Classes of working women in Mozambique: an integrated framework to understand working lives

Abstract

Feminist political economy has illuminated the gendered dimensions of the globalisation of production. Whilst this literature provides essential insights on gendered exploitation in export-oriented industries, women’s work in localised labour markets in the Global South remains underexplored. This paper seeks to address this gap by putting into dialogue three bodies of literature – feminist political economy of globalisation, political economy of development in southern Africa and the social reproduction of the everyday. It proposes an integrated conceptual framework to analyse women’s working lives and applies it to northern Mozambique. It makes two key findings. First, the lives of those working in localised labour markets are shaped by global capitalism through extreme fragmentation of labour regimes forcing people into multiple precarious forms of work – a process that entails the appropriation of women’s productive and reproductive labour. Second, the imperatives of social reproduction shape employment trajectories and expose differentiation among working women, seen for example through gendered constraints to mobility, care obligations and contributions to ceremonies. The implications are that the analysis of women’s working lives needs to capture three key aspects: social differentiation among working women, temporal and spatial dynamics of the everyday and the centrality of the reproduction of the social.

Keywords:

• Social reproduction
• work
• gender
• labour regimes
• Africa
• Mozambique

Introduction

With production becoming globalised, attention has been paid to women’s increased participation in employment across the Global South (Rai & Waylen, 2013). The literature on the feminisation of labour and gendered labour markets provides pivotal insights on women’s participation in employment, highlighting how working conditions reflect women’s disadvantage and reproductive obligations. While the conceptualisation of labour markets as gendered institutions that expose women workers to super-exploitation (Elson, 1999; Elson & Pearson, 1981) is the obvious starting point for those interested in women’s work in the South, this literature is focused on employment in industries integrated in global production networks and focuses on gender inequality at the expense of other forms of inequality that shape differentiation among working women. Does this literature provide a useful theoretical framing to analyse the gendered dynamics of work in localised labour markets at the margins of global production? This is an important question for understanding work in contexts, such as that of Mozambique, where the dominant form of employment entails engagement in multiple occupations including wage work, non-wage work and cash-earning activities that are fully embedded in localised labour markets, that is markets not linked to global production networks and at best hinging on regional networks.

To capture the gendered organisation of work in localised labour markets, we need to consider how women’s work is embedded in processes of social differentiation as well as shaped by the imperatives of social reproduction. Thus, I draw on the literature on trajectories of capitalist development in colonial and post-colonial southern Africa to gain a historical perspective on processes of livelihood diversification and social differentiation, including some of the gendered dimensions highlighted in this body of work (Bernstein, 2010; O’Laughlin, 1998), to contextualise women’s work in the dynamics of capital accumulation in southern Africa. I then combine this literature with insights based on social reproduction, racial capitalism and the international political economy (IPE) of the everyday, particularly inspired by the work of Katz (2001a), Federici (2004), Bhattacharyya (2018), and Elias and Rai (2019), and consider the mutual constitution as well as tensions in the organisation of productive and reproductive work, which further enables a perspective on gendered working lives and differentiation among working women at the margins of global production.

Hence, this paper brings together insights from three bodies of literature – the feminist political economy of globalisation, the political economy of capitalist development in southern Africa, and the social reproduction of the everyday – to construct an integrated framework for the conceptualisation of women’s working lives. It is aimed at addressing two main questions. First, how are working lives in localised labour markets shaped by global capitalism? Second, how do social reproduction imperatives shape working lives?

To address these questions, I draw on empirical evidence collected in northern Mozambique, in the province of Cabo Delgado, in 2011–2012, although the analysis is informed by a longer-term research engagement in Mozambique, which is ongoing at the time of writing. Data collection took place in an urban area (Pemba), a peri-urban one (Metuge) and a rural one (Mueda), which offers a perspective on interconnections across space. I draw on qualitative data gathered through in-depth interviews, life histories and participant observations as well as quantitative data collected through a household survey. As part of the survey, 120 households were randomly selected across the three areas, and one woman was interviewed in each household, with the aim to document women’s experiences and their knowledge (Harding, 1987). A feminist approach provided the epistemological foundations for mixing methods and for reflecting on the positionality of the research participants vis-à-vis researcher and research assistant. Existing data on northern Mozambique is both scarce and biased, as it will be explained in section 5; thus, it was necessary to engage in primary data collection to gather a detailed understanding of the rhythms of everyday life and a mixed-methods approach allowed for extending the type of evidence typically used in (political) economic research (Berik, 1997). All data was collected by the research assistant, Eusebio Tissa Cairo, and myself; throughout the research process we reflected on how questions were posed and interpreted by the research participants in light of the relations of power at play between the research participants (local communities, mostly women, with low levels of education and income on average), the research assistant (a local man from one ethnic group, of the three involved in the research, and with a higher level of education) and the researcher (a woman and an outsider to the local communities) (see Deane & Stevano, 2016). Various dynamics shaped the research activities in ways that, owing to the focus of this paper, cannot be fully explained here (see Deane & Stevano, 2016); whilst gender dynamics created a proximity between the researcher and the research participants, this was trumped by the difference in nationality, race, income and education, which was somewhat managed through a long fieldwork of about nine months and multiple visits that served to build a relationship of confidence and respect with the research participants.

The analysis shows that, although localised labour markets are detached from global production networks and dynamics, the lives of those working in these markets are shaped by the dynamics of global capitalism through extreme fragmentation of labour regimes, commodification of life and erosion of ability to rely on household production for social reproduction. People are forced into multiple, often highly precarious, occupations through processes of livelihood diversification and social differentiation that have important gendered dimensions that account for the appropriation and devaluation of women’s work. The nature of these localised labour regimes is best understood as evolving from the legacy of colonial relations and the ensuing trajectories of accumulation. At the same time, the imperatives of social reproduction shape employment trajectories and the organisation of working lives in ways that allow us to see the differentiation among working women. Those who have remained at the margins of global circuits of production, most often only sporadically integrated into or violently expelled from them in gendered and racialised ways (Bhattacharyya, 2018; Sanyal, 2007), find themselves struggling to ensure their social reproduction whilst the latter as a means to reproduce life governs their social relations and everyday working lives.

This paper makes a contribution to the literature on Mozambique by providing rich empirical evidence on a region of Mozambique that remains under-studied despite attracting significant international interest in recent times owing to the discovery of natural gas and the ongoing insurgents’ attacks that are threatening the profitability of large-scale investments and ravaging people’s livelihoods (Morier-Genoud, 2020; Rawoot, 2020). With rare exceptions, the international media coverage of the conflict in northern Mozambique is seldom grounded in an understanding of everyday life, in fact the existence of people living in the territories that are now of interest to international investors is downplayed and erased; thus, the empirical evidence discussed in the paper, even if it was collected back in 2011-2012, can make an important contribution to centre the lives of people and the organisation of working lives in the emerging accounts on the province of Cabo Delgado in Mozambique.

The paper makes also a conceptual contribution to feminist analyses of work because the fragmentation of working lives is not unique to northern Mozambique but typical, if in context-specific declinations, of racialised peoples in peripheral areas of global capitalism (Bhattacharyya, 2018; Li, 2010; Naidu & Ossome, 2016; Ossome, 2020; Shivji, 2009). I argue that it is necessary to extend the empowerment versus exploitation debate to consider the patterns of social differentiation among classes of working women, which requires the analysis of material realities and social relations that shape women’s lives outside the workplace. Understanding women’s experience of wage work or paid work hinges on seeing how this type(s) of work intersects with the other forms of work and non-work they perform in their everyday lives. In this sense, it is necessary to turn the relation between production and reproduction on its head and see not only how production shapes social reproduction but also the other way around (see also Mezzadri and Majumder, 2020). Furthermore, a social reproduction lens needs to account for the centrality of the reproduction of the social (Hornby & Cousins, 2019; Stevano, 2018), which is a specific dimension of social reproduction that highlights the centrality of making contributions to ceremonies and rites as a means to reproduce kinship and family structures and to ensure membership to a social group and its survival. The analysis illuminates how temporal and spatial dimensions dictate the rhythms of the everyday and materially shape social reproduction, which adds empirical material to the notion that time and space are foundational categories of the everyday (Elias & Rai, 2019). Working lives are structured around precarious and exploitative dynamism created by the spatial dispersion associated with the fragmentation of labour and the means of social reproduction, with households and familied stretched across space in ways that put a strain on family-centred social reproduction.

The paper begins by setting out the components of the integrated framework to conceptualise women’s working lives. It then moves on to outline the characterisation of colonial labour regimes in northern Mozambique (section 3) and then the contemporary ones, highlighting key aspects of continuity and change (section 4). Section 5 analyses the organisation of women’s work and gendered labour processes. Section 6 illustrates the mutual constitution as well as the tensions between production and reproduction. Finally, section 7 concludes.

An integrated framework to conceptualise women’s working lives

The shifting gendered division of labour spurred by globalisation processes is a key area of study in the field of feminist political economy (FPE henceforth) (Rai & Waylen, 2013). Liberalisation of trade and financial markets alongside advancements in information communication technology (ICT) reshaped global production from the 1970s onwards, with the relocation of production from the North to the South (Bair, 2005; Baldwin, 2012). This process triggered significant changes in the global division of labour, in particular it led to the creation of employment in export-oriented agriculture and manufacturing, which was accompanied by an increase in the number of women in the labour force (Barrientos et al., 2004; Benería et al., 2016). From the 1980s, FPE has analysed these processes and described quantitative and qualitative changes in women’s participation in the workforce. A core component of the debate was the theorisation of the feminisation of labour in both quantitative terms, as the number of women employees increased in relation to that of men, and in qualitative ones, as precarity grew whilst working conditions and wages declined (Barrientos et al., 2004; Çağatay & Özler, 1995; Standing, 1999).

A key insight from this literature is the conceptualisation of labour markets as gendered institutions (Elson, 1999). Women’s disadvantage – visible through their responsibilities for care and domestic work, limited participation in public life, lower education and skills – is reproduced in the labour market (Elson & Pearson, 1981). As women enter the labour force on unequal terms, they are exposed to super-exploitation (Ibid.). Through informal and flexible work arrangements women have access to low paid and insecure jobs while the value of non-wage work they perform continues to go unrecognised (Elson, 1999). Through the integration of women in the labour force, export-oriented firms ensure access to cheap labour and hence competitiveness in global markets while reproducing gender inequality (Berik, 2017; Elson & Pearson, 1981; Seguino, 1997). In processes of capitalist development, women’s suffering in wage work is ‘directly rooted in their function as unpaid labourers in the home’ (Federici, 2004: 95).

The cheap labour and exploitation arguments have not been uncritically accepted. Other feminist work has stressed that, although women are drawn into the labour force because of economic necessity (Casale & Posel, 2002; Kabeer, 2012), participation in the labour force can be emancipatory (Kabeer, 2012; Kabeer & Mahmud, 2004). This research investigated the conditions under which employment can increase women’s options and found that formal and semi-formal employment that ensures higher wages and job security can offer an empowerment path through increased decision-making power and economic autonomy (Heintz et al., 2018; Kabeer, 2012).

Although this debate remains central in feminist scholarship, the polarised evidence and views it has generated on the effects of women’s participation in paid productive work obscure important aspects of the dynamics of gendered oppression and exploitation. A theoretical framework for women’s employment across sectors in the Global South is to be rooted in a social reproduction extension of the cheap labour and exploitation theses as a means to, first, centre how exploitative relations materialise at the intersections between reproduction and production and therefore are not alleviated or worsened only through participation in productive work (see Ferguson, 2019) and, second, to capture patterns of social differentiation among classes of working women based on social relations shaped by multiple inequalities. A re-framing of these arguments in these terms serves also to overcome a limitation of the FPE literature on global division of labour, which is overwhelmingly focused on the export-oriented industry, the neoliberal phase of capitalism and gender inequality, at the expense of analyses of women’s work in localised labour markets, colonial legacies and the co-constitution of gender, race, class, migration status, age inequality.¹ Some of the most recent literature on labour in global production networks significantly advances our understanding of the interplay between reproductive and productive work (e.g. Arslan, 2020; Baglioni, 2018; Gore & LeBaron, 2019; Mezzadri, 2016; Mezzadri and Majumder, 2020); this paper builds on the insights of this literature and applies them in the underexplored case of localised labour markets, which are central to the vast majority of the poor’s working lives.

To shed light on processes of social differentiation, a stream of regional literature on southern Africa provides key insights in the context of colonial and post-colonial transformations. Marxist analyses document how colonial labour regimes, hinging on migration in the region, constituted the basis of processes of capital accumulation and shaped post-colonial labour regimes and peasant production in racialised ways (Arrighi et al. 2010; Bernstein, 1979; Wolpe, 1972). However, class relations in rural Africa escape the dualistic dimension typical of industrial Europe, and are more fragmented, with multiple classes of labour relying on varying combinations of agricultural production and wage and non-wage work (Bernstein, 2007; 2010; Cousins et al. 1992; O’Laughlin, 1996). Bernstein (2007, p. 6) defines classes of labour as those who ‘have to pursue their reproduction through insecure and oppressive – and typically increasingly scarce – wage employment and/or a range of likewise precarious small-scale and insecure ‘informal sector’ (‘survival’) activity, including farming in some instances’; the term highlights the lack of homogeneity in the condition of the proletarian and the importance of differentiation among workers. To this effect, O’Laughlin (1996) proposes that the process of rural social differentiation is decomposed into livelihood diversification and class stratification, which may unfold in parallel or not. In other words, deepening social differentiation may be linked to the necessity to diversify sources of household earnings and production, which do not necessarily and linearly correspond to class transitions. This is an important argument, which has the fundamental implication that intra-class inequality matters. These fragmentations are co-constitutive of class relations and show how capitalist development produces differentiation, not homogenisation into undifferentiated wage labour (Bannerji, 2011; Bhattacharyya, 2018; Mezzadri, 2016; Sanyal, 2007), as it will be illustrated in the next sections through an account of labour regimes in northern Mozambique from the colonial to the contemporary neoliberal era.

In Southern Africa, processes of social differentiation accelerated with the integration of male labour into the regional circuits of labour migration (O’Laughlin, 2002). The gendered patterns of labour migration led to the increased prevalence of women-headed, divided and fragmented households (Budlender & Lund, 2011; O’Laughlin, 1998). The literature on households was essential to shed light on the configurations and transformations of familial units underpinning the social fabric in the region (Guyer & Peters, 1987). The absence of male labour contributed to shifting gender roles in household management and farming, a transformation that marks a phase of the household development cycle (Murray, 1987; O’Laughlin, 1998). Shifts in types of crops cultivated – from more to less time-consuming – are also documented and show how lack of labour and time constraints were a challenge for many poor rural households reliant on women’s work (O’Laughlin 2013).

This literature on southern Africa provides an illuminating account of capitalist development in the region as shaped by colonial regimes of accumulation, how it impacted household configurations, family and gender relations in ways that altered household production and social relations. The attention for social differentiation lays the foundation for considering how classes of labour face both similar and different material conditions of everyday life. This literature is less telling on how, in turn, gender and social relations shape not only agricultural production but also gendered participation in casual wage work and cash-earning activities; how women’s embeddedness in family and social relations shapes their own employment trajectories. In short, it is necessary to overturn how we understand the interplay between productive and reproductive work: it may be that productive work is ‘supplementary’ to reproductive work, and not the other way round (Bhattacharyya, 2018). I propose to shed light on these relations by explicitly embedding this analysis in a social reproduction framework that allows us to see further elements of differentiation within and between classes.

Employing a social reproduction framework means taking a historical materialist perspective to consider the relationship between processes of production of value and those of production of human life (Bakker & Gill, 2019; Ferguson, 2018; Winders & Smith, 2019). A social reproduction framework can bridge various divides, such as between the biological reproduction of human life and the social reproduction of the labour force, between the household and the market or the state, and between micro and macro approaches (Katz, 2001a; LeBaron, 2010). Katz (2001a: 711) provides a useful and encompassing definition of social reproduction:

Social reproduction is the fleshy, messy, and indeterminate stuff of everyday life. It is also a set of structured practices that unfold in dialectical relation with production, with which it is mutually constitutive and in tension.

A social reproduction perspective makes two essential contributions to the conceptualisation of women’s work. First, it provides the basis for a unifying analysis of inequality. Labouring classes are divided through differences of gender, race, generation in ways that are central to the reproduction of the capitalist system (Bhattacharyya, 2018; Ferguson, 2018; Mezzadri & Fan, 2018; Mitchell et al., 2004). Various forms of oppression, being it gender, race, nationality and others, are neither separate nor cumulative; they are not even secondary to class oppression, in fact, they are constitutive of dynamics of exploitation and dispossession. The scholarship on racial capitalism strengthens these arguments by highlighting how capitalist development works through expropriation, exploitation and expulsion, where people are integrated in and pushed to the margins of capitalist production in multiple gendered and racialised ways; this leads to the possibility of expropriation not only through dispossession and to exploitation not only through absorption into wage work, but also through expulsion from wage work and relegation to forms of work that are considered to be scarcely productive, unproductive or even non-work (Bhattacharyya, 2018). Second, the geographies of social reproduction help trace connections between local realities and global processes (Katz, 2001a). Through a geographer’s perspective, Katz (2001a, 2001b) reminds us that the local is globally-constituted through material practices that are specific to historical moments and geographical locations. Both the spatial and temporal dimensions of social reproduction connect the everyday of local realities with global socio-economic transformations (Elias & Rai, 2019). This insight is central to capture the connections between localised labour markets and the changing global division of labour.

Drawing on a synthesis of insights from these rich bodies of literature, it emerges that an integrated theoretical framework to study women’s work has three main pillars. First, it adopts the notion that labour markets are bearers of gender and other forms of inequality. Second, it postulates that classes of labour are shaped by processes of social differentiation and are themselves internally fragmented. Third, both inter- and intra-class fragmentation is best understood through a social reproduction lens that captures the imperatives of reproduction, which unfold in dialectical mutuality as well as conflict with production.

Colonial labour regimes in Northern Mozambique

Mozambique is part of the area that supplied labour to the South African economy (Arrighi et al., 2010). However, much of the literature on labour migration in southern Africa focused on circuits of labour towards the South African economy while disregarding regional specificities (O’Laughlin, 1998). The North of Mozambique, where I conducted this study, has a different colonial labour history from the South because it was not integrated in the migratory flows to South Africa. Isaacman (1982) argues that the political economy of northern Mozambique during the late colonial time was marked by forced cotton production, temporary male migration to the sisal plantations of the coastal areas and voluntary migration to the sisal plantations of southern Tanganyika.

Having been placed under the direct control of concessionary companies that, with the assistance of the colonial state, coerced peasant households into cotton production, northern Mozambique became the core area of the forced cotton regime (Isaacman & Chilundo, 1995). Adverse agro-climatic conditions, low prices, peasant resistance and absence of technology, all meant that households’ ability to hire extra-familial labour was key to increase the production of cotton (Isaacman, 1996; Isaacman & Chilundo, 1995). A few households could employ extra-familial labour; these were, for example, local chiefs responsible for monitoring cotton production who were in turn allowed to use peasant labour on their cotton fields and occasionally were paid higher prices (Ibid.). The ability to mobilise agricultural labour was central to the acceleration of processes of rural differentiation in the region (Isaacman & Chilundo, 1995).

Alongside cotton, men from across the region temporarily migrated to the coastal areas to work in the sisal plantations (Alpers 1984; Isaacman 1982; O’Laughlin, 2013). Men worked on the plantations for several months and then returned to their homes (O’Laughlin, 2013). Instead, cotton production took place at the household level, with women and men coerced to grow cotton on separate plots of land. Notwithstanding the different systems of labour extraction, Isaacman et al. (1980) argue that the overall effect was a constraint of household labour availability. Cotton posed a threat to household food production, which resulted into widespread malnutrition and famine (Isaacman, 1996; O’Laughlin, 2013). Women faced the greatest challenges because they had to continue to perform domestic work and other reproductive obligations (Isaacman, 1996). In addition, tensions arose in relation to gendered responsibilities for household agricultural production, which was often performed by women and subsidised capital accumulation by cheapening labour during the colonial time, as it has been documented for Mozambique and the whole southern Africa (O’Laughlin, 2013; Wolpe, 1972).

Peasants resisted through sabotaging cotton production, artistic expression and out-migration. Migration to Tanganyika began as a strategy to escape excessive taxation imposed by the Portuguese and, later, intensified with the embitterment of the forced cotton system (Alpers, 1984). Some migrants settled in Tanganyika with their families while others continued to periodically return to Mozambique in spells of circular migration (Ibid.). The migration to Tanganyika had important economic and political implications in the region. First, returning migrants did not have access to land through kinship ties and used their earnings to buy land (Alpers 1984; Isaacman 1996). Second, while away, migrants got in contact with political movements of resistance to colonialism in Tanganyika, which later paved the way for the birth of the Frente de Libertacão de Moçambique (FRELIMO), which drove the country into the struggle for independence (Alpers 1984; Isaacman et al. 1980).

Through integration in regimes of migrant wage labour and forced labour, processes of social differentiation unfolded and accelerated (O’Laughlin, 2002). Households’ ability to hire extra-familial labour or to migrate, forcedly or as an act of resistance, are indicated as central to processes of rural and social differentiation and increased reliance on cash-incomes (Isaacman & Chilundo, 1995; O’Laughlin, 2002; Pèrez-Niño, 2017). In addition, the imposition of taxes and the creation of shops in rural areas added to the increased need to engage in forms of work somehow remunerated in order to participate in the cash economy (Feijó, 2016). Subsistence production persisted as peasants were not dispossessed of their land, which means, in some orthodox understandings of class formation, that a clear process of proletarianisation did not take place. However, O’Laughlin (2002) argues that proletarianisation occurred nonetheless, through the commodification of life and increasing ties to global markets, which imply reproduction is no longer possible outside the circuits of commodity of production (Bernstein, 2010).²

The fragmentation of contemporary labour regimes

In the transition from colonial to post-colonial Mozambique, the socialist state posited itself as the engine of economic activity for the creation of a unified country, where participation in the labour force, for both women and men, was seen as a key vehicle for the promotion of modernity (Feijó, 2016). During the first phase of the socialist period (1975–1980) FRELIMO sought to socialise production by nationalising large farms and turning small ones into cooperatives (O’Laughlin, 1996). However, the socialist era did not radically dismantle colonial labour relations as, on the one hand, it reproduced the economic dualism of production based on large farms versus small-scale commercial production, which did not reflect the reality of differentiation in rural areas (Harris, 1980; O’Laughlin, 1996; 2017) and, on the other hand, engagement in paid work maintained a disciplining character (Feijó, 2016). The racialised conditions of production and life were in many ways upheld despite the end of the colonial rule. FRELIMO abandoned any plan for a socialist agrarian strategy in 1986, after the death of Samora Machel, and officially embraced the politics of liberalisation and privatisation (O’Laughlin, 1996) that ushered the country into the on-going neoliberal era.

Whilst desirable, accounting for a detailed process of change in labour regimes in northern Mozambique is not allowed by the accessible data; here I focus on providing a snapshot of the contemporary neoliberal labour regimes in the province of Cabo Delgado, which is missing in the current literature, and draw some reflections on how the contemporary conditions of exploitation compare to those predominant in the colonial time.

Contemporary labour regimes revolve around pockets of export and commercialised agriculture, which are mostly developed in peri-urban areas and in the southern districts of the province of Cabo Delgado due to better road infrastructure and proximity to the Nacala corridor, which is the main export channel in northern Mozambique. Private investments in agriculture and agro-industry account for approximately 18 percent of total investments in the province (Muianga et al., 2012). The agro-industry is scattered and overwhelmingly concentrated in timber extraction.³ The majority of companies exporting timber are foreign, but the largest one is the Mozambican Miti Ltd.⁴

Single agro-industries of medium capacity operate in production and processing of key export crops: cotton and cashew-nut. At the time of fieldwork, cotton production was restricted to a few southern districts, where the British multinational Plexus had concession areas. Plexus provides small-scale farmers with inputs and then buys cotton at a pre-established price. In 2012, Plexus branch in Cabo Delgado employed 157 permanent workers and 4691 seasonal workers. If Plexus’ operations in the neighbouring province of Nampula are accounted for, Plexus would be the largest employer in northern Mozambique. On the other hand, cashew nut is mostly grown in the northern districts of the province by small-scale producers. In northern Mozambique, the importance of cashew has been on the rise since the mid-2000s, with the emergence of several small and medium-scale processing plants in the ‘cashew triangle’ in Nampula and surrounding areas (Paul, 2008). In 2012 as well as today, there is one cashew processing factory in Cabo Delgado, Korosho, owned by the Indian multinational ETG World, which also exports raw cashew nut and employing 350 workers, most of whom are women.

A significant share of private investment, estimated at around 78 percent, goes to the tourism industry, which tends to be located in the coastal areas and islands (Muianga et al., 2012). Interviews with entrepreneurs revealed that it is common for private investors to have businesses in more than a sector, thus we observe a concentration of enterprises in the hands of a few owners in a landscape of relatively stunted enterprise development. In addition, significant socio-economic transformations began to take place in Cabo Delgado in 2012 due to the discovery of natural gas. However, owing to the embryonal phase of these investments at the time of fieldwork and significant delays in the extraction operations,⁵ it was not possible to gather evidence to reflect on how this discovery might have shaped labour dynamics.

Beyond these scattered areas of enterprise, the labour regimes in Cabo Delgado are characterised by vast localised informal labour markets. The only statistics available on the informal economy indicate that, in 2004, 77.9 percent of the labour force in Cabo Delgado worked in the informal economy and only 4.3 percent in the formal sector (INFOR, 2004). In addition, although official statistics grossly underestimate workers’ involvement in multiple occupations (see next section), they do report that above 80 percent of those who are involved in secondary activities in northern Mozambique (above 90 percent for the province of Cabo Delgado) do so to increase their overall income (IFTRAB, 2006). Primary research shows that Cabo Delgado residents often engage in various types of work, including farming, casual and seasonal wage work, trading, sale of agricultural produce and other goods, and so forth. The processes of livelihood diversification that accelerated during the colonial time have further advanced, with the progressive erosion of households’ abilities to reproduce themselves through small-scale subsistence agriculture. Commonly, households use earnings from a form of work to finance another cash-generating activity.

Consider this story, collected in the district of Metuge, close to Pemba, which illustrates how processes of livelihood diversification can transform the material conditions of living. Eusebio, from a cotton-producing district, married Catarina, who is from Nampula province, and together they moved to Metuge, where Eusebio had some relatives. Eusebio tried to grow cotton with scarce success, as Metuge is outside the concession area managed by Plexus, and later switched to the production of sesame seeds. Networks of buyers from Pemba and NGOs’ support for seeds’ production contribute to making this cash crop viable in the district. With the earnings obtained from sesame seeds sales, Eusebio hired workers to build a new house, purchased a plot of land as well as a motorbike. In addition, he uses these earnings to run a small shop, to finance Catarina’s petty trading activity, and to hire agricultural labourers. Catarina and Eusebio used to work as casual agricultural labourers, but stopped in 2008, when the earnings from sesame seeds were sufficient not only to free them from the need to engage with agricultural wage work but also to be able to run other cash-earning activities and hire agricultural labourers.

Central to the processes outlined is the significance of past conditions and social relations in shaping trajectories of migration and work. In particular, the link between cotton production in the recent past and contemporary cash crop production, contingent on the proximity to urban markets, highlights the importance of understanding labour organisation through socio-economicdynamics of continuity and change. In a study of cotton-growing households in Nampula, Pitcher (1996) documents that the largest landholdings had been acquired through the market, despite the persistence of mechanisms of matrilineal land inheritance. In the case of Catarina and Eusebio, the ability to acquire land lies at the basis of their path of livelihood diversification. These are ‘classes of labour’ that combine practices of capital and labour whilst being forced to sustain their social reproduction through the multiplication and fragmentation of work, which may ensure survival while further entrenching the reproduction of life and society in highly precarious livelihoods (O’Laughlin, 2013).

Access to land continues to be central to contemporary livelihoods in northern Mozambique, as it is in other parts of Africa (Ossome, 2020), with urban households often trying to gain or maintain access to land in rural areas to produce for own consumption or sale. Around 70 percent of the survey respondents, including both rural and urban households, have access to land. Further, the ability to hire agricultural labour continues to be key to processes of social differentiation. This can be illustrated through an exploration of the veteran pension scheme, which I identify as an important component of the political economic fabric of contemporary Cabo Delgado. With the revolution for independence having kicked off in this part of Mozambique, the province has a high number of living veterans. In 1994, President Chissano introduced a pension scheme to recognise the veterans’ contributions to Mozambican independence. Local governmental authorities had a record of 26,000 veterans receiving the pension in 2012. In my survey, 10 percent of the respondents are recipients of the pension and another 31 percent has someone in their household or family receiving the pension.

Pensions are monthly income transfers whose level depends on when the recipient began to receive the transfer and the position covered in the war. At the time of fieldwork, the lowest pension was 4,582 MZN (equivalent to 88 GBP) while the minimum wage ranged between 2,300 MZN and 5,320 MZN depending on the sector. Some interviewees reported receiving up to 9,000 MZN per month. Pensions are important because they represent one of the very few sources of regular income. In a context characterised by high unemployment, scarce regular employment opportunities – especially for women – and widespread informality, access to a regular income shapes material conditions of living. Although women‘s pensions tend to be lower than men‘s because of women’s lower positions in the military hierarchies, they are one of very few sources of regular income for women.

Pensions shape marital relationships and household dynamics, with younger people relying on older relatives’ pensions and women building their own houses as safety nets for their children and in case of marital disruptions. An interviewee lamented that young women who can rely on their mothers’ or grandmothers’ pensions want to have children without getting married. Many women who receive a pension explained that they built houses for themselves and their children, countering the custom that men are responsible for providing a house for the family.

‘[…] we managed to get a tin roof for our houses, build a well to collect rain water, buy mattresses and chairs, framed beds, hire agricultural labourers and send our children to school’, qualitative interview, two women who receive a pension, Namaua (Mueda), 15th November 2011

As illustrated by the quote, pension recipients hire agricultural labourers to the extent that ‘being a veteran’ is colloquially used as equivalent to ‘being a labour hirer’. Research participants explained that when they look for casual wage work, they go to the houses of the veterans to inquiry as to whether they need labourers. All the pension recipients interviewed hired agricultural labour, at least during the season when agricultural work is more intensive.

In contemporary northern Mozambique, agricultural production remains an important source of sustenance and earnings, but its main role is to provide resources and to free time, through reliance on agricultural labour, to engage in other cash-earning activities. The ability to hire agricultural labour continues to be a significant dimension of social differentiation, although not necessarily of class stratification. Those who manage relative better off positions in the local context, through labour hiring for example, do not always – in fact most often do not – fully take the position of capital.

If we assess the contemporary conditions of exploitation in these fragmented labour regimes against the dynamics that were prevalent in this region during the colonial time, we can observe that colonial exploitation was based on the extraction coerced labour that directly served the interests of colonial capital and on the indirect extraction of value from the work of those left outside of the circuits of migrant labour, although the latter has been traditionally classified as unproductive work or non-work (Bhattacharyya, 2018; Federici, 2004). Whilst unpaid, reproductive and informal work continue to function as a subsidy to capitalist production in contemporary labour regimes, it appears that the current conditions of exploitation are shaped by patterns of expulsion from the possibility of formal wage work and relegation to fragmented and precarious livelihoods.

Women’s employment and gendered labour processes

There is a significant gap between official employment statistics and the reality of employment in northern Mozambique. The official data indicates that 80 percent of women are employed in agriculture and only 2 percent of them have a secondary off-farm occupation in the province of Cabo Delgado (IFTRAB, 2006). However, according to the survey we conducted in 2012, 75 percent of the women interviewed performed casual wage work and/or cash-earning activities, in addition to farming their own or family’s land. These results are in line with studies documenting the importance of rural labour markets and off-farm employment in Mozambique and other countries in southern Africa (Cramer et al., 2008; O’Laughlin, 2002).

Problems with data collection through labour force surveys in settings where employment is informal and casual have been largely discussed in the literature, with particular reference to the difficulty of capturing women’s employment through conventional questionnaires (Bardasi et al., 2011; Berik, 1997). In the Mozambican context, these statistics also feed into the long-standing narrative that poor rural residents are mostly small-scale farmers, a belief that underpins contemporary debates on the lack of ‘work culture’ among the poor and in rural areas (see O’Laughlin, 2016; Revés, 2011). Besides the problematic use of cultural arguments to explain approaches to work, which reinforce racist and classist tropes on ‘lazy Africans’, the portrayal of Mozambicans as farmers who lack the appropriate attitude to perform non-farm and wage work could not be further from the truth once it is acknowledged that labour markets are extremely dynamic and multiplicity of occupations is a necessity for many.

Women work as farmers, casual wage workers, as traders of agricultural produce, preparing food and drinks for sale, as market and street vendors, trading fire wood and charcoal, and as domestic workers. These occupations differ substantially in regularity, earning levels and mobility. Besides farming, whose rhythm is dictated by agricultural seasons and crops, the performance of other types of work ranges from regular (similar to full-time jobs) to sporadic (a few times per year). Often, how frequently a cash-earning activity is performed depends on whether it can be financially sustained, including being able to cover the transport costs when working entails spatial mobility. Cash shortages were often mentioned as the most important reason for abandoning an occupation, a finding that bears significant gendered implications in processes of livelihood diversification. As illustrated by the story of Catarina and Eusebio described above, it is common that women’s participation in remunerated work is sustained through male incomes. This means that the acquisition of goods for sale and/or transportation costs may be covered by a man’s income. Although these intra-household arrangements may reflect cooperation, they also create mechanisms through which women’s work is de facto regulated and controlled by men. Further, the precarity of women’s livelihoods is exacerbated as, in the event of household financial crises, they may be the first to have to quit their occupation. The gendered dimensions of processes of livelihood diversification are crucial to understand women’s working lives.

Regarding responsibility for deciding whether a woman should engage in remunerated work, survey data reveals that 44 percent of the interviewed women said that it is their own decision, 31 percent reported that it is their husband’s decision and another 22 percent said the couple decides jointly.⁶ If only married women are considered, it is clear that women need to obtain permission from their husbands if they want to work. Even when women describe their participation in paid work as resulting from their own initiative, it emerged through qualitative interviews that they had the obligation to run the idea through their husband. In some cases, women take up a cash-earning activity upon suggestion or request of their husband, as part of men-led household economic planning.

Women’s work is framed as ‘helping’ the husband and the family and often dismissed as insignificant. This framing devalues women’s work and contributes to make it invisible. Across many structured and semi-structured interviews, a consideration of women as just farmers emerged, by both women themselves and men. This reflects the socially-recognised role of women as responsible for household agricultural production, but it obscures the importance of other types of productive and reproductive work that women extensively engage with in their everyday lives. Although women’s earning tend to be lower than men’s, they are nonetheless essential to the household’s sustenance, particularly as additional sources of income are necessary to manage fluctuations due to seasonal, casual and informal work. The path of capitalist development in northern Mozambique is one that has not led to the formation of neat classes nor it has created a clear-cut separation between the loci of production and reproduction. Thus, gendered control and appropriation of labour transcends fictitious reproductive-productive divides, going beyond the appropriation of women’s reproductive labour as argued by Federici (2004) and encompassing women’s remunerated work too, and is central to the diversification of work and income sources necessary to ensure social reproduction in a context characterised by extreme fragmentation.

Further, earning levels vary depending on occupation type, regularity and location. Some types of work may be better remunerated than others, but significant variations are observed within occupations, with those able to work more regularly and in more favourable locations having access to higher earnings. Thus, time, availability of other sources of income and intra-household relations are all critical to how women approach remunerated work and make use of their earnings. Women report very different ways to making income contributions to their household, from negligible contributions to substantial ones, at times equivalent to being the sole or main earners in the household. Their contributions not only depend on how much they earn, but also on who else takes economic responsibility for the household, whether there are marital conflicts that may lead to concealing income, and on other extra-household economic obligations. Recognising this heterogeneity highlights how working women are a socially differentiated category, whose traits of difference are largely shaped by social reproduction dynamics and imperatives.

Social norms on household income provisioning indicate that this is undoubtedly men’s responsibility. Yet, it is evident that practice deviates substantially from this normative ideal. These mismatches are accounted for by considering women’s engagement in remunerated work as exceptional, non-ideal, but also as a form of help towards the husbands and, in some cases, as empowering.

‘Nowadays a woman cannot sit in the house and wait for her husband!’ Qualitative interview, woman, Metuge, 14th May 2012

‘These days we are living off my earnings. He [the husband] lost his job and is looking for a new one. I use my earnings to buy caril (relish that goes with the staple in a meal), but it is difficult. When he had a job, he used to buy food.’ Qualitative interview, woman, Pemba, 7^th July 2012

Many women describe the ability to earn an income as liberating from the reliance on a man’s income. However, they also expect to receive a cash allowance from the husband to purchase food and household goods.⁷ It is in the dialectical relation between norms and practices that is possible to capture the centrality of social and gender relations in shaping labour processes.

Consider Beata’s story, which shows how women’s work trajectories are shaped by changes in marital relations. Beata lives with her six children after her husband passed away in 2011. Beata reported three sources of income during the interviews: casual wage work in agriculture, trade of firewood and the earnings from the eldest son’s small stall. Beata explained that she began to work as a casual agricultural labourer and to run her cash-earning activity after divorcing the first husband, then stopped when she re-married and only started again after the passing of her second husband. Indeed, widowed and separated women are over-represented among those engaging in wage work and cash-earning activities. By the same token, more women in rural areas engage with multiple forms of remunerated work than the urban counterparts, which shows the resilience of norms on gendered responsibility for household provisioning. When men have access to higher and more stable incomes, via better paid and more regular jobs available in urban areas, then it appears that norms that depict men as the primary income providers trump the necessity for women to participate in remunerated work.

The gendered nature of processes of livelihood diversification and gender norms on household provisioning, however contrasting with the reality of women’s work, shape how women’s work is devalued and appropriated. Reproduction imperatives lead to the unfolding of material and social practices that deviate from norms and, through these deviations, it is possible to see how gendered constraints persist and intertwine with income and material deprivations in shaping working lives.

Social reproduction and work

The gendered character of the organisation of work intertwines with processes of livelihood diversification and social differentiation, leading to gendered and differentiated classes of labour. How do we capture the differentiation that shapes women’s working lives? Borrowing Katz’s (2001a) framing, a social reproduction lens allows for capturing practices that are mutually constitutive and in tension. It is through the co-constitution of production and reproduction as well as through the conflicting pressures on time that we can see how work ought to be understood not only as shaped in the workplace, but as reproduced through the practices of everyday life. The blurred separation between the workplace and other loci of life in the context of multiple and fragmented livelihoods help us capture these fundamental connections.

The multiplicity of occupations and the fragmentation of livelihoods have important spatial dimensions. Recent urbanisation, mostly towards the provincial capital, and short-term movements within the region, to neighbouring regions and, less frequently, to Tanzania are at the centre of the organisation of productive and reproductive activities. The recent urbanisation means that most residents of the main urban centre of Pemba describe themselves as being from rural areas and have family members living in the rural areas of origins. The fabric of social relations along family and kinship ties is a core component of social reproduction in this context, where public provisioning is extremely lacking. Thus, social obligations exist to pay visits to relatives and support each other in various ways, as I will illustrate below. Further, the ability to travel is essential to perform certain types of remunerated work such as petty trade.

Many women work as traders of agricultural produce, alcoholic drinks and a variety of other goods. Moving around is time-consuming as well as space- and time-specific. Traders need to travel to specific places – where they can acquire and sell the products – and at specific times – when it is possible to acquire good-quality products at convenient prices. In some cases, this may mean travelling in the very early hours of the morning and returning home in the evening or a few days later. Indeed trading is an onerous occupation performed almost exclusively by young women. Traders may walk when transport costs cannot be sustained, although some types of long-distance trading do require means of transport and only women who can afford these costs are able to take part. Thus, being able to work as a trader entails meeting certain conditions in terms of time availability, physical strength and/or money. In addition, traders either rely on elderly children, sisters or neighbours who carry out housework while they are away or are women who do not have young children, are divorced or separated and therefore less constrained by a range of typical household responsibilities.

These tensions are constitutive of everyday life and are visible through a social reproduction perspective. Another example is provided by casual wage work in agriculture, which is a last but important resort for the poorest households facing food and income shortages. Agricultural work is time- and energy-consuming and dictated by agricultural cycles. Long working hours and season-specific work intensity create dynamics of over-burden and exploitation that affect disproportionately poorer households and women, especially if they are the primary earners. Women need to reconcile long working hours with constant levels of care and housework and, given that casual wage work in agriculture is paid upon completion of a task, the longer it takes to complete the task the lower the wage for the amount of time spent. One outcome of these conflicting labour demands is that women perform casual wage work in agriculture more sporadically then men.

Obligations towards care provisioning are crucial. Visiting and providing help to ill and elderly relatives is an important social responsibility. These practices entail emotional and physical labour contributions towards housework and farming. They can take up weeks or even longer-term reconfigurations of living arrangements. A common practice is the relocation of older children to their grandparents’ households, as a form of labour and care provisioning. Children provide company, carry out housework and help with farming. Housework and care provisioning can be commodified when the exchange takes place outside family ties and with a wealthier household, who receives a child from a poorer one and takes responsibility for the child’s sustenance, which may include school costs, while asking the child to carry out domestic work.

The emotional as well as physical labour performed to assist a sick person is entrenched in gendered systems of responsibility that hold women accountable for caring for ill relatives of their family of origin and, in some cases, also for female relatives of the husband.⁸ This labour poses significant demands shown by the finding that the longest work interruptions collected during fieldwork were all attributed to the necessity to care for someone ill. For instance, Ernestina, who owns a small bar in the town of Pemba and has been running it every day for the past ten years, explained that a few years ago she had to close the bar for six months to be with the husband who was ill and needed care. Other stories document the necessity to abandon farming or reduce the size of cultivated land in order to be able to provide care to someone who is ill. The spatial fragmentation mentioned above is an important component of these processes as often care provisioning entails temporary relocation. Thus, it is crucial to note the importance of both spatial and temporal dimensions of social reproduction.

Finally, an important aspect that illuminates the importance of using a social reproduction perspective to understand work dynamics and working lives is the social obligation to make contributions to funerals, weddings and initiation rites (Hornby & Cousins, 2019; Stevano, 2018). These can be offers of food or cash to the family hosting the ceremony. This practice confirms the significance of Wolf’s notion of ‘ceremonial fund’ (cited in Cousins et al., 2018, p. 1065), which emphasises the centrality of the reproduction of the social, not only in addition to, as suggested in the original formulation, but as part and parcel of the reproduction of life and of capital-labour relations. The reproduction of the social is a specific dimension of social reproduction that refers to the practices that are critical to the reproduction of social and cultural relations in contexts where the welfare state is minimal or absent and therefore the maintenance of the structures of kinship and family is central to the cohesion of the social fabric and any form of safety net. Ceremonies lie at the foundation of social and lineage-based relations: economic contribution to ceremonies is an obligation and a sign of membership and affiliation as much as hosting a rich ceremony confers a higher socio-economic status. Often people look for piece-rate wage work if they need to attend a ceremony and lack the resources to make a meaningful contribution. The social significance of the reproduction is revealed by two empirical realities in Cabo Delgado: on the one hand, monetary and in-kind contributions to ceremonies emerged as one of the most common types of inter-household economic exchange and, on the other, the gendered practice of saving was often aimed at being able to make a contribution to initiation rites. Despite the heterogenous forms of gendered intra-household income management, women retain some level of autonomy in the use of their own earnings, which entails making savings where possible to make cash contributions in the occasion of initiation rites that, being organised along gender lines, do not typically entail resorting to men’s earnings. The reproduction of the social is crucial for the maintenance of social relations that ensure membership to social groups and their survival.

As the working poor are pushed to the margins of global production, the means to ensure their social reproduction are constantly eroded; yet, it is precisely the imperative of social reproduction that shapes their working lives and creates dynamics of differentiation among working women. Better means to ensure social reproduction, on grounds of socio-economic status, caring responsibilities, age and social relations, enable women to engage in remunerated work more steadily. However, it is critical to bear in mind that the possibilities for ‘better work’ are incredibly limited as the working poor, as a gendered and racialised group in the context of global neoliberal capitalism, are increasingly pushed into a condition of disposability and their exploitation materialises through their expulsion from formal wage work and their relegation to highly precarious work. The compulsion to find cash-earning activities while retaining some access to land creates also spatial dispersion that places the working poor in a circuit of precarious and exploitative dynamism, which stands in opposition to misleading ideas of poverty as associated with laziness, unemployment and time abundance. The everyday lives of the working poor are marked by constant movement, change and disruption and, while ‘small’ winners are produced, the conditions of exploitation of classes of working women, and of labour in general, rest on the reproduction of this precarious and exploitative dynamism. Whilst the small winners who can hire agricultural labourers and engage in more lucrative and steady cash-earning activities may improve some material conditions of everyday life, it is not evident that most of them have or will manage to free themselves from the precarity of life. What conditions enable this shift is a question for future research that needs to be addressed against the backdrop of the changing nature of surplus populations.

7. Conclusions

Drawing on empirical evidence on women’s work in localised labour markets in northern Mozambique, this paper put forward an integrated conceptual framework to understand working lives. By building on different bodies of literature with important elements of complementarity, showing how the IPE of everyday lens is essential to capture dynamics of production and reproduction as well as social differentiation among classes of working women in localised labour markets in peripheral economies, and how, vice versa, the feminist analysis of localised labour markets can add to literatures focused on export-oriented industries, global supply chains and global capitalism. The framework is based on an understanding of labour markets as bearers of gender, class, age, race inequality. Classes of labour are socially differentiated and internally fragmented, and a social reproduction lens helps us identify these patterns of differentiation and fragmentation both locally and globally.

The paper makes two key arguments. First, looking at women’s work in localised labour markets, as opposed to markets integrated in global production networks as in much feminist political economy literature, I interrogate how work in workplaces and contexts that are seemingly detached and marginalised from global dynamics of production is shaped by transformations in the global economy. Through assessing the conditions of exploitation in contemporary labour regimes against those prevalent during the colonial time, it is evident that the commodification of life, which accelerated during the colonial time, is at the basis of a long-term process that culminated in the fragmentation of labour and the erosion of means of social reproduction during the neoliberal era. From extraction through coerced labour during the colonial time, the contemporary dynamics of exploitation revolve around expulsion from formal wage work. Thus, the necessity to engage in multiple and highly precarious occupations in localised labour markets is itself part and parcel of global racial capitalism, which comes to shape the lives of those excluded from global production networks. Importantly, the gendered dimensions of processes of livelihood diversification entail the appropriation of women’s productive and reproductive labour.

Second, I question how classes of working women come to be differentiated. Women’s working lives are shaped by gender relations and social material practices that are foundational to the reproduction of the economic and the social. How work is approached, organised and experienced depends on the interconnections between the organisation of productive and reproductive activities, remunerated and unremunerated work. The imperatives of social reproduction are mutually constitutive and in tension with productive activities, as they determine gendered constraints to mobility and migration, work interruptions and obligations towards care provisioning and ceremonies underpinning the reproduction of social groups.

Working lives are structured around what can be termed as precarious and exploitative dynamism, which captures the necessity and compulsion to engage in multiple and precarious occupations that are framed by specific spatial and temporal dimensions. The fragmentation of labour and the means of social reproduction creates forms of spatial dispersion, with households and families stretched across space, and temporal tensions that have the overall effect of further complicating family-centred social reproduction and differentiating working lives. Temporal dynamics shape social reproduction through both historical processes and everyday life (on the relevance of time in social reproduction perspectives see also Bargawi et al. 2020 and Mezzadri and Majumder 2020).

Although the discussion of these dynamics has focused on a study of northern Mozambique, insights can be drawn to consider working lives of both women and men in all those peripheral areas of global capitalism where gendered and racialised peoples cobble together livelihoods at the margins of global production; they constitute the majority of the working poor and a better understanding of their everyday lives ought to be central to investigating the present and the future of neoliberal capitalism.

Acknowledgements

I am indebted to all the Mozambican people of Cabo Delgado who took part in this research, in particular the women who gifted me with their time, knowledge and experience. I think of them every day, especially now that they are living through a time of conflict in the region. This work would have not been possible without the assistance of Eusebio Tissa Cairo. I am thankful for the insightful exchange with my fellow guest editors of this special issue, Alessandra Mezzadri and Susan Newman, and all other contributors. I have also received useful comments by the audience of the Journal of Agrarian Change seminar, where I presented an early version of this paper in 2019.

Disclosure statement

No potential conflict of interest was reported by the author.

Additional information

Notes on contributors

Sara Stevano

Sara Stevano is a Lecturer in Economics at SOAS University of London; she works in the fields of feminist political economy and development economics.

Notes

1 Inequality is intentionally defined as co-constituted, rather than intersectional, to stress that different axes of inequality – gender, race, class, age, and so forth – are not independently constituted, but in fact co-determine inequality (Bannerji, 2011). This conceptualisation of inequality also suggests that an exclusive focus on gender inequality fails to capture the co-constituted nature of inequality.

2 Bernstein (2010) reminds us of the concept of commodification of subsistence, attributed to Robert Brenner, to indicate that farmers lose the ability to reproduce themselves outside of commodity and markets relations despite not having been necessarily dispossessed of their land.

3 Investment Promotion Centre (CPI) data published in 2011 and updated by the author in 2012.

4 Information on private companies and multinationals in Cabo Delgado are based on interviews with entrepreneurs or representatives of the company conducted by the author in 2012.

5 Mozambique could become one of the top exporters of gas in the world but there have been delays in the extraction of gas due to a host of reasons, including in finalising agreements between the government and investors, and now further delays are expected owing to the conditions of uncertainty and insecurity created by the ongoing insurgency.

6 The remaining three percent are respondents who answered that they do not know and one respondent saying the decision was with her brother.

7 Around 72 percent of the women interviewed receive a cash allowance to meet the household basic needs, most often from their husbands and, in a few cases, from their brothers. A more detailed discussion of the gendered system of cash allowances is provided in Stevano (2019).

8 This also illustrates partial shifts from the matrilineal systems characteristic of northern Mozambique, where women hold the greatest responsibilities towards their family or kinship, to patrilineal-like arrangements, where women are responsible towards the husband’s family.