Introduction
This article investigates the historical logic of the mode of capital accumulation in Mozambique. By historical logic we mean the objective explanation of the specific characteristics based on an understanding of the country’s historical foundations, and the conflicts and tensions within and related to its structures of accumulation.
The historically specific characteristics of the mode of capital accumulation here are the specific forms, stages and methods that the development of capitalism in Mozambique followed, given its particular class-structured conditions. This investigation is focused on the capitalist restructuring of the economy that followed the adoption of the neoliberal structural adjustment and stabilisation programme, which spelled the end of a period of state-centred accumulation.1
Recent analyses of the Mozambican economy have often been articulated around linkages, which are variably understood as the forces of production and the potential connections between economic processes, like the replication of the pattern of specialisation that underlies the systemic crises of accumulation (Brum 1976; Cramer 1999; Biggs, Nasir and Fisman 1999; UNIDO 1999; GoM 2015), or around agencies, that is the social organisation of the forces of production or the social entities that do or do not bring about the linkages. Agencies include classes, the state, entrepreneurs, institutions, economic regimes, and how they succeed or fail to perceive and deliver optimal economic outcomes (Macuane, Buur and Monjane 2018; Guebuza 2004; Pereira Leite 1999, 1995, 1993). The successive crises of accumulation (Castel-Branco 2017; Wuyts 1989) are more commonly explained in terms of one or other of these two elements, often at the cost of the other and of analytical clarity.
These analyses tend to follow two main lines of enquiry. The first line focuses on elite behaviour and issues of transparency, neopatrimonialism, corruption and political will, usually emphasising the need for change in political behaviour and relationships of agents, institutional reform, privatisation and the elimination of cronyism (Macuane, Buur and Monjane 2018; Machel 2012, 2011; Nhachote 2010: Africa Intelligence 2021). The second line focuses instead on the world system that creates and replicates the patterns of specialisation and the dynamics of crisis in Mozambique and normally argues in favour of the transformation of the productive forces, often seen in isolation from broader social and historical conditions. Recommendations typically propose economic policy reforms through a technocratic approach that does not analyse the crises of accumulation and the dynamic social structures of accumulation associated with them (Brum 1976; Cramer 1999, 2001; GoM 1997, 2015; Weiss 1992). The problem with these two lines of enquiry is that they do not allow us to understand the objective logic of the mode of accumulation at play in Mozambique and have, therefore, limited analytical and heuristic power: they do not explain why things are as they are and therefore cannot use their analysis to identify and mobilise the energy for change.
This article takes a different perspective. Conditions of capital accumulation are historically determined and structured by class. To understand them, we need to analyse the dynamic, dialectic relationship between agents and linkages, here denominated linkagency, following Fine and Rustomjee (1996). These social structures and dynamics encompass the forces of production and their social organisation, and the power relationships that influence how policy responds to economic pressures and class relations.
The article explores the essence and historically specific traits of the capitalist experience of Mozambique and their logical location in specific historical conditions. This analysis does not imply that the recent trajectory of the Mozambican economy was and has been unavoidable, nor that it cannot be changed, but only that it can be logically derived from the options emerging from the existing historical conditions of accumulation. The understanding of this historical logic enables socially transformative action to be articulated and drawn from the concrete analysis of the mode of accumulation and its contradictions, rather than from a utopian and idealistic perspective of the state and of the society, or from a vision of socially and politically sterilised actions that are good for growth or for other economic indicators.
The arguments of the article are developed in three sections. The first summarises the method of political economy that is adopted here. This is followed by a section on the main characteristics of the mode of capital accumulation in Mozambique, which we describe as an extractive economy, and its internal tensions, conflicts and dynamics of expansion and crises. The third part explores the historically class-structured conditions under which the extractive economy developed, locating, within a broader context, the arguments in favour of state support for the emergence of a national bourgeoisie.
Linkagency: a method of political economy investigation of economic history
In the 1980s, Mozambique underwent a severe economic crisis, which followed the collapse of the state-centred project of development. The analysis of this period, often referred to as the ‘crisis of the 1980s’, poses some methodological and theoretical questions.
The ‘crisis of the 1980s’ was officially explained by four factors: war, natural disasters, the deterioration of international terms of trade of exports relative to imports, and errors in economic management, namely excessive centralisation and excessive state intervention in the market (GoM 1987, 1988, 1990; World Bank 1990, 1985).
The international financial institutions (IFIs) deployed these arguments to justify the introduction of a neoliberal agenda of stabilisation and structural adjustment. Within this simplistic analytical framework, if the ‘exogenous’ factors – such as natural disasters and terms of trade volatility – are eliminated because they are beyond the reach of public policy, we are left with two ‘endogenous’ factors: war and economic mismanagement. The ‘solutions’ to these two factors are respectively peace and the peace process, and economic restructuring aimed at dismantling the state-centred system of accumulation through privatisation, deregulation, the commodification of public services, market liberalisation, and fiscal and monetary austerity.2
This schematic and highly ideological analysis, which abstracts from the specific characteristics of the Mozambican economy, has been an effective tool in advancing the neoliberal agenda. Given that it cannot explain the dynamics of crises and how they are linked to economic expansion, it fails to shed light on the questions and challenges of transition.
An example of this is the World Bank study by Biggs, Nasir and Fisman (1999) on the structures and performance of the manufacturing sector in Mozambique 12 years after the sector was ‘restructured’ through privatisation and liberalisation. In one part of the report, it is argued that the specialisation of the Mozambican economy in primary exports shows its strengths and how markets have positively reacted to liberalisation, while in another part of the same report it is argued that this pattern shows the weaknesses of the economy that need changing, through state intervention. Another example is the Mozambican government’s industrial policy document (GoM 1997) advocating for the development of national small and medium private enterprises in import-substitution industries without any consideration for the social accumulation basis of the economy – the extractive, minerals–energy and agro-commodity export core of the Mozambican economy – or for the state of the privatised firms and the industrial strategy’s dependence on inflows of foreign capital (Castel-Branco 2017, 2020a). In a further example, the government in 2015 proposes an industrialisation strategy (GoM 2015) aimed at overcoming the structural limitations of the economy that is based on and articulated around expanding its extractive core and the linkages to it.
Heterodox approaches propose alternative explanations of the crisis, but, sometimes, these collapse into modified versions of dependency and path-dependent theories. Such approaches place the emphasis on the limitations of the Mozambican economy’s pattern of specialisation that make it dependent on exports of primary goods to rigid and volatile markets, in order to finance imports of investment goods, working capital and of consumer goods required for economic change (Brum 1976; Tibana 1991; Weiss 1992; Wuyts 1989). The argument is that extreme dependency on volatile export earnings from primary products constrains the ability of the economy to invest in structural change and grow faster. Any sudden and very significant acceleration of investment in structural change would inevitably build the momentum of a crisis of accumulation. This analysis can be extended to the crisis of 1991–93, when Mozambique was renegotiating a second adjustment and stabilisation programme with the World Bank and the International Monetary Fund (IMF). During this crisis, gross domestic product (GDP) contracted by 4% per year because of the interruption of aid inflows, following a period of significant economic recovery, in which GDP had grown by 5% per year, fuelled by large inflows of foreign aid, in 1987–90 (Castel-Branco 2002). Tibana (1991) shows a statistically significant correlation between export earnings from commodity exports and general economic performance in subsequent years.
For many of these scholars, the fundamental questions are the structure of linkages embedded in the historically built pattern of specialisation, the nature of the forces of production and how they are related to the world economy. According to these, the Mozambican economy can only be freed from its structural constraints to long-term growth through a transformation of the patterns of specialisation.
The issues raised by heterodox perspectives are real and highly relevant and, therefore, are part of the explanation of the crises of social and economic reproduction. However, there are equally important questions that they cannot address. First, this can become a circular argument because the success of industrialisation, which is the universally recommended solution, is bound to be conditioned by success in expanding commodity export earnings within the context of global volatile commodity markets, which would lock the country into exactly the same problem it was trying to escape from. Second, the argument fails to explain why and how this pattern of specialisation, which fails to provide stable reproduction and accumulation of capital, recurs in Mozambique, a country that underwent several radical political changes in the last 70 years. (These included the introduction of import-dependent, import-substition industries in the 1950s and 1960s through the multiplication of maquiladora industries – import substitution at the final stage of assembly; the crisis of colonialism and the introduction of a state-centred mode of accumulation in the late 1970s to mid 1980s; the neoliberal adjustment programme from the late 1980s; and the neoliberal economic nationalist movement from the the early twenty-first century.)
The heterodox critique of the patterns of specialisation was the main justification for the government’s adoption, in 1979, of the Prospective Indicative Plan (PPI) for 1980–90 to eradicate underdevelopment in one decade, which failed and led to the ‘crisis of the 1980s’. The PPI’s chances of success could have been improved with more balanced macroeconomic management, namely between consumption and investment/accumulation (Wuyts 1989; Doriye and Wuyts 1993), and a more carefully researched transition strategy particularly regarding agrarian and economic transformation (O’Laughlin 1981, 1996), but these ‘correctives’ are not independent of the broader social and historical context of the development of the productive forces.
In order to go beyond a simple demonstration of the correlation and causality between crises and the patterns of economic specialisation – or linkages – we need to locate those patterns and linkages within their class and other social and historical agencies. This can be operationalised through the linkagency approach (Fine 1992; Fine and Rustomjee 1996), which connects three components: linkages, which are the economic activities, processes or phenomena that, broadly speaking, lead to other economic activities, processes and phenomena; agencies, like social classes, the state and markets, which do or do not perceive, create, take advantage, articulate and bring about the linkages; and a social theory that unites them, which explains how linkages and agencies interact. Linkagency analysis connects the explanation of what happens – linkages – with that of who is responsible – agencies – and why and how that is done: the interaction between the two.
Fine and Rustomjee (1996) observe that many analyses of the South Korean industrialisation – often utilised to discuss South Africa – narrowly discuss either linkages or agencies, focusing either on developmental linkages – for example, import substitution or export promotion – or on the agency of the state that delivers them – for example, by way of industrial policy. Here the state is seen as an agency independent of – or at least able to free itself from – class interests, capable of choosing the ‘right’ combination of linkages, and this is advocated as a necessary condition for the adoption of successful development policies.
The linkagency approach, by contrast, argues that linkages and agencies are not independent of each other and of the historically class-structured conditions that create them. The agencies that bring about developmental linkages are structured as class and other social interests and relationships and are related to structures of accumulation, in the same way that Marx (1976) refers to the relationhsip between the forces of production and their social organisation. Within this approach, agencies are not free to roam unconstrained by economic conditions and structures (Kohli 1994; Fine 1992; Fine and Rustomjee 1996).
Following Fine (1992) and Fine and Rustomjee (1996), this paper applies the linkagency method to the analysis of Mozambique’s relevant economic history, exploring the objective and historically class-structured conditions under which the patterns of accumulation were and are formed. This method offers a radical alternative to the neoliberal analysis of the crisis that became the official narrative, while locating the structuralist, path-dependency analysis focused on the pattern of productive linkages in a class context. Rooted in a dialectic understanding of the specific historical conditions under which Mozambique’s capitalism develops, the present research looks at how agencies emerge from the forces, pressures, tensions and crises of the structures of production (linkages); how they react to and restructure them; and at why and how international and domestic capitalist forces and the state combine and, in so doing, structure the social dynamics of economic accumulation.
Key characteristics of the mode of capital accumulation in Mozambique
In this section, we argue that the Mozambican mode of reproduction and accumulation of capital is articulated around what we call the extractive economy (Castel-Branco 2010, 2014, 2017). Capitalism is about the reproduction of the social conditions for viable capital accumulation (Marx 1976 [1867]), hence our analytical focus on the mode of reproduction and accumulation of capital. The Mozambican capitalist classes have been undergoing extremely rapid transformation. Class formation has been occurring at an accelerated pace, generating dynamics of class struggle that are clearly conditioned by broader dynamics within capitalism at the regional and international levels.
Capital accumulation and the extractive economy
Capital accumulation is a social process structured by class, involving different, if related, processes. One is the expansion of wage labour, not least through land expropriation and the separation of the direct producer from the means of production, which creates social conditions that compel labour to sell their labour power. Another is the expansion of constant capital, like machines and infrastructures, the expansion and diversification of commodity production to sustain profit, and the expansion and intensification of the role of interest-bearing capital (Marx 1976 [1867], 1978 [1885], 1981 [1894]; Luxemburg 2003 [1913]). Studying the mode of capital accumulation in Mozambique entails individuating the particular social and historical dynamics that explain how profits are made, appropriated and deployed at the national level, and how this level is conditioned by and interacts with the world market. For this, we need a method of research, political economy, operationalised through the linkagency, that goes beyond the limitations of the traditional divide between macroeconomics and microeconomics and between sectors of the economy. The linkagency approach enables study of the dynamic structures of accumulation which are concurrently material (economic conditions, or linkages) and social (the social interests, groups, tensions and struggles that develop around the material).
In Mozambique, the historically specific mode of capital accumulation has led to the consolidation of an extractive economy. This mode of capital accumulation emerged from the combination of material conditions and agency conditions. Material conditions include the severe social and economic crisis of the 1980s that signalled the end of the state-centred system of accumulation, the patterns of specialisation inherited from colonialism, and the economic threat posed by the post-apartheid and post-sanctions expansion of South African monopoly capital. Withing the linkagency method, we can call these material conditions linkages. The agency conditions – both social and political – include the imposition by international financial institutions of a Washington-Consensus type of monetarist stabilisation and neoliberal structural adjustment,3 the mobilisation of national political elites to form the social basis of the new class of owners of capital, and the inability of the trade unions and other social movements to understand the historical significance of the regime change taking place at the time and to articulate labour strategic responses to it (Castel-Branco 1996, 2017, 2020b).
The extractive economy thus also entails a social process of accumulation focused on the creation of the national capitalist classes. Mozambique is a relatively latecomer to this process and aimed to develop national oligarchies that combine political and financial power, ownership of basic resources and access to rents from multinational capital (Castel-Branco 2014, 2017) while largely ignoring – or marginalising from policy focus – the social conditions of reproduction of labour (Ali 2017; Ali and Muianga 2016; O’Laughlin 1996, 2016; O’Laughlin and Ibraímo 2013: Castel-Branco 2021) and of reproduction of the process of capital accumulation in the medium and long term (Castel-Branco 2017, 2020a).
This process of accumulation is based upon systematic and recurring waves of ‘expropriation’ of the economic property and political power of the state as part of state philosophy and policies of promoting the emergent national capitalist bourgeoisie and absorbing the social and financial costs of private capital accumulation (Castel-Branco 2014, 2017; Guebuza 2004).
‘State expropriation’ is a concept playing with the general historical fact that private property and wage labour are created by expropriating direct producers of the ownership and control of land and/or the property of other means of production, and the specific historical fact that during the period of state-centred accumulation the Mozambican state owned the land and the main means of production and developed a policy framework focused on promoting its own accumulation. The creation of the Mozambican capitalist classes required the swift and cheap privatisation of state property, as well as the refocusing of state policy on promoting private accumulation. By giving away the control of land, natural mineral and energy resources and other fundamental means of production (such as ports, railways and industries), and by absorbing private debt and the costs of private accumulation (for example, through tax incentives, private debt guarantees and public–private partnerships) the state helped to ‘expropriate’ itself of basic, material resources and of the instruments and power of economic policy.
In the 1990s, the first wave of state expropriation led to the privatisation of about 1500 state-owned firms and to state-owned shares in as many private firms, including an implicit and informal subsidy to Mozambican buyers, who paid a mere 20% on average of the agreed asset prices (Cramer 2001; UTRE 1999, 1996; World Bank 1996). This wave created a class of Mozambican owners of assets which however lacked both the finance (because of monetarist policies focused on controlling money supply and inflation) and the industrial policies and institutions to help develop productive capacities. Thus, most of the asset owners failed to reproduce, and never moved beyond basic mercantilist capitalism. Throughout the 2000s and 2010s, the second wave of expropriation of the state denationalised, privatised and allowed the financialisation of real assets in strategic natural resources – land, forestry, coal, natural gas and oil, mineral sands and other mining reserves. This attracted multinational capital and provided significant rents and some productive linkages to the emerging domestic capitalist bourgeoisie (Bias 2010; Catembe.com 2015; Machel 2012, 2011; Nhachote 2010; Macuane, Buur and Monjane 2018; Castel-Branco 2014). The third wave of expropriation of the state consisted of denationalisation of strategic infrastructures (like railways and ports) and the overutilisation of the sovereign debt space, which had been created during two decades of austerity and GDP growth, through massive fiscal subsidies, public–private partnerships, private debt guarantees and the creation of new infrastructure for the core of the extractive economy.4 It also aimed at reducing costs and risks for large investors, making large private investment profitable and linking domestic and multinational private capital (Massarongo 2016; Massarongo and Muianga 2011; Castel-Branco 2014). These patterns of private capital accumulation caused a series of fiscal crises, which were dealt with through a sequence of austerity programmes – the fourth wave of state expropriation – that also opened up the space for the penetration of private financial capital in new areas, leading to the commodification of public goods and services, such as health, education, social security, national defence and security and public works (Castel-Branco 2020a, 2017, 2014).
Pattern of specialisation
As observed by heterodox economists, the Mozambican economy is stuck in a historical trajectory of dependence on exports of primary products. Its capacity to expand is dependent on primary exports and this dependency structures the social system of accumulation. The narrow specialisation of the economy generates continuous crises of accumulation because of four key elements: the volatility of international markets for primary commodity export; the secular tendency for the terms of trade of primary commodities to deteriorate in relation to those of imports of more specialised goods and services; the cyclical crises of profitability that result from tensions associated with overexploitation of armies of casual workers in employment systems that reproduce poverty, absenteeism and low productivity; and the difficulties posed by these patterns of production for the generation of developmental linkages and industrialisation (Brum 1976; Wield 1977a, 1977b; O’Laughlin 1981; Wuyts 1980, 1981; Tibana 1991; Castel-Branco 2010, 2017; Ali and Muianga 2016, 2017).
The two decades from 1995 to 2015 witnessed rapid economic growth, which reproduced and expanded the trajectory inherited from colonialism. This economic growth was based on the reprimarisation of production, consistent with the rapid expansion of foreign investment and of the overall size of the economy (Castel-Branco 2014, 2017; Langa 2017; Mandlate 2017; Maia 2019).
Over the last two decades, this general trend has deepened even further. For example, between 2004 and 2016 the manufacturing output, which had never been highly diversified and articulated (Brum 1976; Wield 1977a), narrowed down in two ways: the number of sectors and products decreased, and capital was concentrated on activities requiring less processing. Six sectors, representing 90% of industrial output, had levels of concentration of production of 70–90% of their output on four or fewer products, most of them requiring only basic processing (INE 1990–2019) (see Table 1). The only new statistically significant products introduced in industrial production in Mozambique since the mid 1960s are aluminium, coal, natural gas, minerals and tobacco, which represent approximately 80% of total industrial output in Mozambique (ibid.; Castel-Branco 2003, 2017).
Percentage of concentration of industrial production in four or fewer products per subsector, and the overall share of each group of subsectors in total industrial production, average for 2004–2016.
| Percentages of concentration of industrial production per subsector in four or fewer products | |||
|---|---|---|---|
| Between 70 and 79% | Between 80 and 89% | 90% or more | |
| Industrial subsectors and main products (four or fewer per subsector) | Food, beverages and tobacco (flour, sugar, beer and green tobacco leaves) Textiles, clothing and leather products (ginned cotton, cotton yarn and sacks) Non-metal minerals (cement) | Metal engineering (metal and mechanical consumer goods) Chemicals and oil products (natural gas) | Metallurgy (aluminium) |
| Output of the subsectors in the group as a proportion of total industrial output | 17% | 5% | 68% |
Private investment, both foreign and national, is extremely concentrated in the extractive economy. The Mozambican economy is structured around an extractive core: minerals–energy and agricultural primary commodities for export, and the services and infrastructures that are directly associated with it, such as roads, pipelines, logistic systems, construction, engineering, financial, accounting, transport and communications (Castel-Branco 2014, 2017). The extent of this concentration is shown in Table 2. Between 1996 and 2018, the extractive economy absorbed 95% of total private investment, 96% of all private inflows of foreign capital, 85% of national direct investment (NDI) and 88% of national commercial bank loans. This suggests that foreign direct investment (FDI) is the engine determining the size and allocation of foreign and national finance for investment.
Allocation of private investment between the extractive economy and other sectors, by source of financing, 1996–2018.
| Sources of private investment | Extractive economy | Other sectors and activities | ||||||
|---|---|---|---|---|---|---|---|---|
| Core of the extractive economy | Associated infra-structures and services | % of source | % of total private investment | % of source | % of total private investment | |||
| % of source | % of total private investment | % of source | % of total private investment | |||||
| Foreign direct investment (FDI) | 83 | 31 | 13 | 5 | 96 | 36 | 4 | 1 |
| Foreign commercial loans | 73 | 29 | 23 | 9 | 96 | 38 | 4 | 1 |
| National direct investment (NDI) | 58 | 3 | 27 | 2 | 85 | 5 | 15 | 1 |
| Domestic commercial loans | 66 | 12 | 22 | 4 | 88 | 16 | 12 | 2 |
| Total | - | 75% | - | 20% | - | 95% | - | 5% |
Between 2006 and 2017, GDP grew at an annual average of 7%. During that time, four sectors grew faster than GDP: the extractive industries, at an annual average growth rate of 23%; financial services and financial intermediation at 16%; and construction and vehicle trade and repairs at 10%. At the bottom of the table were agriculture at 3.2% and manufacturing at 3%. Within the agricultural sector, the annual rate of growth of food production for the domestic market stayed at a modest 1.2% (–0.9% per year below the rate of growth of the population), while the annual rate of growth of commodities for exports, wood, tobacco, bananas and sugar, varied between 6 and 7% (BoM 2020, 2003–2019; INE 1990–2019, 2020).
These data suggest that the trajectory of economic growth has been determined by the extractive economy. This growth finances consumption for higher-income groups, whereas the lack of capacity to expand per capita food production for the domestic market has a markedly negative impact on cost of living for the working classes.
The trends in the balance of payments reflect these patterns of production. Between 2005 and 2017, the extractive core of the economy provided 90% of exports, namely 72% from minerals and energy and 18% from semi-processed agro-commodities. As regards imports of goods, means of production (machinery and spares, construction materials, fuels, minerals and metal products) represented 66% of the total, while food products (16%) and motor vehicles (11%) were also very important. For the same period, business-related services (transport, construction, and business and consultancy services) represented 90% of imports of services (BoM 1995–2019; INE 1990–2019).
From analysis of the balance of trade, we conclude that commodity exports provide the capacity for financing economic expansion, which, in turn, is dependent on imports of strategic goods and services for the functioning of the economy and for feeding the people. This suggests that the pattern of specialisation of production and the dependency on imports of productive capacity and food have been reproduced, which minimises the possibility of sustained fast economic expansion.
Commodity volatility on the world market causes economic booms and busts in Mozambique. Any long-term planning of investment has been dependent on the capacity to sustain external debt, attract massive aid or generate large fiscal surpluses during boom periods – but these surpluses have never occurred (Brum 1976; Castel-Branco 2020a, 2017; Wuyts 1989). The booms and busts have a destabilising effect on the exchange rate which in turn affects the price of food, which is largely imported. Exchange rate policies have tended to focus on stability, supported by balance of payment financing by the IMF or by eroding foreign reserves, inevitably leading to overvaluation and overshooting in the adjustment periods. The obsession with stability is derived from the political need to minimise the inflationary impact of the booms and busts on food prices, which means that we are in the presence of an ‘anti-riot exchange rate policy’ (Massarongo 2017, 2013; Castel-Branco 2017, 2020a).5
Dependency on international private finance
The rapid expansion of the economy, based on its extractive core, depends on inflows of large-scale international private finance. The dependency on foreign inflows of private capital is multidimensional, as foreign capital determines the quantity and the direction and allocation of investment, as well as the focus of policy and the control of strategic resources. We can assess the power of foreign private capital by looking at some data, summarised in Table 3. Not only was 76% of all private investment in Mozambique financed by external inflows of capital, but direct investment by domestic firms (NDI) only financed 6% of total private investment.
The combination of Tables 2 and 3 suggests that FDI determines the economic structures and linkages. Intentionally or not, FDI defines the direction of investment, leading multinational companies to wield a disproportionate power to influence state priorities and policies. This has consequences for the dynamics of class formation, competition and struggle.
Within the Mozambican private financial system, total deposits in savings accounts are less than the average annual inflow of FDI. In 2019, total deposits in savings accounts averaged US$4 billion, while the average annual inflows of FDI approached US$5 billion (BoM 2020, 1995–2019; World Bank 2020a, 2020b; CPI n.d.; Massingue and Muianga 2013). This strategy of accumulation causes dependency on large inflows of private foreign capital, entrenching financial dependency within the national development strategy.
Financial dependency creates several critical structural problems. On the one hand, the economy must attract international private capital, and the state makes sure that this happens in four interlinked ways: by making strategic natural resources available at low cost; by ensuring that capital controls the management of relevant public infrastructure (like railways and ports) through different types of concessions; by guaranteeing tax incentives and facilities for repatriation of capital; and by increasing public indebtedness for infrastructure investment and for guaranteeing private debt. The magnitude of tax incentives prevents the formation of large fiscal surpluses in boom periods, which reduces the policy tools to cope with export market volatility during periods of economic expansion and to sustain sovereign debt (Castel-Branco 2010, 2014, 2017). On the other hand, the scale of the concessions allows multinational companies to transform part of the real assets – mining, natural gas or land – into financial assets such as shares, holdings, debt securities and others that are then sold and resold in stock markets with the sole purpose of generating financial capital directly out of transactions of financial assets, without concern for or involvement in production (Marois 2012; Fine 2010, 2012).
Financial assets are resold on world stock exchanges, guaranteeing accelerated recovery of cost and the generation of financial rents independently of the productive deployment of the real assets. In other words, national strategic resources are financialised for corporate profit (ibid.; Catembe.com 2015).
Alongside this process of financialisation, the economy’s weak industrial base makes it hard to develop productive linkages with large private investment. Domestic capital is left with few options: it can invest in secondary activities associated with FDI, such as transport services, construction, maintenance and cleaning and residual finance; it can participate in the shareholder structure of investment projects, often through links with the state; or alternatively it can engage in speculation with concessions mostly in agricultural and real estate land or mining (ibid.; Bias 2010; Langa 2017, 2015; Langa and Mandlate 2013; Massingue and Muianga 2013; Melina and Xiong 2013).
A classic example of the control exerted by foreign capital over policy and strategic resources is offered by the contract over the port and logistics terminals for the extraction of natural gas in Pemba and Palma, Cabo Delgado. The contract was allocated by the government of Mozambique to a private company, Portos de Cabo Delgado (PCD), headed by a senior figure from the national political and financial establishment. The government justified the absence of public tender by stressing the urgency of expediting the process in order to establish the logistics infrastructure for natural gas provision. PCD was then outsourced to the financial group Orlean Invest, headed by the Italian Gabriel Volti and the Nigerian Atiku Abubakar, to implement the project associated with the concession. The contract confers monopoly power on the concessionaire and, in this way, on the sub-concessionaire with respect to the Pemba and Palma terminals. It also prevents the government from exercising or granting rights to provide logistical and port support for oil and gas exploration up to 300 miles south of Pemba, which includes the port of Nacala. The fact that the outsourcing contract forms part of the original concession contract between PCD and the government suggests that PCD was a rentier intermediary from the start. Besides, the sub-concessionaire is a financial group and not a corporation with experience and reputation in ports and oil and gas logistics, pointing to the fact that the core business here is not logistics but finance. As has come to be expected with such financialised deals, no new investment or construction work has been funded after three years by private funds, and the project, if it ever moves forward, will require public guarantees of private debt or direct funding by the public treasury to cover immediate financial needs, which are estimated at US$200 million (Mosse 2016).
Therefore, the expansion of this economic model exacerbates rather than alleviates the tensions and structural limits of the economy, while fostering a process of domestic bourgeois class formation based on rent, supported by international capital interests at the material and ideological level.
International finance and public debt
The rapid expansion of inflows of private international capital is associated with the accelerated expansion of public debt and the financialisation of the state, as public policymaking is subordinated to the logic and interests of financial markets, in a context where finance dominates the operation of the economy (Fine and Pollen 2018; Mikuš 2019). Private finance penetrates and comes to dominate service provision in basic needs services that were traditionally managed by the state: education, health, social security, national defence and security, roads, collective transport and the provision of water and electricity.
The close relationship between foreign investment and public debt results from the combination of several factors, one of these being high capital costs of projects focused on intensive and expensive infrastructure and other capital investment, inducing borrowing from capital markets. Table 4 shows that during the period of rapid expansion of inflows of foreign capital, the ratio of private debt to FDI of private firms increased significantly.
Evolution of inflows of FDI (US\(millions) and ratio of private debt to FDI (as a percentage of FDI).
| 2000–2004 | 2004–2009 | 2009–2011 | 2011–2012 | 2012–2013 | 2013–2014 | 2014–2015 | 2015–2016 | 2016–2017 | 2017–2018 | 2018–2019 | 2011–2016 | 2000–2019 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Total FDI (US\) millions) | 1324 | 2188 | 2189 | 3664 | 5635 | 6697 | 4999 | 3868 | 3128 | 2319 | 2678 | 24,863 | 36,671 |
| Annual average FDI (US\(millions) | 265 | 438 | 1095 | 3664 | 5635 | 6697 | 4999 | 3868 | 3128 | 2319 | 2678 | 4973 | 2037 |
| Private debt as proportion of FDI (%) | 0 | 60 | 70 | 62 | 96 | 86 | 89 | 80 | 74 | 72 | 82 | 85 | 82 |
Another factor is the long time lag between investment and project start-up, such that high levels of private debt and significant financial costs accrue for investors at the early stages of the projects.
Third, government policy prioritises reduction cum socialisation of private risks and costs to make investment in the extractive core viable and attractive for investors. The state gives away strategic natural resources at low cost to attract multinational corporations such as Vale do Rio Doce, Riversdale and Rio Tinto in coal; Sasol, Anadarko, Eni, ExxonMobil and Total in natural gas; Kenmare in heavy sand; Gemfields in rubies; and BHP/South32 and Mitsubishi in aluminium. To make these projects financially viable for shareholders and to guarantee the participation of emerging national capitalists in the capture of rents, the state finances its FDI attraction strategy through public debt: the government offers tax incentives, making tax revenues from exports of minerals, energy and agro-forestry either volatile or non-existent, promotes public–private partnerships and public investment in key infrastructral projects and ensures public guarantees for private debt. Inflows of foreign private capital, which make up 76% of private investment, are also responsible for 70% of public commercial sovereign debt, as direct public investment in infrastructure represents 31% of commercial sovereign debt and private debt guarantees and public–private partnerships represent 39% (ibid.; Castel-Branco 2020a, 2017; GoM 1995–2019).
The conversion of private into sovereign debt has four major advantages for multinationals and for financial markets and private holders of securities: it socialises risks and costs; it guarantees the returns; it creates new speculative options, as speculating against the risk of default becomes part of financial dealings (Castel-Branco 2020a); and, ultimately, debt can be swapped for real assets, such as land or mineral and energy resources.
The state offers exceptionally large mining and energy concessions to multinational companies, partly because of its limited negotiating capacity and lack of experience, but also to achieve two other objectives: to show its political commitment to supporting large multinational capital, and to leverage the entry of emerging national capitalists in these large businesses.
This creates three additional problems that were probably not anticipated in the government’s initial calculations. First, multinational companies establish monopoly control over strategic resources, which gives them power to influence national and local authorities, control territories and their peoples, decide what to do with infrastructure, and determine development options and the public strategy and policy, which have been utilised in favour of shareholders and debt bond holders (Castel-Branco 2014, 2017, 2020a; Macuane, Buur and Monjane 2018; Massarongo 2016, 2017). Second, the state has had to buy back its minority shareholding in the projects based on resources which not long ago were state-owned. To finance these operations, the state contracts foreign debt by selling Eurobonds at a high premium, favourable to international financial capital that speculates with expectations of expansion and collapse, but which represents a continuous drain on the state’s financial capacity (Castel-Branco 2020a). Third, the mega-scale of concessions has made it possible to financialise the real assets of Mozambique’s subsoil. Financialisation allowed investors to significantly shorten the time required to recover their investment, generate profits and reduce risks (ibid.).
In 2015, the Mozambican Tax Authority, ATM, announced that there were more than two dozen transactions between multinationals involving national mining and energy assets (Catembe.com 2015), usually justified by the need for recapitalisation of the companies and new partners when they move from prospecting to extraction. The total estimated value of these transactions exceeds the total value of private investment projects approved since 1990 (US\)35 billion) which, if duly taxed, would give the state sufficient revenue to cover two budget years (ibid.).
The coal industry offers a good example of the process of financialisation of natural resources in Mozambique. Riversdale was an Australian company, listed on the S&P/ASX 200 index, specialising in coal mining in Africa, with three major shareholders: India’s Tata Steel, Brazil’s steel company CSN, and Passport Capital (now defunct). In 2010, the Anglo-Australian company Rio Tinto, which was the second largest corporation in the world’s mining and metals industry, only behind BHP, acquired a majority stake in Riversdale.6 In the same year, Riversdale signed an agreement with the Chinese company Wuhan Iron and Steel to jointly develop the coal concession in Benga, near Moatize, Tete province. A year later, and after no significant additional investment, Riversdale sold 51% of its shares in the Benga concession to Rio Tinto for around US$4 billion, the equivalent of four times the total value of its share price on the Australian stock exchange prior to the acquisition of the coal concession in Tete. Riversdale’s stock revalued by a factor of eight in one year, due solely to the financialisation of real assets and speculation with those assets on the Australian stock exchange (VCCircle 2011). In the meantime, Wuhan Iron and Steel sold its stock in Riversdale to Rio Tinto for US$1.2 billion, which is twice as much as they have invested as shareholders of the project (ibid.). Ten years later, production on this coal site has yet to start.
Financialisation of real assets was also directly supported by sovereign debt. In the case of electricity, the Mozambican government purchased the Portuguese shares in Hidroeléctrica de Cahora Bassa (HCB) in Tete for €900 million through a loan from the European Investment Bank. The subsequent debt was managed by Mozambique’s second largest private bank, Banco Comercial e de Investimentos, SA (BCI). Similarly, the construction of the Katembe bridge and a 100-kilometre road to Ponta do Ouro were financed through a US$682-million loan from the Exim Bank to be paid in 20 years at 4% interest (Castel-Branco 2014; Macuane, Buur and Monjane 2018; Catembe.com 2015; Dreher et al. 2017).
Figure 1 shows that from 2011 to 2016 the stock of sovereign debt grew significantly faster than economic output, jumping from 40% to 120% of GDP. This acceleration is linked to the acceleration of private investment, because of mechanisms of public support to private investment. The structure of sovereign debt changed from predominantly concessional to largely commercial and therefore became more expensive, more short-term and more difficult to renegotiate.
Sovereign debt and the financial system
The rapid increase of sovereign debt has made the domestic financial system more speculative and thus inadequate for supporting the expansion, diversification and articulation of national productive capacities. In view of the fast-growing pressure posed by the debt service on current public expenditure (Figure 2), the state borrowed from the domestic financial sector (Figure 3), with which it traded short-, medium-, and long-term public debt bonds. Figure 2 also illustrates the close relationship of the trends in external and internal public debt, whereas Figure 3 shows the government gradually becoming the largest recipient of credit from domestic banks.
However, government borrowing is crowding out not private investment in general, but only domestic small and medium investment, because domestic public debt is helping the government to pay for infrastructure, private debt guarantees, public–private partnerships and other expenditures that support large multinational, extractive projects and financialisation of real assets by large corporations. For this reason, government debt operates like a mechanism of transfer of surpluses, present and future, from the whole economy to its extractive core and to the capitalist class with rentier features that thrives on it.
In the period of rapid expansion of private investment cum public debt (2011–17), the financial sector adjusted and took advantage of the economy’s dynamics for its own benefit. The spread between interest rates and inflation was kept high – always above 8% – with an annual average of 15%, and was insensitive to monetary policy when reference rates fell (see Figure 4). This means that the financial system crowded out small and medium businesses that operate outside of the extractive core of the economy.
At the same time, while credit to the economy continued to grow, the share of credit allocated to productive sectors and trade fell by 14% and 21% respectively, whereas the share of bank loans to consumption of durable goods (housing and motor vehicles) and purchase of government debt bonds increased by 24% and 19% respectively (BoM 1995–2019, 2020). Speculation over public debt and consumption by the wealthiest groups in society have become much more important to the business and profitability of banks than any productive investment. Through its impact on the financial system, the current mode of accumulation prevents the development of alternatives to the extractive economy.
Social inequality and poverty
The extractive economy is focused on the creation of national capitalist classes and, consequently, is promoting the financialisation of the state and of real assets. Workers’ social conditions and social reproduction are largely ignored or marginalised in policy focus. Austerity programmes are systematically utilised to curb the frequent crises of accumulation and reproduction that ensue from this social and economic system, thus protecting the profitability of financial assets and markets, restricting the ability of the state to pursue broader economic and social programmes, and promoting the commodification and the penetration of financial capital in key sectors such as health, education, social security and public works. The expansion of this economy increases the scale of these structural problems, as growth and crisis are organically and dialectically related.
The extractive economy has been more efficient at creating millionaires than at reducing poverty. Between 2002 and 2016, inequality in distribution of nominal income increased, with the Gini coefficient jumping from 0.47 to 0.54, and the income share of the poorest 60% declining from 28% to 23%.7 For the wealthiest 10%, the income share increased from 40% to 50% (Trading Economics 2020). Between 2010 and 2015, the number of Mozambican millionaires increased by an average of 5% per year, the number of multimillionaires increased by a factor of 5 and, for the first time, a Mozambican resident was classified as a ‘centimillionaire’ (New World Wealth 2019).8 On the other hand, between 2000 and 2019, the percentage of people in Mozambique living below the poverty line declined by only 0.7% per annum on average, while the absolute number of poor increased by almost 2.5 million, in an economy growing, on average, at 6.5% per year (World Bank 2017, 2020a, 2020b; INE 2016, 2017; MEF 2016). Mozambique’s elasticity of poverty reduction with respect to GDP growth is only –0.11, which means that Mozambique’s GDP needs to grow 9% for poverty to reduce by 1%. This is one-third of the average elasticity for Sub-Saharan Africa and one of the lowest rates in the region (Sembene 2015).
Mozambique’s millionaires are the product of the speculative and extractive mode of accumulation that we have analysed. Their newly acquired wealth is deployed on and around the capital-intensive extractive core of the economy, on financial and real estate speculation and on luxury consumption (Machel 2011, 2012; Macuane, Buur and Monjane 2018; Nhachote 2010; Castel-Branco 2017). The persistence of poverty is generated from the increasing casualisation and informalisation of work, land expropriation and massive displacement of people, destruction of livelihoods in the areas of expansion of mining and fossil fuel extraction, limited employment opportunities in the main areas of economic growth, and accelerated depreciation of the purchasing power of poorer groups because of the inability of the economy to produce and provide cheap wage goods and services (Ali 2017; Ali and Muianga 2016, 2017; Ali and Stevano 2019; Muianga 2017; Wuyts 2011; O’Laughlin and Ibraímo 2013).
Crises and the historical class-structured logic of the mode of accumulation in Mozambique
The previous section described the key characteristics of the mode of capital accumulation, how objective economic conditions (linkages) and corporations and social classes, decisions and policies (agencies) relate, and the mechanisms of the dialectic relationship between expansion and crises. Where does this mode of accumulation come from, why, and how is this logically linked with class structured historical conditions? More importantly, why has it persisted and reproduced itself over time? This section addresses these questions. We argue that the final crisis of the state-centred mode of accumulation, in the 1980s, opened the door for the argument in favour of national capitalism to dominate the political debate and practice, and that the focus on the development of national capitalist oligarchies conspired with the specific historical conditions to produce an extractive and porous mode of capital accumulation (Castel-Branco 2014).
The question has been asked as to why the response to the crisis was the creation and empowerment of the emerging national bourgeoisie and not a different version of a socialist alternative. While this cannot be exhaustively addressed here, we would suggest that a bureaucratic and authoritarian state-centred mode of accumulation together with a dualistic perspective on the agrarian question (O’Laughlin 1981, 1996; Castel-Branco 1996) were instrumental in preventing progressive political, economic and social alliances from developing as powerful alternatives.
Crises and collapse of the state-centred mode of accumulation
In the second half of the twentieth century, Mozambique’s economy was affected by at least five periods of economic crisis, defined by the occurrence of negative rates of expansion of GDP. Some were more severe and long-lasting than others, namely in 1963–64, 1970–72, 1974–77, 1982–87 and 1991–93 (Castel-Branco 2017). During every crisis, negative growth rates followed periods of rapid expansion of GDP and investment, and a recurring pattern was created where a period of rapid economic growth lasting two to five years was followed by another period of negative growth lasting two to three years, except for the crisis of the 1980s, in which negative growth lasted for five years. Each time, growth was not sustained because of the exhaustion of the economic capacity to intensify or maintain the operation of the existing productive capacity and to expand it, which was generally caused by fiscal and current account deficits that accumulated during the expansion period (ibid.).
This raises three questions: (i) why did such deficits arise? (ii) why did the deficits become so unsustainable that the economy was forced to contract? and (iii) why was this pattern reproduced for five decades, despite radical changes in ideology, institutions, politics and policy, broadly understood as agency? The remainder of our analysis of the crises will focus on the post-independence experience and on the debate about the reasons for the collapse of the state-centred mode of accumulation.
The Mozambican government’s response to the crises and collapse of the colonial system (1970–72 and 1974–77), based on an analysis of the need to change investment, production and trade structures, was its PPI of 1980–1990 (mentioned above), an economic programme to eliminate underdevelopment and lay the foundations for socialism, defined as state-centred accumulation. With an extremely ambitious 10-year investment plan targeting an average annual rate of growth of 17%, the PPI was based on three pillars: the socialisation of the countryside, through the rapid development of productive forces organised in state-owned agricultural enterprises or cooperatives; industrialisation, related to agro-industrial transformation and mineral and energy projects, mostly located in the Zambezi valley; and a national programme of education and vocational training to meet the needs of the two previous pillars.
Three slogans dominated the economic ideology at the time: one was that accumulation came first and social transformation next, which led to the reproduction of the economic structures and social relations of production inherited from colonialism; the second was that state enterprises were the most advanced form of socialist property, which in practice marginalised socialist alternatives, like cooperatives, associations and others; and the last was the dualistic perspective on the agrarian question, reflected in the absolute priority given to the establishment of large state-owned farms reproducing monoculture and massive seasonal labour systems, and in the idea that peasants could wait because they could take refuge in pre-capitalist, non-market economies. This led to poverty and famine in the rural and urban areas, and social disengagement of the peasantry from the revolution (CEA 1977, 1979; First 1983; O’Laughlin 1981, 1996; Bowen 2000; Castel-Branco 1996).
The state was the central agency for implementation of the PPI, through a central planning system, the use of public budget and a public financial system, the creation of technical institutes and state enterprises. It was expected that the PPI would be, largely, financed through international cooperation, mainly by the Council for Mutual Economic Assistance (COMECON), and by the subsequent generation and accumulation of surpluses of state-owned enterprises, of state fiscal revenues and of foreign currency earnings (Wuyts 1989; Castel-Branco 1996).9
However, by the early 1980s it had already become clear that the PPI-driven ‘big push’ had failed. Table 5 shows the percentage rates of change of some key economic indicators, comparing the first half of the 1980s with the year before the start of the PPI (1979). The figures show that the ‘big push’ was maintained for only two years and, by 1984, the performance of the economy was significantly worse than in 1979. Investment grew meteorically, as one would expect from a programme designed to quintuple the size of the economy in 10 years, but only for two years. During acceleration, exports – primary products – grew slowly (7%), but imports, dominated by means of production and food, increased six times faster (44%), such that by 1984 exports could only cover 18% of imports. This suggests that a current account crisis may have been a leading direct cause of the collapse of the PPI. Like exports, GDP grew slowly during acceleration, but collapsed rapidly afterwards. The investment-to-GDP ratio increased dramatically in two years but, as surpluses did not materialise fast enough, the economy ran out of fuel, investment collapsed and so did the entire structure of the PPI.
Real rates of growth (at constant prices) of some key economic indicators between 1979 and 1984, in percentage (cumulative rates for each period).
| 1979–1982 % | 1982–1983 % | 1983–1984 % | 1979–1984 % | |
|---|---|---|---|---|
| Gross domestic product (GDP) | 9 | –12 | –9 | –13 |
| Gross investment | 270 | –64 | –25 | 0 |
| Exports of goods and services | 7 | –27 | –32 | –46 |
| Imports of goods and services | 47 | –25 | –30 | –23 |
| Export to import ratioa | –44 | –20 | –10 | –60 |
The economic crisis and its neoliberal analysis, discussed earlier in this article, combined and conspired to generate a crisis of political identity, which marked the end of the collective, state-centred project of development and opened the doors for a nationalist, capitalist project to emerge. After that, all that was needed was a theory that argued that the development of a national, capitalist bourgeoise was indeed an imperative for national liberation.
Mozambique’s capitalist manifesto
In 1987, in a busy public lecture in one of Maputo’s hotels, Armando Guebuza, at that time a member of the politburo of Frelimo’s Central Committee and minister of transport and communications, presented a document that can be regarded as his capitalist manifesto for Mozambique.10 Guebuza argued that to the party and the state pertained the function of protecting Mozambique’s political and economic sovereignty, and by extension the strategic resources of Mozambique, against imperialist globalisation. His theory predicted that this task would demand the development of a nationalist, domestic capitalist class capable of representing national interests and dictating the course of national development. Human capacities and financial capital were needed to this end. His proposal was that human capacities would come from the party and the state – where the experience of managing the country and nationalist loyalty had been accumulated – as well as from the younger generation of nationalist families with a strong track record of patriotism. Capital would have to come from foreign sources because of the financial weaknesses of the emerging national bourgeoisie, the constraints imposed by fiscal and monetary austerity and the deep economic crisis of the country. Additionally, it would be necessary to find a mechanism by which the emerging national capitalists could benefit, directly, from the inflows of foreign capital.
Guebuza also warned about the risks posed by the privatisation of state-owned or state-sponsored enterprises, which could lead to an extensive transfer of assets to foreign capitalists or to Mozambicans from minority ethnic groups. This argument reflected real class-structured aspirations, which emerged under specific historical conditions of economic and political crisis, and from transition from state-centred to private accumulation.
During the national liberation war (1964–74) Frelimo had defined the ‘liberation of the land and of men and women’ as the main goal of the national democratic revolution, reflecting the need for liberation from colonial occupation – liberation of the land – and from capitalist relations of production – liberation of men and women (Machel 1973, 1974, 1977). Under the new regime, these slogans acquired a radically different meaning: the liberation of men and women meant freeing themselves from the barriers that colonialism and imperialism had created to the development of national capitalism, and the liberation of land became associated with the political imperative and power of the new national bourgeoisie to expropriate land from the state and peasants and make it available for the process of domestic capitalist accumulation. Men and women were ‘free’ to become capitalist or self-employed, or to sell their labour power. Therefore, the alternative to imperialism was no longer socialism, but national capitalism (Guebuza 2004).
Historical conditions of development of national capitalism in Mozambique
Mozambique’s capitalist manifesto was rooted in several historical references, which are fundamental to understanding the current dynamics of accumulation. First, the integration of Mozambique into global capitalism had been carried out during colonialism, which led to the mass expropriation of land and other resources. It formed the rural semi-proletariat dependent on wage labour markets but tied to the land as a way of guaranteeing their availability at low cost and of passing part of their social reproduction costs to the unpaid, uncommodified labour of their families. It integrated the economy into the world capitalist system as an exporter of primary goods and importer of capital and consumer goods, and, generally, subordinated economic and social activity to the dictates and logic of capitalist accumulation (CEA 1977, 1979; First 1983; O’Laughlin 1981, 1996; Wuyts 1980, 1981). The reliance upon the extraction of absolute surplus value through coercive means such as forced labour, labour market monopsony, compulsory crops at controlled prices and the role of household unpaid production for subsistence that guaranteed cheap and available labour power and the dependence upon volatile commodity markets shaped the pace and direction of the development of productive forces and of the corresponding social relations of production, constraining the options for future development (ibid.).
The formation of Mozambican capitalist classes occurred within these structures of capital accumulation inherited from colonialism. (ibid.). Despite radical and fast changes in property rights and land tenure, from large state farms and cooperatives in the first decade of independence to different forms, shapes and sizes of private property afterwards, the basic structures of agrarian capital accumulation and its role in overall accumulation of capital did not change (Ali and Muianga 2016, 2017; Castel-Branco 1996; O’Laughlin 1981, 1996, 2016; O’Laughlin and Ibraímo 2013; Wuyts 1981, 1989).
Historically new are the scale and speed at which Mozambican capitalist classes are developing, their alliance and subordination to large-scale multinational capital, mainly investing in minerals and energy sources, and their expropriation of the state through large-scale transfers of property and rents and through financialisation of the state. The structural and dialectical dynamics of expansion and crises of the colonial economy, which were an essential part of the collapse of the PPI, are reproduced at a much larger scale (Castel-Branco 2014, 2017).
Mozambique has reshaped the forms rather than changed the structures of capital accumulation in which the historically specific struggle to build a national, black bourgeoisie played a crucial role. The transition from a state-centred to a private-centred mode of accumulation required massive privatisation of state assets. Privatisation was seen as a panacea to attract private investment, increase efficiency and productivity, reduce fiscal imbalances and create a national capitalist class (Cramer 2001; Castel-Branco 1996, 2002, 2003).
However, the programme of privatisation faced several key challenges. Most of the productive assets were run down or technologically obsolete, requiring considerable amounts of finance for rehabilitation and modernisation. The economic crisis combined with monetary and fiscal austerity to make the availability of finance contingent to inflows of foreign aid and/or foreign private capital. Since the beginning, and before they could become competitive, emerging private firms were exposed to strong international competition due to the liberalisation of markets (labour, capital goods and consumer goods). At the same time, the withdrawal of the state from active industrial strategies left domestic industries and firms to fend for themselves, which did not help those requiring considerable restructuring There was a serious risk that emerging national, black elites could be crowded out as a result of access to state assets being privatised (Cramer 2001; Castel-Branco 2002).
The state, which had neither the political space nor the institutional capacity to address the problems with finance, liberalisation and industrial policy, because of the structural adjustment and stabilisation programme, faced no powerful pressure for progressive alternatives from the broad working classes and opted to focus on the single goal that could be actively pursued through policy: the formation of an emerging class of national owners of productive assets. The post-independence Mozambican government, similarly to earlier colonial authorities, had limited and structured the development of national capitalist classes, making their development dependent upon the state.
The emergence of national capitalist classes was promoted through three policies. A pool of public assets was created for exclusive access of veterans of the liberation war, and association with veterans became one of the criteria in the evaluation of tenders for the purchase of public assets. Mozambican nationals were allowed to pay a mere 10% of the selling price to acquire the asset, while the remaining 90% could be paid as their enterprises developed. Evidence shows that, on average, 80% of all assets were sold to nationals and only 20% of the total sale value of these assets was paid to the state. This amounted to a large implicit public subsidy, without industrial policy, to develop a national class of owners of mostly rundown or obsolete means of production (UTRE 1996, 1999; Cramer 2001). This is what we have called the ‘first wave of state expropriation’ (Castel-Branco 2017, 2014). By 1995, approximately 60% of these privatised firms had either gone bankrupt, been resold to foreign investors, or transformed from factories into warehouses, and the state had not realised the expected fiscal returns of privatisation (UTRE 1999; Cramer 2001; GoM 1992, 2000a, 2000b).
Privatisation was also an uneven process, which favoured industries in oligopolistic markets financed by foreign capital over the emergence of small and medium industries. For example, privatisation revived the cashew-processing industry, characterised by a high level of fragmentation along the production chain and financed basically from customer credit, local banking or domestic corporate profits. The whole industry was then liquidated by the liberalisation of raw cashew nut exports. The sugar industry, on the other hand, was characterised by high levels of homogeneity and vertical integration along the production chain, as plantations and sugar processing facilities were sold to sugar multinationals (Illovo, Tongaat Hulett and a consortium of different capitals). International commercial banks funded its rehabilitation and the industry was able not only to resist liberalisation pressures but also to build a series of market distortions that supported the sector’s development. These included a flexible pricing policy to protect it from imports at dumping prices and world market volatility, the coordination of competitive investment and introduction of barriers to the entry of new companies, to ensure economies of scale, and the negotiation of prices, production levels and market shares to minimise excess output that would have to be sold outside preferential quotas at dumping prices. Sugar companies established a producers’ association that acted as an oligopoly to negotiate policy, prices, market quotas and labour conditions with the government and other interested parties (international financial organisations, other industries etc.), while promoting the development of medium and small cane growers, who helped to reduce adjustment costs in the face of world market volatility and break the organisation and bargaining power of sugar cane plantation workers. Big business, involving commercial banks and multinationals, therefore resisted liberalisation and managed to subordinate public policy to its corporate interests, showing that the power of capital over the market and the state is vital for its expansion, consolidation and profitability, even at the expense of the whole economy (Castel-Branco 2002; Cramer 1999, 2001; Fine 1997; Fine and Poletti 1992; Fine and Rustomjee 1996).
A third point is that because social differentiation, class formation and tensions – including the formation of capitalist and working classes – had developed along racial lines under colonialism, then from the point of view of economic nationalism national liberation would not be complete until a strong, black capitalist domestic bourgeoisie developed (Castel-Branco 2007; Tutashinda 1978; O’Laughlin 1996; Shivji 2009). For this reason, the promotion of capitalism in Mozambique adopted an ethnic-national dimension and rationale, which had an objective base in history and would later become useful to validate the capitalist project historically and politically and to hide its class dimension.
Fourth, the process of private capital accumulation was focused solely on the formation of the national capitalist bourgeoisie, disregarding the conditions of social reproduction of labour, which constrains the continuous expansion of capitalism since it expropriates without creating employment and effective demand, generates misery and new social tensions, makes domestic markets less potent and reduces the sources of impetus for economic development to the dynamics and interests of multinational capital and foreign markets (Marx 1978 [1885]; Luxemburg 2003 [1913]; Bukharin 2003 [1915]; Lenin 2010 [1917]; Mandel 1980).
Fifth, the new capitalist bourgeoisie emerged as the self-appointed representative of the nation’s highest aspirations and of the African renaissance, being intentionally separated from, though historically dependent upon and associated with, global capitalism. The continuation into the future of the broader aspirations of national liberation – economic liberation – became ideologically and politically articulated in terms of the role of the national bourgeoisie, with the government of the national state becoming a committee for empowerment and management of its interests.
Sixth, the end of apartheid in South Africa in the early 1990s and the subsequent lifting of economic sanctions brought about another set of challenges and opportunities for the development of Mozambican capitalist classes. The challenges were threefold: trade competition from a more competitive and diversified industrial base, the restructuring of regional economies by means of the internationalisation of South African firms and financial system, and the reduction of migrant-labour-related financial inflows to the rural areas, which affected the dynamics of accumulation and labour reproduction in the south of Mozambique. The opportunities arose from the expansion of South African investment, geared towards the core of the extractive economy (aluminium, natural gas, forestry, sugar and tourism), related infrastructure (Matola and Maputo ports, the Maputo–Witbank toll road, and railways) and financial services (ABSA Group, FirstRand Group and Standard Bank). The challenges limited the scope for industrialisation through backward and forward linkages, while the opportunities were created in finance and the extractive core of the economy, where linkages with Mozambican firms developed only in services, maintenance and the simplest industrial activities (Castel-Branco 2020b, 2017; Castel-Branco and Goldin 2003; Biggs, Nasir and Fisman 1999; Maia 2019).
Mozambique’s capitalist manifesto in the global debate about the state and the national bourgeoisie
Many elements of this nationalist debate resonate in other countries in Africa, Latin America, Asia and advanced capitalist economies. The role of national bourgeoisies, their political legitimacy and the promotion of their class interests via the state in a context of increasingly unequal development of capitalism in the age of neoliberalism are truly global issues. The argument for building a strong national bourgeoisie while maintaining foreign capital markets and FDI as drivers of growth fits well with the extensive literature on late development and industrialisation, which has rejected market fundamentalism in favour of state-guided, national capitalist industrialisation (Amsden 1989, 2004; Amsden, Di Caprio and Robinson 2012; Chang 1996, 2002; Jones and Sakong 1980; Kim 1997; Kholi 1994).
At an academic conference, organised by the Human Sciences Research Council and held in Johannesburg in October 2007, Professor Adebayo Olukoshi, then executive secretary of the Council for the Development of Social Science Research in Africa (CODESRIA), presented a paper on the intellectual, political, social and economic challenges of establishing a new relationship between post-apartheid South Africa and the rest of Africa. His analysis of the debate about the rapid expansion of South African big business across the continent and its impact on the interests of accumulation and reproduction of national capital in each of the other African countries highlighted two issues. First, the historical heritage, the racial basis and the constraining impact of South African capital on development options and possibilities in each of the other countries raised the central question of to what extent was South African white monopoly capital African. Second, what could African countries do to resist more successfully the unrestrained penetration of South African big businesses, which were not only constraining the development of national capitalists but also shaping the regional patterns of development? Whereas South African white monopoly capital was classified as imperialist, it was also argued that the ability to restrain its expansion and protect national alternative development options was correlated with the promotion of the national capitalist bourgeoisie, which would have the capacity to draw on the national state to its own advantage – that is, to ‘capture the state’. This is consistent with Guebuza (2004), who argued that Mozambicans should not be afraid of becoming rich, as they deserve to and can do so, and with the African Union’s NEPAD economic development programme (African Union 2001), which claims for the private sector the leading role in promoting development and economic growth in Africa, reducing the role of the state to that of a provider of the institutional conditions for private-sector development. These include the fundamentals of the Washington Consensus, namely privatisation, liberalisation and macroeconomic stability (meaning fiscal and monetary austerity), together with public–private partnerships in developing the so-called human capital and provision of social safety-nets.
Thus, the African alternative to non-African capital became the promotion of national, i.e. African, industrial capitalism. It follows that, by historical construction, economic nationalism faces the contradictory imperatives of having to develop a critical rhetoric against neoliberalism and imperialism and legitimising the use of state power to promote national, private accumulation of capital, which also requires facilitating the alliance with international capital in neoliberal terms. This reduces the so-called African renaissance to an expansion and consolidation of African capitalist financial oligarchies, rejecting the fundamental criticism of capitalism as a system of accumulation and the struggle for alternatives to capitalism (Castel-Branco 2007; Tutashinda 1978; Shivji 2009). This is a risky political agenda, vulnerable to the volatility and speculative nature of global capitalism, and one that is contrary to the demands and aspirations of the African working people (Shivji 2017).
Conclusions
This article has investigated the objective historical conditions under which the characteristics of the Mozambican mode of capital accumulation, the extractive economy, have developed. It analysed the historical logic of capital accumulation in Mozambique through the lens of linkagency, following Fine and Rustomjee (1996), thus helping to operationalise Marxist political economy analysis. The article showed the power of the existing structures of accumulation to resist change and to impose limits and volatility, through systemic crises of accumulation and reproduction, on the direction and speed of economic transformation. Any transformation of the extractive economy requires careful macroeconomic management and planning so that the process of change does not quickly run out of steam, as was the case with the PPI (1980–1990) and the period of fast FDI growth (2009–2014) associated with the extractive core of the economy.
The paper argued that without locating these problems within their class dynamics we cannot fully understand the essence of such a pattern of specialisation and the challenges associated with it, nor can we deal with it in any meaningfully transformative manner. Hence, methodologically, the paper stressed the importance of analysing the historical logic of the mode of capital accumulation from a political economy perspective, of which the pattern of specialisation is a specific, concrete manifestation rather than the essence.
It could be asked whether a different type of economy could have emerged from the ashes of the crisis of the state-centred system of accumulation. This apparently obvious question might not be, upon closer analysis, the most essential and relevant, as it cannot lead to objective answers because it is based on suppositions. The current system of accumulation is what emerged out of different sets of possibilities, and our first intellectual and social responsibility is to understand why this was so and what we can learn from it.
Class-structured conditions, which encompass the historically determined structures of accumulation and the specific linkagencies that develop, are critical engines of development trajectories. Social analysis may have to focus on the understanding of such class-structured conditions, tensions and struggles to articulate what the struggle is about – for example, protection of the emerging national capitalist bourgeoisie or the alliance and emancipation of the broadly defined working classes or working people, following Shivji (2017), and support for the organisation and mobilisation of class struggle.
In this context, we need to recognise that a legacy of the bureaucratic and authoritarian state-centred mode of accumulation, combined with a dualistic perspective on the agrarian question, were instrumental in preventing progressive political, economic and social alliances from developing as powerful alternatives from the working classes, as these have implications for the motivation and forms through which working people may or may not engage politically with any form of revolutionary change.
To conclude, we would like to emphasise three areas of action for change. First, it is the importance of rigorous theoretical analysis of class-structured dynamics of conflict, tension and struggle which should show us that structures of accumulation can be changed, but they cannot be changed successfully at will alone, as humans write their history but not under historical and social conditions chosen by themselves. Dynamic social structures of accumulation impose limits to change that need to be conquered and surpassed, and create the social relations, social groups and social interests and struggles that will help to overcome them. This also reveals the issues around which social movements and class struggle can be organised, and the social and class alliances that are required for successful change. Finally, this analysis should provide a way of articulating short- and long-term reformist and revolutionary struggles and change. We know, for example, that an extractive economy cannot sustain broad-based and substantial improvements in social welfare and taxation. While narrowing the struggle to better social redistribution is pointless, inserting that struggle within a path of revolutionary transformation of the patterns of accumulation and labour reproduction links day-to-day struggles that affect day-to-day life to the mobilisation of workers and peasants with the essential and longer-term goal of revolutionary transformation of the capitalist mode of accumulation.
Second, no such social changes and transformation can take place without the organisation, mobilisation and articulation of social movements that give political expression and provide the strategies and tactics for political and social struggle. Achieving this requires social and political theory rooted in the specific historical and social conditions of struggle.
Third, Mozambique’s social system of accumulation and associated contradictions and struggles are located in a particular geographic space, that of southern Africa, and particular time, that of financialised neoliberal capitalism, such that successful revolutionary analysis and struggle require the understanding of the conditions and issues around which the internationalist and solidarity dimension of the struggle can be organised, mobilised and consolidated. Capital is articulated globally and so too should be the resistance of labour and alternatives to capital.