1994 and ended in 2004, when he relinquished power in favor of Armando Guebuza, due to internal party pressures. During the Chissano era, peace was consolidated after the devastating civil war, and the country experienced double-digit economic growth and relative political stability, apart from Renamo’s complaints that the elections of 1994 and 1999 were rigged. Chissano’s ruling coalition accommodated the interests of many different factions and individuals both within and beyond the ruling party, although not extensively, as it mostly included the Renamo leader Afonso Dhlakama as part of the peace-building process and a few of Renamo’s generals. The Chissano era has been characterized as one of political liberalization, with less emphasis on the state-party link. Religious groups, ethnic-based lobbying and different economic interests were allowed to flourish, combined with a controlled attempt at decentralizing power after 1996. The more hard-line factions within Frelimo contested this political liberalization, as they saw it as a challenge to the foundational value of ‘national unity’ organized around Frelimo’s continued dominance. Within the ruling coalition, the economic reforms involved a relatively controlled process of primitive accumulation that led to the emergence of a domestic business class closely linked to the Frelimo ruling coalition.39 The privatization process was an important element in creating a national business class linked to the party. Carlos Castel-Branco argues that the process of accumulation by the local elites was led by the state in the post-war period and had two distinct stages. In the first stage, the state empowered local elites, de-capitalized and without business skills and experience, through favorable privatizations of public assets and enterprises. In the second stage, it selected from the previously empowered local elites those whose loyalty was more certain when it came to forming a national oligarchy, which over the last decade has been linked through the Frelimo state to international capital and investments in natural resources.40 Despite one of the most ambitious privatization programmes in Africa, public enterprises continue to be important in financing the political settlement in Mozambique, formally by providing top managerial jobs to key members of the Frelimo coalition, and informally by paying for party activities.41 Chissano’s coalition-building approach was technocratic, consisting of an extensive economic, political and administrative reform agenda and the appointment of technicians with little political influence over government positions. This approach broke with that of previous regimes, which had given precedence to politics over technical matters, and drew heavy criticism from internal party factions linked to Frelimo’s ideological socialist hardliners, liberation fighters and older members of the party. The reform programme included strengthening industrial sectors that could provide returns in terms of rents and political support, such as the labor-intensive and rural-based sugar industry.42 It also involved a ten-year public-sector reform programme launched in 2001, funded by the World Bank and donors, which focused on the administrative apparatus of the state, salary reform, and the introduction of a meritocratic and managerial public service. Some of these measures, such as the fight against corruption, conflicted with the nature of the coalition-building approach. Thus, Chissano’s period was also known for the emergence of corruption in the public sector, as epitomized in the phrase, ‘the goat eats where it is tied’, and for its laissez-faire (deixa andar)43 attitude to public service. The Chissano era became known as corrupt in part because of its departure from Frelimo’s strict ideology of Marxist-Leninism and party discipline, but also because of the broad-based process of accumulation that the regime permitted in order to accommodate social groups and factions that could threaten the ruling Frelimo party electorally, and the fear of losing control and power to Renamo. Another way of distributing power and rents was through the process of decentralization. This process was initiated in 1994 but was reversed after the general elections of 1994 and the unexpected electoral successes of Renamo in the name of maintaining ‘national unity’. The Frelimo-controlled parliament approved instead a controversial law that opted for gradualism in the creation of municipalities, 33 being set up in 1997, and 43 in 2007 and 53 in 2013. After boycotting the first municipal elections, Renamo won five municipalities in 2003, thus introducing the first experience of alternation of power in the country. Decentralization was also important for Frelimo local elites, who from the beginning of the movement’s foundation in 1962 claimed to exert power at the local level but were tamed by the strong discourse of ‘national unity’ and centralized party control. Under Chissano, a deconcentration law was approved in 2003 granting more administrative powers to the provincial and district governments. Regarding the horizontal dimensions of the political settlement during the Chissano era, for Renamo accommodation meant the appointment of district administrators proposed by the party, pecuniary benefits for its leadership and the inclusion of some of its officials in the military command, as provided for in the General Peace Agreement signed in 1992. The post-civil war institutional arrangement also included power-sharing between Frelimo and Renamo, as reflected in the power of parliamentary parties to appoint representatives to the National Electoral Commission, the Constitutional Council, the Council of State and the Security and Defence Council. Parliamentary parties were also entitled to have state subsidies funded from the public budget, calculated according to their parliamentary representation. These arrangements provided some level of participation and access to state resources for Renamo cadres under an arrangement sometimes called ‘Frenamo’.44 However, the process of resource distribution to Renamo was clientelistic in nature, bolstering the authoritarian and centralizing leadership style of its leader Afonso Dhlakama and generally benefitting Frelimo. The resources made available to Renamo and Dklakama through the peace agreement and the practice of parliamentary inclusion did nothing to contribute to the institutionalization of Renamo, nor did it benefit a wider range of the movement’s membership. Despite its repression of Renamo, competitive elections proved to be one of the main causes of fear for the ruling party coalition. In the first post-war democratic elections in 1994, Chissano took 53 percent of the votes, while Dhlakama received around 33 percent. Renamo won in the central and northern provinces, but due to the winner-takes-all system, Chissano could maintain control. This was repeated in the 1999 general elections, when Chissano again won the presidency, but with a very slim margin of 4 percent over the Renamo candidate. However, there have been strong suggestions of electoral fraud and even that the elections for both president and parliament were rigged, indicating frail political support for Frelimo among the population, particularly in the central and northern regions where Renamo won comfortably.45 A substantial part of Renamo’s guerrilla forces has not been integrated into the regular army, allegedly because it was agreed in the General Peace Agreement that Renamo could keep a certain number of its forces under its control to protect its leaders. This residual force came to play a role in the political crisis that erupted in 2012, following the approval by Frelimo of a contested electoral law and Renamo’s allegations that its ex-guerrilla fighters were being compulsorily retired from the army and kept from high-ranking positions within the security establishment.46 Despite the relatively inclusive nature of Chissano’s government, Renamo was systematically excluded from access to resources such as jobs in public enterprises, benefits from privatizations and access to opportunities for linkages with megaprojects and other foreign direct investment projects. The main source of accommodation for Renamo was through the narrow opportunities for participation in Parliament and its ability to propose appointees. Renamo was facing a process of weakening, which accelerated after 2004 when Guebuza took control. Moreover, below the thin veneer of tolerance and inclusiveness under Chissano, the harassment of opposition party members did not cease, especially locally, and access to high-ranking public positions and economic opportunities was still substantially dependent on the ruling party’s blessing. At the beginning of the democratic era in 1994, Chissano had assumed that the controlled but limited inclusion of Renamo would reduce the movement’s influence over the years. A decade later, Guebuza would consider this slow-death strategy too slow, given the need to accelerate control over the territory in light of the prospect of opening up natural resource exploitation in former Renamo areas. One dramatic consequence of Guebuza’s strategy would paradoxically trigger a ‘second coming’ for Renamo. Turning to the last dimension of the political settlement, the financing of the regime took place within the context of a reduced national tax base. Development aid accounted for more than 50 percent of the state budget, and the Chissano government’s ‘controlled’ economic and political reforms provided a strong argument that allowed the country to benefit from the Highly Indebted Poor Countries Initiative, leading to a substantial reduction of its external debt. Public procurement and access to natural resource licenses (land, mineral resources) and to public office and employment, all based on the demands of party loyalty, provided the ruling elite with rents with which to buttress the regime. However, this system of rent distribution, although relatively wide in terms of intra-elite inclusion, was centralized and based on the direct control of the state apparatus, which provided considerable power to the president and his technocracy. This contributed to sidelining the lower-level factions within the Frelimo party, mostly in the provinces. Chissano’s reliance on the technocracy came at the expense of reducing the party’s influence over decisions on rents and rent accumulation. Moreover, the high level of dependence on development aid implied the existence of an accountability architecture that forced the government to be more accountable to donors than to national constituencies, society or the party.47 This was heavily criticized inside the party, and, combined with its narrow victory in the 1999 general elections, paved the way for pressures for a new leadership that would privilege party politics over technocracy, and national (party) accountability over accountability to external actors, i.e. preserving Frelimo members’ privileges. It was under Chissano’s regime that foreign direct investment inflows started with large-scale investments, such as the Mozambique aluminum project (Mozal) and the production of natural gas in Pande and Temane by Sasol, which included building a pipeline from Temane in southern Mozambique to Secunda in South Africa. It was also under Chissano that negotiations to resume coal mining were initiated, with Riversdale buying one of the first concessions very cheaply. It was later sold on to Rio Tinto without the state having any legislation in place to tax the capital gains. However, these efforts provided narrow and dispersed opportunities for rents. Mozal, for example, had a project to create linkages with local companies that only benefitted a few members of the political elite.48 Some of this group also seems to have benefitted from the building of the Sasol pipeline, but exactly how is still unclear. During the Chissano era, decision-making power was not only shifted from the party to the state technocracy, which was the gatekeeper for access to rents. The state technocrats also benefitted from their proximity to President Chissano, whence the increasing levels of corruption. It was in this context that Armando Emilio Guebuza was elected secretary-general of the party in 2002 and confirmed as Frelimo’s candidate for the 2004 general elections. Guebuza adopted the fight against corruption of the Chissano era as a platform for his political campaign in the 2004 elections, which he won by a large margin. The Guebuza era: from internal contestation to monopolization In his bid to consolidate the leadership of the party and be elected president, Guebuza revitalized the party structures (party cells) from the grassroots up to the highest levels of the state and parastatal companies. He also promoted the provincial party secretaries as local political leaders, restored the notion of Frelimo as the ruler of the country, and revived the foundational idea of national unity organized in and through the Frelimo party. Party membership as a condition for access to public office was promoted at all levels of society, with an increasing number of party cells in public entities. Mobilization also increased, and in 2012 the general-secretary of Frelimo claimed that the party had more than four million members, although it has never attained such a figure in any election, despite the statutory obligation for members to vote in all elections. This strengthening of the party was important for the balance of power both within and outside the ruling coalition. Within the ruling coalition, Guebuza positioned himself in opposition to the Chissano era’s inefficient public service, with its widespread corruption, and he promised to boost national capitalist development. This did not necessarily mean that all factions within the ruling party coalition would benefit, since the available economic opportunities were distributed among the narrow coalition of those closest to Guebuza, who controlled the economy and rents with a tight fist, in contrast to the Chissano era.49 Despite the revitalization of the party down to the local level, the distribution of power was highly unequal between the top and bottom of the party structure. Within the party, control was effective but not simple, since there were conflicts between higher- and lower-level factions. Frelimo has historically used influential political figures, among them liberation struggle veterans, to exert control over the local-level membership, and even state structures. During elections for both party and public positions, central brigade leaders were responsible to the Frelimo top leadership for ensuring that the candidates preferred by the party leadership at the central level were selected (and elected) in local elections. This created discontent locally, with people staying away from elections. The imbalance of power was also reflected in the way power was distributed within the state and the process of decentralization. Municipalities run by the opposition complained that the central government was blocking their bids to take over the functions provided for in the legislation. The budgetary allocation across government layers also showed the asymmetry of power, with up to 50.3 percent of recurrent expenditure located in the central government between 2010 and 2014, and over 80 percent of capital expenditure, including expenditure for projects implemented at the local level. This has implications for the resources available for the decision-making of local party factions as well as opposition-run municipals, and consequently for their relative power to influence politics. Concerning the horizontal dimension, which is where the relative strength of the excluded factions is located, Frelimo’s and Guebuza’s main political adversaries, Dhlakama and Renamo, were defeated in the 2004 general elections by a margin of more than 30 percent. This margin allowed Guebuza to dismiss their importance as political interlocutors, to cease the limited flow of resources initiated during the Chissano era, and to initiate the sidelining of Renamo’s former guerrillas in the armed forces. Frelimo’s victory in the 2009 elections, which gave it more than two-thirds of the parliamentary seats and more than 70 percent of the vote for Guebuza, consolidated his success as a powerful president and party leader. However, this electoral success was misleading, since it was based on a low voter turnout and increased control over those voting. The turnout for national elections went from 87.9 percent in 1994 to 69.5 percent in 1999 to a low of 36.42 percent in 2004, before rising a little to 44.4 percent in 2009.50 Access to state resources and the capacity to win all parliamentary and presidential elections since coming to power and thus to provide positions and opportunities for organized members consolidated Guebuza’s control not only over the party but also society. However, this did not go unchallenged. Outside the party, Renamo resorted to war after its claims to keep its members in the armed forces and under its command were no longer accepted by the Guebuza government, and after disagreement over a new electoral law approved unilaterally by the Frelimo caucus in Parliament in 2012. In 2014, an agreement between Renamo’s leader and Guebuza allowed Renamo to participate in the 2014 elections, in which its presidential candidate, Afonso Dhlakama, and the party contested the results and had considerable success. Renamo claimed that they had won in six provinces, though more likely in four, in which they demanded to appoint their governors or else they would rule by force. Frelimo and the government refused to respond positively to Renamo’s demands, arguing that it would undermine ‘national unity’. After failed negotiations and alleged attempts by the national army to assassinate Dhlakama, the country was plunged into a military crisis affecting mostly the central region provinces. Regarding the third structural dimension of the political settlement, the financing of the regime, under Guebuza this still relied on donor funding, but there was a greater incentive to achieve autonomy. However, this autonomy was not directed towards following a clear development agenda, as the international development partners wanted. As noted, the logic of the Frelimo state was now one of a bureaucracy more concerned with control than effectiveness.51 Even party members were expected to show strict loyalty to the party, which insisted that those holding state positions were above all accountable to the party. This was not only a political matter, since it also had implications for the economy. The focus on short-term economic gains prevented the government from investing in a consistent strategy to create or strengthen the productive sectors. The country lacked a coherent development strategy apart from poverty reduction action plans, whose scope was narrower. Only in July 2014 did the government adopt the National Development Strategy (NDS) 2015–2035, whose main objective is to improve the living standards of the population through the structural transformation of the economy and the expansion and diversification of the production base: industrialization and agriculture being at the core of the programme.52 The NDS acknowledges that the national economy generally and the rural economy in particular depend heavily on the exploitation and use of natural resources. While the NDS signals that natural resources will be key sources of rents and economic accumulation in the future, their contribution to the economy thus far has been low. According to official data, the contribution of the extractive industry to GDP is increasing at very impressive rates, from 2.2 to 5.1 in the 2011 to 2015 period, but its share is still very low.53 The growth in the extractive sector results from a combination of global and national factors. Changes in the global economy during the 2000s made it highly profitable for foreign companies to develop Mozambique’s natural resource endowment, so there was a clear pull factor. But for the Guebuza ruling elite there was also the push factor of breaking free from aid dependence when criticisms of elections became linked to aid conditions, as when Guebuza was re-elected in 2009. After the general election, in 2010 this culminated in a group of donors forcing through a ‘donor boycott’ involving the suspension of general budgetary support until a set of targets for reforming the biased electoral and parliamentary system had been agreed.54 The combination of revenue sources initially gave the Frelimo party and the Guebuza administration ample room to breathe: natural resource licenses, capital gains taxes (in force since 2013), continued high levels of aid disbursement despite donor criticism, new loans from both the World Bank and various emerging powers like India, China and Brazil, and sovereign bond flotations. The negotiations over major natural resource investments were generally clouded in secrecy, with limited or zero state knowledge or parliamentary participation and oversight.55 Another way to reduce donor dependence and influence was through the selective promotion of pockets of efficiency in key sectors such as state revenues, where special incentives were created to strengthen tax administration. Over the last ten years, this has been responsible for a remarkable increase in revenue collection to reduce dependence on foreign aid, with implications for the government’s increasing ability to resist donor leverage. Thus, between 2005 and 2015 internal revenues increased from 57.8 percent to 75 percent (and state revenues from 56.7 percent to 71 percent), while foreign aid grants decreased from 28.4 percent to 9 percent of total budget revenues.56 The National Institute for Petroleum is another example of creating potential pockets of efficiency for rent management related to the new extractive sectors. But excellence was not necessarily translated into following the rules during the Guebuza era. The sale of a stake in the ENI gas concession in 2013 is a good example of how capable pockets of efficiency can exist formally but only be used selectively. When ENI announced the sale of part of its stake in the Rovuma Basin to the China National Petroleum Corporation, it was, as the regulation suggests, subject to the approval of the Mozambican government. Protracted discussions with the government and ultimately President Guebuza were centered directly on determining the amount of capital gains tax to be paid on the $4.21 billion sale. The Mozambican parliament had by then approved a capital gains tax of 32 percent, but the president had not formally signed the law into effect.57 By negotiating directly with the president, ENI managed to reduce the tax to $400 million, or a rate of 9.5 percent instead of 32 percent. This followed similar decisions on capital gains tax that saw a far lower rate of revenues entering the state’s coffers than the capital gains tax stipulated,58 with continued speculation that the Guebuza regime was using the executive’s discretionary power to accumulate and fund the party and personal ambitions. Therefore, the beginning of the exploitation of natural resource projects, mostly coal and the huge gas discoveries in the Rovuma basin, provided the necessary elements for the expansion of formal and informal rents. Although the contribution of gas and mining to state revenues remained low, two elements in particular held out the prospects of high rents. The first was the optimistic projection that the country would be a major LNG exporter, with a considerable level of fiscal revenues entering the state’s coffers over the next decade, amounting to US$ 500 billion by 2045,59 although these figures are contested.60 The second is that capital gains taxes started to flow after some companies began selling their assets in Mozambique. The process started during Chissano's mandate with the purchase of Riversdale’s coal mines by Rio Tinto in 2011, which until then had not been taxed. The government discovered that these transactions could be taxed, introduced a legislative revision in 2013, and collected revenues in subsequent transactions of natural resource assets totaling US$ 1.3 billion. However, although Mozambique has been compliant with the Extractive Industries Transparency Initiative since 2012, it is not clear how these revenues were used. Growing internal revenues and the availability of resource rents during Guebuza’s second term were also combined with an increase in non-concessional loans (internal and external) to offset the reduction in development aid. Reduced aid inflows stemmed from the global financial crisis and changing donors’ policies, among them the weakening enthusiasm for direct budget support, emerging ideas of aid for trade and a focus on business development. The natural resources boom stimulated the inflow of foreign direct investments in the country and the interests of the emerging economies in the areas of mining, gas and agriculture, with sizeable investments from China, Brazil and India. However, foreign direct investments were concentrated in some areas and were capital- rather than labour-intensive. As has been observed, thirteen megaprojects accounted for 58 percent of the private investment approved between 1990 and 2012 (of US$ 20.2 billion), which employed 20,000 people.61 In summary, under Guebuza’s mandate rent distribution continued, mediated by the control of the state apparatus as the gatekeeper for public procurement, public-private partnerships and contracts with public enterprises.62 An additional element brought about by the national resources boom was access to foreign direct investment geared towards the extractive industries, with low linkages to the local economy, no relevance for the development of productive sectors and low employment rates.63 Moreover, the few linkages established depended on previous experience and institutional capacity, which was only available to those companies that were part of the state’s supply chain or other megaprojects. As a result, a vicious cycle emerged regarding those who benefited from the extractive economy, namely those with political and institutional links that favoured access to such projects.64 It was during this period of feverish expectations of LNG revenues that Guebuza had to relinquish power, having fulfilled two terms in office. The emergence of Nuysi: from bad to worse? Guebuza did not want to give up power just when the vast natural resource endowment finally seemed secure. His power was consolidated at the tenth party congress in 2012, despite increasing dissatisfaction within the ruling party, mostly because of his centralized control over rent opportunities and the prospective opportunities emerging from the state and the new extractive economy. Channeling rents or access to rent opportunities to his closest allies, Guebuza succeeded in securing the election of members loyal to him in the party’s decision-making structures: the Central Committee and the powerful Political Commission. But the combination of internal and external dissatisfaction with Guebuza’s monopolization of the prospective resources led to increasing opposition within the party that heavily influenced the process of choosing Frelimo’s new presidential candidate and ultimately the president of the party. Guebuza had first tried to appoint a loyal successor, but failed to do so because his opponents, mostly the liberation struggle veterans and more liberal members, rallied to support the candidacies of Filipe Nyusi and the former prime minister and Chissano protégé Luisa Diogo. The latter had been aligned to Guebuza as defence minister, and it was during his time as minister that the conflict with Renamo had escalated and a series of secret loans for defence purposes were taken out.65 Nyusi was supported by longstanding military figures of the gas-rich province of Cabo Delgado, chief among them the former Minister of Defence, Joaquim Chipande. Nyusi defeated Luisa Diogo in the internal elections, received the party’s nomination and was elected President of the Republic in the 2014 general elections. In March 2015, Guebuza’s opponents succeeded in electing Nyusi, as president of the party, in line with the Frelimo tradition of combining the presidency of both country and party, using the argument that this avoided fragmenting the political power underpinning the quest for ‘national unity’ in and through the Frelimo party. But Nyusi does not control the Central Committee or the Political Commission, making it difficult for him to assert his formal power. This might change in the eleventh congress of Frelimo being held in September 2017. Increased foreign direct investments and the prospects for natural resource revenues also stimulated the increase in public indebtedness, including so-called secret and illegal debts, amounting to more than $2 billion, which funded dubious investments in the tuna fishing company EMATUM, PROINDICUS for maritime security and Mozambique Assets Management for oil and gas logistics between 2012 and 2014.66 Most of these loans, it was eventually revealed, were channeled into purchasing military and surveillance equipment, and it is suspected that some of them were used for the personal benefit of members of the ruling elite. Between 2009 and 2014 the government also arranged another undisclosed loan of $221 million for the Ministry of Interior, whose objectives are still unknown,67 besides accruing a massive internal state debt because bills for recurrent costs, construction, medicine, ammunition and weaponry over the last years were not honored. This public indebtedness was only possible due to the guarantees of the availability of natural resource rents in what was expected to be the near future, possibly post-2025. This level of public indebtedness results in part from the relative success of the country in reducing its dependence on donors and their leverage over its policy decisions. It also reflected the schisms within the ruling elite and its centralization of power, in the sense that only a reduced number of individuals close to the president and inside the Frelimo Party, the government and parliament knew what was going on. The government even bypassed parliament in the process, which is supposed to grant authorization for loans beyond the levels approved in the state budget. However, donors and international financial institutions such as the World Bank and the IMF are still important sources of state revenues, and they influence investors’ decisions. The discovery of these concealed debts angered the Bank and Fund, who decided to suspend their support to the state budget until an investigation into and audit of these loans were carried out. This plunged the country into a serious budgetary and foreign exchange crisis that affected the prospects of economic growth, with implications for state revenues. Without resources from its international partners, amounting to around $500 million annually, the government was forced to revise the approved budget for 2016 and introduce austerity measures. Despite government claims that social policies would not be affected, the revision of the budget resulted in cuts in social expenditures and increases in defence and security spending.68 The government was initially reluctant to allow a forensic audit as demanded by the IMF and other development partners, which would reveal the real beneficiaries of the concealed loans and end with them being taken to court if necessary to satisfy the demands of the country’s development partners. Eventually the government agreed to an independent audit being carried out by the international company Kroll Inc., which, however, faced difficulties in assessing information, delaying the delivery of results from February to May 2017.69 The secret loans were taken out during the last years of Guebuza’s term, with Nyusi as an insider. However, it involved a narrower group of the ruling elite, creating an oligarchy, with Guebuza at its forefront. Power struggles within the ruling coalition, along with uncertainties regarding the securing of property rights in the future, began to challenge the concentration of power within Frelimo. This is evident under the Nyusi government, with persistent appeals of Frelimo senior members to Guebuza to allow more space for the new president to deal with the impending problems the country faces. It was only when the regime had its back against the wall and was facing the full effects of a financial meltdown that there was a breakthrough in the civil war. International mediators, demanded by Renamo, initially failed to negotiate a proposal for a high-level meeting between Nyusi and Dhlakama with an agenda that included decentralization and the integration of Renamo members into the army, as well as their disarming. As these efforts failed and it became clear that Mozambique would default on all its secret loan payments from the end of 2016, the regime stood on the brink of defaulting altogether, as the state coffers were empty, and the IMF demanded the withholding of all international aid. In late December 2016, Nyusi and Dhlakama agreed on a truce for the festive season, later extended indefinitely, with promises by both parties to reach a peace agreement and end the conflict. Frelimo was on the brink of breaking up, but as the ruling party has done every time it has faced possible defeat, its leadership and the different factions stuck together, since they know that their survival depends on the unity of the party and its control over the state and society. This became clear in early 2017, when Frelimo probably could have challenged the two banks that initially granted the key secret loans with relative ease in the London courts. The two banks, Credit Suisse and the Russian bank VTB, had clearly failed to do due diligence, thus making the loans ‘illegitimate debt’ or a type of ‘odious debt’, as lenders have a fiduciary responsibility to borrowers regarding ‘loan pushing’.70 However, following this route would have required the Frelimo government to make public all information about who took out the loans and benefitted from them. This would have implicated the leaderships around Nyusi and Guebuza and possibly have split the party. In January 2017, the Mozambican administrative tribunal declared that the secret debts were both unconstitutional and contrary to budgetary laws, making the secret loans illegal. However, since April 2017 the loans have been included retrospectively in the 2015 State General Accounts report, since they were ‘proven to be in the public interest’.71 Through this move the state accepted the guarantees and will pay to relinquish individual responsibility, making legal what was formerly illegal. The paradox is that the secret and illegal debts, which were based on the prospect of natural resource windfalls, became legal overnight through their inclusion in the state budget, being passed on to the people of Mozambique in the name of maintaining party unity. The state guarantee is therefore more than just a guarantee of debt, it is also what allowed the civil war to end, as the war-mongers among the ruling Frelimo elite were protected from financial liabilities. The move also allowed the continued drive towards maintaining ‘national unity’ organized in and around the continued dominance of the Frelimo party, not only as the sole legitimate government of the country, but also as the continued guarantor of peace, at least in the short term. Conclusion So far expectations of future rents is the only significant factor stemming from the natural resources boom, but natural resource rents per se do not explain the current situation in which the ruling elites tend to favor more short-term gains, instead of more developmental policies, like the diversification of the economy and the building of strong productive sectors. To understand how the prospect of rents from natural resource sectors have influenced the political settlement, we have argued that one has to look at the dynamics of power both within and outside the ruling elite and the incentives they create for elites to use the control of power and access to economic benefits to achieve narrow and short-term gains rather than inclusive and longer-term goals. In this article, we have sought to contribute to the discussion about Mozambique’s crisis by showing how the expectations related to the natural resource boom that contributed to the crisis are enmeshed in complex political processes. Theoretically, the article draws on the political settlement approach, which offers a comprehensive framework for understanding how the distribution of power is organized and potentially changes over time. We argue that the prospect of natural resource rents exacerbated already existing power struggles and distribution of resources within and outside the ruling elite, informed by the continuation of the ideology of ‘national unity’, which in the last twenty years changed and evolved from a moderately inclusive political settlement in the Chissano’s era to an exclusive one in the Guebuza era. This exacerbated internal and external conflicts and competition over power culminating under Nyusi's presidency with the Frelimo party struggling to stay united for political survival. Internal ruling elite struggles over access to potential short term resource rents have limited Nyusi's capacity to reach an agreement with Renamo and to tackle the root causes of the political crisis with Renamo. The argument in this article, while clearly drawing on previous scholarship on political settlements as well as discussions of the resource curse phenomenon, makes a number of original contributions. First, where Jedrzej Frynas, Geoffrey Wood and Timothy Hinks add expectations of resources to the list of the effects of the resource curse, they do not relate this to how expectations of future natural resources become intertwined with the organization of and possible changes to the evolution of political settlements or of the political order.72 This article directly links the question of expectations related to resource windfalls to the evolution of the distribution of power over time. We argue that, as expectations increased, it put severe stress on the ability to maintain stability in the organization of power, as different factions juggle for access to prospective resources. Second, whereas much of the existing literature in African studies on political settlements is rooted in sectoral analysis, at its core the political settlements approach is a macro-level form of study. This article provides a macro-analysis of the distribution of power and what constitutes such power in societies, linking this analysis to expectations of natural resource rents. Finally, this article specifically engages with a broader discussion of how ideas and ideology intersect with a political settlements analysis. We argue that, in the case of Mozambique, historically based foundational ideas like ‘national unity’, linked to the continued dominance of the Frelimo party as providing the sole legitimate government of the country, cannot be separated from a more interest-based political settlement analysis, as the two are intertwined in practice. Footnotes 1. The 1992 General Peace Agreement ended a sixteen-year insurgency by Renamo against the Frelimo government. 2. Adrian Frey, ‘War in peace: The return of civil war in Mozambique’, Club of Mozambique, 3 May 2016, <http://clubofmozambique.com/news/opinion-war-in-peace-the-return-of-civil-war-in-mozambique/> (11 October 2016). 3. For the main facts of the debt crisis, see Sapo Notícias, ‘Cronologia: Da Ematum à incorporação das dívidas nas contas de Moçambique’, 18 April 2017, <http://noticias.sapo.mz/info/artigo/1502039.html> (30 April 2017). 4. Adrian Frey, ‘Standard & poor’s: Mozambique’s debt to fall from 137% to 118% by 2020’, Club of Mozambique, 12 February 2017 <http://clubofmozambique.com/news/standard-poors-mozambiques-debt-fall-137-118-2020/> (18 April 2017). 5. Frances Coppola, ‘Mozambique is about to default on its ‘tuna bond’, Forbes Investing/Foreign Affairs, 17 January 2017, <https://www.forbes.com/sites/francescoppola/2017/01/17/mozambique-is-about-to-default-on-its-tuna-bond-again/#585fc8e1431a> (1 May 2015). 6. David Rosenfield, ‘The coal mining sector in Mozambique: A simple mode of predicting government revenue’ (Conference Paper No. 9, Institute of Social and Economic Studies, September 2012). 7. Cayley Green and Lisa Otto, ‘Resource abundance in Mozambique: Avoiding conflict, ensuring prosperity’ (Occasional Paper, South African Institute of International Affairs, February 2014). 8. Boniface Mabanza, ‘Mozambique at a crossroads: The natural resources trap’ (Kirchliche Arbeitsstelle Südliches Afrika, 2013) <http://www.woek.de/web/cms/upload/pdf/kasa/publikationen/Mabanza_2013_Mozambique_at_the_crossroad.pdf> (8 October 2016). 9. Green and Otto, ‘Resource abundance in Mozambique’. 10. Jedrzej George Frynas, Geoffrey Wood and Timothy Hinks, ‘The resource curse without natural resources: Expectations of resource booms and their impact’, African Affairs 116, 463 (2017), pp. 233–260. 11. Peter Fabricious, ‘The resource curse comes to Mozambique: Mozambique is teetering on the brink of a major sovereign debt default, which has economic, developmental and security implications’, Institute for Security Studies, 9 June 2016, <https://www.issafrica.org/iss-today/the-resource-curse-comes-to-mozambique> (16 August 2016). 12. James Gordon, ‘Curse or cure? The political implications of Mozambique’s natural resources’ (Norwegian Centre for Conflict Resolution Report, February 2015), p. 2. 13. Ibid. 14. Celso Marcos Monjane, ‘Rethinking the politics of distribution in Africa: The case of Mozambique’ (Newsletter of the African Political Conference Group, Volume 11, 2015), pp. 5–6. 15. Answers to the question of whether or not the resource curse exists are highly polarized. See Christa N. Brunnschweiler and Erwin H. Bulte, ‘The resource curse revisited and revised: A tale of paradoxes and red herrings’, Journal of Environmental Economics and Management 55 (2008), pp. 248–264; Rod Alence, ‘Where did Africa’s resource curse go?’ (Southern Africa Resource Watch Working Paper, Johannesburg, 2012). 16. Frynas, Wood and Hinks, ‘The resource curse without natural resources’, p. 233. 17. Thomas Selemane, ‘Mozambique political process bulletin’ (Issue 53, Center for Public Integrity and The Association of European Parliamentarians with Africa, 15 February 2013), p. 1. 18. African Development Bank, Organization for Economic Co-operation and Development, United Nations Development Programme, Economic Commission for Africa, ‘African economic outlook 2013: Structural transformation and natural resources’ (OECD Publishing, 2013). 19. United Nations, ‘Natural resource management and extractive industries in Mozambique: A UN Mozambique study’, 2013 <mz.one.un.org/por/content/download/11060/99059/file/Study_Eng.pdf> (21 August 2016). 20. Damião Trape, ‘Recursos naturais são a força motriz da industrialização-Guebuza’, Sapo Notícias, 25 July 2013, <http://noticias.sapo.mz/aim/artigo/843225072013183531.html> (10 October 2016). 21. Southern Africa Resource Center, ‘Managing revenues and optimizing the benefits of coal and gas resources in Mozambique’ (Policy issues note, SARC, November 2013), p. 11. 22. Benedito Cunguara, Gorka Fagilde, James Garrett, Rafael Uaiene and Derek Headey, ‘Growth without change? A case study of economic transformation in Mozambique’, Journal of African Development 14, 2 (2012), p. 105–130. 23. Carlos N. Castel-Branco, ‘Economia extractiva e desafios de industrialização em Mozambique’, in Luis de Brito, Carlos Nuno Castel-Branco, Sérgio Chichava and António Francisco (eds), Economia extractiva e desafios de industrialização em Mozambique (Instituto de Estudos Sociais e Económicos, Maputo, 2010); João Mosca and Tomás Selemane, ‘El dorado Tete: Os mega projectos de mineração’ (Centro de Integridade Pública, Maputo, 2011). 24. Human Rights Watch, ‘What is a house without food? Mozambique’s coal mining boom and resettlement’ (USA, 2013). 25. Lars Buur and Celso M. Monjane, ‘Elite capture and the development of natural resource linkages in Mozambique’, in Melanie Pichler, Cornelia Staritz, Karin Küblböck, Christina Plank, Werner Raza, and Fernando Ruiz Peyré (eds), Fairness and justice in natural resource politics (Routledge, London, 2017), pp. 200–217. 26. The notable exceptions in the general resource-related literature have been Bonnie Campbell, ‘Revisiting the reform process of Africa mining regimes’, Canadian Journal of Development Studies 30 (2010), pp. 197–217; Jonathan Di John, ‘Is there really a resource curse? A critical survey of theory and evidence’, Global Governance 17 (2011), pp. 167–184; and Michael W. Hansen, Lars Buur, Anne Mette Kjær and Ole Therkildsen ‘The economics and politics of local content in African extractives: Lessons from Tanzania, Uganda and Mozambique’, Forum for Development Studies 43, 2 (2016), pp. 221–228. 27. For a critique of this observation, see Di John, ‘Is there really a resource curse?'. 28. Institutional economists suggest that institutions mediate the relationship between natural resources and development outcomes; see, for example, Jeffrey A. Frankel, ‘The natural resource curse: A survey’ (Working Papers no. 15836, National Bureau of Economic Research, 2010); Christa N. Brunnschweiler, ‘Cursing the blessings? Natural resource abundance, institutions, and economic growth’, World Development 36, 3 (2008), pp. 399–419; and Naazneen H. Barma, Kai Kaiser T. Minh Le, Lorena Vinuela, Rents to riches? The political economy of natural resource-led development (World Bank, Washington, 2012). 29. Lindsay Whitfield, Ole Therkildsen, Lars Buur, and Anne Mette Kjær, The politics of African industrial policy: A comparative perspective (Cambridge University Press, New York, NY, 2015). 30. Mushtaq H. Khan, ‘Political settlements and the governance of growth-enhancing institutions’ (Paper prepared for the UK Department for International Development, London, 2010). 31. Lindsay Whitfield and Lars Buur, ‘The politics of industrial policy: Ruling elites and their alliances’, Third World Quarterly 35, 1 (2014), pp. 126–144; Whitfield et al., The politics of African industrial policy. 32. Tom Lavers and Sam Hickey, ‘Conceptualising the politics of social protection expansion in low-income countries: The intersection of transnational ideas and domestic politics’, International Journal of Social Welfare 25 (2016), pp. 388–398. 33. Tom Lavers, ‘Understanding elite commitment to social protection: Rwanda’s Vision 2020 Umurenge Programme’ (UNU-WIDER Working Paper 2016/093, August 2016), p. 2. 34. Tony Hodges and Roberto Tibana, A economia política do orçamento em Moçambique (Principia, Lisboa, 2005); Luís de Brito, ‘Instituições políticas e unidade nacional’, in Luís de Brito, Carlos Nuno Castel-Branco, Sérgio Chichava, Salvador Forquilha, António Francisco (eds), Desafios para Moçambique 2016 (Instituto de Estudos Sociais e Económicos, Maputo, 2016), pp. 23–31. 35. Bernhard Weimer, José Jaime Macuane and Lars Buur, ‘A economia do political settlement em Moçambique: Contexto e implicações da descentralização’, in Bernhard Weimer (ed.), Moçambique, descentralizar o centralismo: Economia política, recursos e resultados (Instituto de Estudos Sociais e Económicos, Maputo, 2012), pp. 31–75. 36. Frelimo, ‘Central Committee Report to the Third Congress of Frelimo’ (Mozambique, Angola and Guine Information Centre, London, 1978), p. 34. 37. Ibid, p. 39. 38. Ibid, p. 38. 39. On the privatization process, see Anne M. Pitcher, ‘Sobreviver à transição: O legado das antigas empresas coloniais em Moçambique’, Análise Social XXXVIII (2003), pp. 793–820; Anne Pitcher, Transforming Mozambique: The politics of privatization 1975–2000 (Cambridge University Press, Cambridge, 2002). 40. Carlos N. Castel-Branco, ‘Estado e a capitalização do capitalismo doméstico em Moçambique’ (Boletim Ideias 73, Instituto de Estudos Sociais e Económicos, Maputo, 10 de Junho 2015). 41. See Anne Pitcher, Party politics and economic reform in Africa’s democracies (Cambridge University Press, Cambridge, 2012). 42. Lars Buur, Carlota Mondlane and Obede Baloi, ‘Strategic privatisation: Rehabilitating the Mozambican sugar industry’, Review of African Political Economy 38, 128 (2011), pp. 235–256; Lars Buur, Carlota Mondlane and Obede Baloi, ‘The white gold: The role of government and state in rehabilitating the sugar industry in Mozambique’, Journal of Development Studies 48, 3 (2012), pp. 349–362. 43. Literally ‘let go’. 44. Magazine Independente, ‘FRENAMO em divórcio incendiário!’, 2 November 2009, <http://macua.blogs.com/moambique_para_todos/2009/11/frenamo-em-div%C3%B3rcioincendi%C3%A1rio.html> (29 April 2017). 45. Carrie Manning, ‘Mozambique’s slide into one-party rule’, Journal of Democracy 21, 2 (2010), p. 152. For different interpretations of the elections see David Pottie, ‘The Mozambique elections: Renamo demands a recount’ (Southern Africa Report, Volume 15, No. 3, May 2000); João M. Cabrita, Mozambique: The tortuous road to democracy (Palgrave-St. Martin’s Press, Basingstoke, 2000); Eric Morier-Genoud, ‘Mozambique since 1989: Shaping democracy after socialism’, in Abdul Raufu Mustapha and Lindsay Whitfield (eds), Turning points in African democracy (James Currey, Oxford, 2009), pp.153–166. 46. Victor Igreja, ‘Amnesty law, political struggles for legitimacy and violence in Mozambique’, International Journal of Transitional Justice 9, 2 (2015), pp. 239–258. 47. See Carlos Nuno Castel-Branco, ‘Aid dependency and development: A question of ownership? A critical view’ (Working Paper No. 01/2008, Institute of Social and Economic Studies, 2008). 48. Buur and Monjane, ‘Elite capture and the development of natural resource linkages in Mozambique’. 49. Todd Chapman, ‘Former foreign minister on internal politics’, 27 February 2009, Wikileaks/Public Library of US Diplomacy, <https://www.wikileaks.org/plusd/cables/09MAPUTO320_a.html> (14 February 2017); Todd Chapman, ‘Mozambican businessman talks of corruption at the highest levels of government’, 28 January 2010, Wikileaks/Public Library of US Diplomacy, <https://wikileaks.org/plusd/cables/10MAPUTO86_a.html> (14 February 2017). 50. See Weimer, Macuane and Buur, ‘A economia do political settlement em Moçambique’. 51. Tony Vaux, Amandio Mavela, Joao Pereira, and Jennifer Stuttle, ‘Strategic conflict assessment Mozambique’ (Commissioned by DFID Mozambique, April 2006); Jason Sumich, ‘The party and the state: Frelimo and social stratification in post-socialist Mozambique’, Development and Change 41, 4 (2010), pp. 679–698. 52. República de Moçambique, ‘Estratégia nacional de desenvolvimento 2015–35’ (Governo de Mozambique, Maputo, Julho 2014). 53. Data from the National Statistics Institute. 54. Authors’ personal notes, 2010. 55. Eneas Comiche, ‘Prestação de contas e transparência na gestão de recursos naturais ponto de vista do parlamento Moçambicano’ (Paper Presented at Nordic-Mozambican conference on inclusive growth, Maputo, 2012). 56. República de Moçambique, ‘Proposta do orçamento do Estado para 2016’ (Governo de Moçambique, Maputo, 2015); Ministério de Planificação e Desenvolvimento and Ministério das Finanças, ‘Cenário fiscal de médio prazo 2007–2009’ (Governo de Moçambique, Maputo, 2006); Ministério de Planificação e Desenvolvimento and Ministério das Finanças, ‘Cenário fiscal de médio prazo 2014–2016’ (Governo de Moçambique, Maputo, 2013). 57. Deloitte, ‘The Deloitte guide to oil and gas in east Africa: Where potential lies’, 2013, <http://www2.deloitte.com/content/dam/Deloitte/global/Documents/Energy-and-Resources/dttl-er-deloitte-guide-oil-gas-east-africa%20–08082013.pdf> (23 June 2017). 58. See Center for Public Integrity, ‘Questions raised about the $400 million payment from ENI’, Newsletter, 18 August 2013, <http://www.cip.org.mz/cipdoc/261_CIP_PressRelease_August_02_en.pdf> (10 October 2013). 59. International Monetary Fund, ‘Republic of Mozambique: Selected issues’ (International Monetary Fund, Washington, 2016). 60. See Center for Public Integrity and Resources for Development Consulting, ‘Rovuma basin economic model’ (Power point presentation, Maputo, 2015). 61. Carlos N. Castel-Branco, Nelsa Massingue, and Carlos Muianga, ‘Padrões de investimento privado e tendências especulativas na economia moçambicana’ (Boletim Ideias 75, Instituto de Estudos Sociais e Económicos, Maputo, 23 Junho 2015). 62. Borges Nhamire, ‘Como as PPP são usadas para beneficiar as elites e prejudicar o Estado: O caso da concessão para fornecimento de chapas de matrículas envolvendo as famíias Guebuza e Machel’ (Investigation report, Centro de Integridade Pública, Maputo, 2015); Borges Nhamire and João Mosca, ‘Electricidade de Moçambique: Mau serviço, não transparente e politizada’ (Investigation report, Centro de Integridade Pública, Maputo, 2014). 63. Castel-Branco, Massingue and Muianga, ‘Padrões de investimento privado e tendências especulativas na economia moçambicana’. 64. Castel-Branco, ‘Estado e a capitalização do capitalismo doméstico em Moçambique’. 65. See Assembly of the Republic of Mozambique, ‘Parliamentary inquiry committee on the situation of the public debt’, Final report, 30 November 2016. 66. Ibid. 67. Ibid. 68. João Mosca and António Francisco, ‘Orçamento: Rectificativo ou ratificativo?’, Savana 1178 (5 August 2016), pp. 14–15. 69. Kroll Inc. was commissioned by the government with funding from Sweden. An edited version of the report executive summary was made public on 24 June 2017, by the Attorney General Office. 70. See Joseph Hanlon and Tim Jones, ‘Debt and development’, in Paul Haslam, Jessica Schafer and Pierre Beaudet (eds), Introduction to international development: Approaches, actors, issues, and practices (Oxford University Press, Ontario, 2017). 71. Prime Minister Carlos Agostinho do Rosário quoted in Joseph Hanlon, ‘Has government said it won’t repay secret debt’, Mozambique News Reports & Clippings, 17 April 2017, <http://www.open.ac.uk/technology/mozambique/sites/www.open.ac.uk.technology.mozambique/files/files/Mozambique_367-17Apr17_Did-govt-say-no-pay_BdM-Consumption-falling.pdf> (23 April 2017). 72. Frynas, Wood and Hinks, ‘The resource curse without natural resources’. © The Author 2017. Published by Oxford University Press on behalf of Royal African Society. All rights reserved This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

African AffairsOxford University Press

Published: Jul 31, 2017

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off