Introduction
Job creation is a key focus of public policies in Mozambique (GoM 2015, 2020; MITESS 2016). The current Government Five-Year Program (PQG) 2020–2024 plans the creation of about three million new jobs by 2024. However, there is a contradiction between job creation and quality of employment in the prevailing productive structures in Mozambique. The extractive structure of the economy (Castel-Branco 2014) perpetuates the reliance on primary commodities for export (minerals–energy and agro-industrial), with minimal or no processing, and with weak linkages in the economy. The economic structure has failed to generate stable and regular work, income and dignified working conditions, and to ensure public provisioning of basic goods and services for most of the population.
By taking a political economy approach informed by a social reproduction lens, we look at the organisation of work in Mozambique’s agro-industry, and at its tensions and intersections beyond the productive sphere, under the contemporary regime of capital accumulation. A social reproduction approach illuminates the mutual constitution and tension between production and reproduction, where the latter captures practices that are necessary for the reproduction of life and capitalist relations (Katz 2001). This approach allows us to understand the organisation of wage work and to see its interconnections with multiple forms of work and how it is embedded in workers’ lives. We draw on a triangulation of qualitative and quantitative data. The analysis is based on original primary evidence collected in the cashew-processing and forest plantation agro-industries with workers and households in two provinces of Mozambique: in Gaza from 2018 to 2019, and in Niassa from 2014 to 2016. We complement this evidence with secondary data from official surveys, namely the 2014/2015 Household Budget Survey (HBS), and the 2007 and 2017 Population and Housing Census (henceforth, Census).
Two main blind spots underpin the focus on job creation and characterise the dominant understanding of labour markets in Mozambique. First, dualistic views of work and labour markets, premised on upholding productive/unproductive, paid/unpaid, formal/informal divides, conceal the interconnected nature of work in Mozambique. Second, and reinforced by the aforementioned dualisms, how employment interrelates with dynamics of social reproduction remains a neglected aspect of analysis and policy. These interdependences, however, are central amid low-wage casualised employment, high turnover of the labour force and poor public provisioning, of which rural and peri-urban agro-industry is a prime example. This flawed conceptualisation obscures the reality of work for most of the workers and leads to misguided policies.
We argue that the type of employment created in agro-industry reflects the prevailing mode of organisation of production, in which the private sector’s ability to make a profit is premised on its access to cheap and disposable labour. Informal, irregular and insecure work arrangements are a necessary condition of contemporary capitalism for those at the origin of global circuits of production. Workers and their households remain responsible for their own social reproduction, attained through diverse interdependent paid and unpaid work. Well-being is not limited to having a job and a wage; in fact, livelihoods and well-being may be endangered by having a low wage and/or having to do multiple precarious jobs. Unless the co-constitutive interrelations between production and reproduction are understood and addressed, the fragmentation of livelihoods will intensify the social reproduction crisis.
This paper is organised in six sections. The next section, ‘Work, accumulation, social (re)production and embedded contradictions’, looks at the relationship between work, accumulation and social (re)production. The following section, ‘Labour markets in the socioeconomic system of Mozambique’, explores the dynamics of labour markets, existing methodological gaps and challenges. The fourth section, ‘Organisation and social conditions of work in agro-industry in Mozambique’, analyses the organisation of work and quality of employment in agro-industry and their interdependent linkages with (re)productive work, and poses some questions for socioeconomic transformation in Mozambique. The last section concludes.
Work, accumulation, social (re)production and embedded contradictions
The labour force has historically been structurally integrated in the social system of accumulation, as evidenced across southern Africa (O’Laughlin 1981; Bernstein 2010). Although treated as two separate spheres, ‘when viewed, therefore, as a connected whole, and in the constant flux of its incessant renewal, every social process of production is at the same time a process of reproduction’ (Marx 1976, 711). However, the Marxian concept of social reproduction has been extended by feminists to capture the reproduction of human life (Laslett and Brenner 1989). Using a Marxist-feminist social reproduction perspective allows us to understand the social relations through which life and labour power are reproduced, and their interconnections and intersections within the global system of accumulation. The interdependence between productive and reproductive spheres is exposed in low-paid work – which forms the very basis of capital’s extraction of surplus value or the source of dominance – at the origins of global value chains. As Cousins et al. (2018) put it, in light of an extensive literature on post-colonial development in southern Africa, tensions between social reproduction and accumulation have long been noted in southern Africa.
A key contradiction, as highlighted by Fraser (2017) and further articulated by Rao (2021) in the global South, is that although social reproduction is necessary for capital accumulation, capitalism squeezes and destabilises social reproduction by devaluing it. In addition, patterns of exploitative relations are classed, gendered and racialised in ways that reflect colonial relations (Stevano, Ali, and Jamieson 2020). In a capitalist mode of production, workers are alienated; this is explained by Marx in his Economic and philosophic manuscripts of 1844 as losing control over processes of production and work, their autonomy and the product of their work which is appropriated by capital, through the power exercised in the accumulation process (Elster 1986). These patterns hold true for most differentiated workers in southern Africa, which we will illustrate using the example of Mozambique.
Mozambique’s extractive economic structure (Castel-Branco 2014) has failed to generate quality employment for the majority. The pattern of growth reproduces the reliance on export-oriented natural resources and agricultural commodities, with weak productive linkages and in the context of fragmented labour markets. The country’s economy was considered a success story amid high rates of economic growth, with an annual average of around 8% in the last decade and a half, with the exception of the last five years, when it was halved (with the lowest annual rate – about 2% – in 2019) (Castel-Branco 2017; INE 2020a). This structure has been sustained by the concentration of production and private investment in the extractive core of the economy and in the associated infrastructures and services. The extractive nature of the economy absorbs around 95% of total private investment and is marked by the specialisation of exports in a limited range of primary commodities of the extractive core: about 64% in mineral and energy resources (aluminium, electricity, coal, gas and heavy sands), about 15% in agro-industrial commodities (tobacco, sugar, wood, cotton, cashews and bananas), and about 21% in various other products (especially shrimp) (Castel-Branco 2017).
Despite high economic growth, poverty has remained high, with more than half of the population considered poor. Recent estimates show that the cost of a basic food basket may have increased by between 55% and 70% from 2014 to 2016, which is reflected in a national poverty rate of 55–60% – representing a salient increase from 46% estimated by the 2014/2015 HBS (Mambo et al. 2018; INE 2016). Furthermore, inequality in consumption has increased (the Gini coefficient increased from 0.42 in the 2008/09 HBS to 0.47 in the 2014/15 HBS) (INE 2016). Public provision of basic goods and services is limited in a context of commodification of public services, expropriation of resources and of the labour force, a limited tax base with tax exemptions granted to megaprojects, and several waves of expropriation of the state in various periods of its history through privatisation and public debt (Castel-Branco 2017). In addition to these, the country’s dependence on the export of primary commodities and the importation of basic goods and services, and its vulnerability to external crises or price fluctuations in primary commodities internationally, have an impact on investment, on income, on inflation of basic goods and services, and on interest rates.
The rural areas of Mozambique’s economy have been a strong foundation of capital accumulation, and labour power has been organically integrated in the process of accumulation (Castel-Branco 1994; O’Laughlin 1981; Wuyts 1978). In the colonial era, the capital accumulation system was based on the expropriation of the labour force and extraction of surplus value, through forced labour and the imposition of taxes (O’Laughlin 1981). The South of Mozambique was a reserve of cheap labour for the South Africa mines, while in the Centre and North regions the cheap peasantry labour was integrated in the plantation economy and as commodity producers, respectively (Ibid.). The extraction of surplus value through cheap and available labour was only possible through the maintenance of a land-based peasantry, which, at the same time, engaged in wage labour and produced food for own consumption and for the market. O’Laughlin (2002) explains that, since the mid 1980s, there has been no separation between household agricultural production and wage labour (locally or through migration): rather, members of rural households combine food production with various sources of income. There are many other ways of organising work, but there are very few peasants living exclusively from agricultural production.
Nevertheless, the interdependence between various forms of work has been neglected since the colonial period, owing to the persistence of an ideological preconception based on a dualist method of analysis of the nature of the economy and labour markets (GoM 2020; World Bank 2019; O’Laughlin 2017). This dominant dualism draws on the idea that the economy is separated into two sectors – one deemed capitalist/modern and the other pre-capitalist/traditional – and divides work into dichotomies, such as formal/informal, paid/unpaid, wage work/own-account work (INE 2019; World Bank 2012; Ali 2017; O’Laughlin 1996; MITESS 2016). As such, the idea that wage work belongs to the capitalist sphere and unpaid household work (e.g. peasant farming) belongs to the traditional sector persists. This perspective ignores important ‘informal’ forms of subordination of work to capital such as unpaid work and the structural interdependence between the ‘formal’ and ‘informal’ spheres.
This dualism is evidenced in Mozambique’s previous and current five-year government plan and employment policy (GoM 2020; 2015; MITESS 2016), which emphasise the creation of new jobs but fail to address the mechanisms through which quality employment can be generated. For instance, some priority measures in the ‘creation of new jobs’ pillar of the employment policy state that
it is intended to stimulate support for the informal economy, facilitating its transformation into a formal economy, thus promoting more decent and stable jobs … and to facilitate the transition of informal sector workers to the formal sector … through [for instance] facilitating greater access to finance for agriculture and livestock. (MITESS 2016, 26)
Labour markets in the socioeconomic system of Mozambique: beyond the dualism, dichotomies and invisibility of work
Data collection and mainstream analysis of labour markets in Mozambique are dominated by methodological gaps, limiting the understanding, design and effectiveness of public policies (Ali 2017; Cramer, Oya, and Sender 2008). A dualist view underlies official surveys and public policies, such as employment policy (MITESS 2016). As explained earlier, this approach pigeonholes heterogeneous forms of work into dichotomies such as formal/informal, paid/unpaid, wage work/own-account work. The interconnections are ignored by, for example, excluding unpaid work and neglecting casual forms of wage work.
Moreover, only one survey of the labour force has been conducted in Mozambique to present, Integrated Labour Force Survey (IFTRAB) 2004/2005. Official statistics on labour markets are collected through short modules on employment that do not capture information about: (i) the diversity of forms of recruitment and work, (ii) the range of occupations (mostly in rural areas) and their dependence on wage labour and vice versa, or (iii) the conditions of unemployment or occupational mobility of people (Ali 2017; Cramer, Oya, and Sender 2008). Despite these gaps, it is possible to pose some questions for a discussion on labour markets in Mozambique.
The demography is skewed towards the youth, with half of the population aged 17 years old or below and an average life expectancy of 54 years old. The total population increased by about 35% over the 10-year period 2007–17, and in 2017 it was around 28 million inhabitants (INE 2009, 2019). Half of the country’s population is of working age and three-fifths of this is classified as economically active population (EAP), of whom around half has no education at all (INE 2019).
Most of the population (nearly two-thirds) lives in rural areas; around two-thirds of the total EAP is classified by the 2017 Census as employed in agriculture, and nearly the same proportion of people classified their main activity as ‘peasant’. This is illustrated in Table 1, which also shows that in the past 10 years, the majority of the EAP continued to be involved in agricultural activity (in spite of a slight reduction), followed by the tertiary sector (which remains static), while the total EAP share in the manufacturing sector diminished.
Economically active population (EAP) by sector of activity in Mozambique, 2007 and 2017 (%).
| 2007 | 2017 | |
|---|---|---|
| Primary sector | 76 | 67 |
| Secondary sector | 6 | 5 |
| Tertiary sector | 13 | 13 |
| Other services | 5 | 7 |
| Unknown activities | 0 | 9 |
Nevertheless, the Census questionnaires focus on the main activity/occupation in the week of reference (that is, the last seven days), and the exclusive categories and lack of multiple-answer categories result in the neglect and underestimation of wage labour patterns mostly in casual and informal forms, particularly in agriculture (Ali 2017). In addition, 9% of the EAP are reported as ‘unknown’ – a category higher than the 5% of the EAP in the manufacturing sector – which clearly limits a wide analysis of labour market structures. For instance, one may wonder whether the ‘unknown’ category includes informal or casual wage work, in multiple physical or digitally mediated workplaces (which is an emerging pattern of work but not yet collected by official labour statistics).
Additionally, the 2017 Census shows that the majority of the EAP is involved in informal activities (around 88%), and that only 12% of the EAP has ‘formal’ wage work (Figure 1). Nevertheless, the methodological gaps and analytical preconceptions tend to confine wage work to formal arrangements, neglecting informal/temporary forms of wage work (e.g. agricultural and household employees) (Ali 2020a). Also, they neglect the interrelations between occupations (wage/unpaid/own-account work), which have historically been diversified/interdependent as working people have had to engage in multiple occupations to make a living (Ali and Stevano 2019; O’Laughlin 1981).
Economically active population in the labour process in Mozambique, 2017. Source: Ali (2020a; author’s estimates based on the 2017 Population and Housing Census).
A disaggregated analysis by gender based on the 2017 Census (Figure 2) reveals there are fewer women in wage work, relative to men, but more women are classified as ‘own-account workers’ and ‘unpaid household workers’. However, research deploying conceptual and methodological approaches equipped to capture irregular, informal and multiple forms of work has documented a complexity of women’s employment in Mozambique that is obscured by these statistics (Cramer, Oya, and Sender 2008; Oya and Sender 2009; Stevano 2019, 2021). At least two key findings emerge. First, rural labour markets not only exist but are heterogeneous and particularly important for the poorest, who often include women who are divorced, separated or widowed and therefore lack financial support through their families (Cramer, Oya, and Sender 2008; Oya and Sender 2009; Stevano 2019). Second, women engage in a multiplicity of occupations, often in addition to farming, which are integrated in processes of livelihood diversification that are gendered (Stevano 2021). Importantly, these studies provide insights on methodological approaches that can capture a more realistic picture of employment landscapes in Mozambique, through the use of mixed methods, particularly combining surveys with longer reference periods and qualitative interviews that can uncover longitudinal dimensions of work trajectories as well as time constraints.
Economically active population in the labour process by sex in Mozambique, 2017. Source: Ali (2020a; author’s estimates based on the 2017 Population and Housing Census).
Furthermore, the structure of the monthly minimum wages reveals a conflicting picture in terms of the legislative allocation as shown in Figure 3. Throughout the last decade, agriculture and fishery – the activities which according to the 2017 Census constitute most of the EAP (around 70%) – have been allocated the lowest monthly minimum wages, together with public administration, defence and security. Financial services, extractive industry, and electricity, gas and water are allocated the highest monthly minimum wages (MITESS Various; Hanlon 2018). Also, the sectors with the lowest minimum wage experienced a lower wage growth compared to those with the highest allocated minimum wages. Real wages estimated on the basis of food inflation dropped dramatically in 2015 and 2016 whereas food prices rose by 18% and 36%, respectively (Muianga et al. 2018). It is worth noting that in 2020 negotiations over minimum wages were suspended amid the Covid-19 pandemic. Moreover, there are some tensions with regard to the categorisation of activities and their respective allocation in terms of minimum wages. This is the case for the classification of agro-industry as opposed to the manufacturing industry, which is an object of contestation. Cashew-processing industries are not considered to be part of the manufacturing sector and, as such, the sectoral minimum wage is closer to minimum wages in agriculture than those in the manufacturing sector.
Monthly minimum wage by sector in Mozambique, 2010–20 (in MZN). Source: Authors’ estimates based on MITESS (various years) and Hanlon (2018).
The methodological gaps in labour market data collection limit the broader analysis of work, leading to misguided public policies. For instance, those without a formal job in rural areas tend to be classified as ‘peasants’/employed in ‘subsistence’ agriculture, ignoring these workers’ engagement with casual/informal employment (Ali 2020a). Besides, the heterogeneity of work(ers) is overlooked; one example is the treatment of the workforce in rural areas as homogeneous, as if merely dependent on a hoe, notwithstanding the increasing evidence of non-agricultural work even for those with access to land (O’Laughlin 2002). As such, heterogeneous workers who are ‘peasants’ in the deemed ‘traditional sector’ and simultaneously ‘wage workers’ in the deemed ‘capitalist sector’ are invisible (Ali 2020b). A social reproduction perspective enlightens the understanding of work realities, deconstructing these gaps and improving the quality of labour market data. A combination of improved official labour market statistics and case studies, informed by such a broader perspective of work, is necessary to design policies that can address the problems created by low-quality employment. For example, policies aimed at improving wages and working conditions as well as at strengthening the public provision of education, health care, social and child care and infrastructure are much needed. However, this is only a first, if necessary, step for a radical shift towards a politics that values social reproduction and promotes a redistribution of power from capital to labour. The peripheral nature of the Mozambican economy significantly constrains the policy space available to the national government, especially in the context of the late emergence of domestic capitalist classes whose interests are deeply enmeshed with the inflow of foreign capital (Castel-Branco 2014), and made starkly visible by the failure to protect the livelihoods of ‘essential’ workers during the Covid-19 pandemic (Stevano, Ali, and Jamieson 2020). Political change within the country needs to be accompanied by significant transformations in global governance.
Organisation and social conditions of work in agro-industry in Mozambique
The development of agro-industry and the concentration on production of primary commodities for export, through the acquisition of large amounts of resources (land and water) at low cost and access to cheap labour, are central aspects of the prevailing agro-industrial productive structures. Employment in agro-industry reflects the productive structures and is marked by fragmented social conditions of work and livelihoods, as will be demonstrated in this section.
Cashew is one of six industries – along with wood, tobacco, sugar, flour and prawns – accounting for a third of Mozambican exports; exports of these industries rose by 1205% between 1995 and 2011 (Sutton 2014). It is estimated that the sector employs around 13,000 workers, 65% of whom are women (Mozacaju 2017). The factories are mostly located in rural and peri-urban areas. With investments in large-scale, capital-intensive industries in the late colonial period, Mozambique attained the status of the world’s first producer, processor and exporter of cashews in the 1960s (Leite 1999). Soon after independence, the socialist government identified the cashew sector as key to the Mozambican economy and subsidised the processing industry by introducing a ban on the export of raw cashews in 1978 (Cramer 1999). During the civil war in the 1990s, cashew production, almost entirely in the hands of small-scale producers, declined substantially and the sector was liberalised. The export ban was replaced with an export tax, which was gradually reduced in the 1990s, while Incaju – the governmental institution that deals with the cashew sector – and the processing factories were privatised (McMillan, Rodrik, and Welch 2002). These reforms have been the object of intense scrutiny and debate as they led to a modest increase in producers’ prices but also to the complete collapse of the processing industry (Aksoy and Yagci 2012). Thus, the sector entered a phase of ‘rehabilitation’. Protests organised by the sectoral trade union, SINTIC, in 1999 led to an increase in the export tax and the introduction of a seasonal export ban in 2003. Meanwhile, new factories began to appear in the early 2000s. This new generation of factories is very different from the old one: these factories are labour-intensive, located in proximity to producing areas, and operate within a radically transformed cashew global value chain, which is dominated by new players, such as Vietnam and India (Mishra and Martin 2017). Nevertheless, in early 2021 the largest cashew processor and sector employer, OLAM, employing around 3000 workers, withdrew its activities from Mozambique, with a significant impact on workers in the sector (Carta de Moçambique 2021).
Forest plantations have been encouraged by the government but have faced different dynamics throughout history. In the colonial and post-independence period, the state sustained the planting of forests to reduce pressure on the native forest. In the early 1990s, some of the productive units were privatised, which led to a decline in plantation activity (MINAG 2015). From the mid 2000s, large-scale forest plantations of exotic/fast growing species (eucalyptus and pinus) were established, particularly in the central and northern regions of Mozambique, for the export of wood and its derivatives. Job creation was one of the promises made to justify the expropriation of land. Forward linkages are limited, with the exception of wages, and the existing backward linkages (e.g. sub-contracting companies to prepare the land, cutting trees and planting) are also restricted and not well established. In 2012–13, almost half of planted forests were concentrated in Niassa province, facilitated by Malonda Foundation (MINAG 2015). Moreover, from 2014, forestry companies entered a phase of crisis due to financial difficulties, limited prospects for expansion, and uncertainty over the installation of a factory to produce paper pulp. In an effort to cut labour costs, the companies implemented a significant reduction in labour recruitment, lowered working conditions, turned to outsourcing or sub-contracting of services, and dismissed workers. More than half of the forestry companies in Niassa disappeared. Currently, around half a dozen large-scale forestry companies are operating in the country. Green Resources and Portucel have the biggest land concession, although plantations cover only a small part of the concession area (Ali 2020b).
Our analysis of work in the cashew processing industry and in forest plantations is based on primary evidence collected between 2014 and 2019. Data collection in the cashew industry entailed three exploratory focus groups with workers, 30 individual semi-structured interviews with workers, observation of the organisation of production in the factory and a dozen interviews with other stakeholders, including managers, trade unions and public officials. In forest plantations, data were collected at (sub)national levels through around 80 semi-structured interviews with workers, two dozen interviews with their households, observations of organisation of production processes and local dynamics, and more than two dozen interviews with key stakeholders, including trade unions, managers, outsourcing agents, government, community leaders and peasants’/land/environmental (inter)national organisations.
System of work and employment conditions
Notwithstanding the clear differences in product and activity in the cashew processing factories and in the forest plantations, there are some relevant commonalities in the organisation of work, pay systems and working conditions. Both sectors organise production and work around a task-based system, which is the central mechanism underpinning work casualisation, intensification and absenteeism. Figure 4 summarises key issues of pay and working conditions characterising the experience of work(ers) reported in the interviews. The workforce is male-dominated in forest plantations and female-dominated in the cashew industry. In the forestry companies studied, women and men perform nearly the same type of work with the exception of some activities generally done by men such as cutting down trees, opening and clearing the fields; some men also perform types of work that are typically considered women’s work, such as in the nurseries. In the cashew processing companies, most of the productive processes are handled by women who predominate in the peeling and selection processes; men are mostly concentrated on machinery operations, storage and packaging. As Figure 4 shows, the type of employment created by the country’s agro-industries is mostly of low quality along three key dimensions: (i) irregular and unstable work; (ii) low wages; and (iii) insecure and precarious social conditions of work.
Key issues in the system and social conditions of work. Source: Authors, based on interviews with cashew processing and forest plantations workers, Gaza (2014–16) and Niassa (2018–19).
Irregular and unstable work
Agro-industry employment is mostly casual. The irregularity of work reflects the organisation of production and the type of investment made in its processes. Both agro-industries are labour-intensive, although some productive processes are (semi-)mechanised, and adopt a predominantly target-based system of production. The contractual system is imprecise and unclear. For instance, in the cashew processing industry, some factories provide a vague type of written contract classified as ‘uncertain time’ and ‘various services’, while other employers recruit the workforce informally, without written contracts. In fact, a mix of workers – with and without written contracts – may be found in the same factory. In forest plantations, most of the field workers do not possess written contracts and are outsourced, and those who do have temporary contracts (no longer than three months). Nevertheless, a written contract does not ensure stability of work or a guaranteed wage, owing to perverse mechanisms that are detailed in the next section. The weak contracts and/or their enforcement translate into a very fluid workforce, where the workers appearing on a company’s register do not necessarily correspond to those found at the workplace on a given day. This happens because workers may skip days of work, reflected in high levels of absenteeism especially in the cashew processing industry (Mishra and Martin 2017), and/or because other workers may be admitted to work on an informal basis.
Low wages
The employment structure is polarised, with a few deemed skilled workers who are relatively well paid and the majority deemed unskilled temporary workers with low wages. Most of the workers are paid below the statutory minimum wage (shown in Figures 3 and 4). This is a key area of discontent stressed by different workers (casual/permanent, female/male) in both agro-industries. The intensification of work, enacted through the imposition of very high production targets, is a great concern for both forest plantation workers and cashew processing workers who need to attain the daily production target to receive the minimum wage. For example, around 78% of the casual workers in a forestry company studied are paid below the sector minimum statutory wage. A range of activities (e.g. planting) that used to be carried out by more than one worker are now performed by a single worker. Furthermore, more than 90% of the interviewed workers reported concern over a lack of knowledge on how wages are calculated, as expressed by a forest worker:
I have to stay until it is dark in the forest … I couldn’t dig 300 holes with a pickaxe, and so I don’t know whether I will receive any wages, they mark me absent but I’m going to work and I do the same job … I would like to know how much I should receive for each hole I dig! (Interview with permanent forest worker A, male, Litunde, 11 December 2014)
One of the biggest problems is that you work for two days and end up being paid for only one, because you do not manage to meet the targets … many workers end up getting only MZN 3000–3500 per month … [but the contract states MZN 4300]. If a worker makes a complaint, they may be sacked. (Interview with cashew worker B, female, 1 May 2019)
Insecure and precarious social conditions of work
The organisation of agro-industrial production does not provide workers with the social conditions for a dignified life. First, work hours are frequently very long. Most workers reported pressure on time available to do complementary activities, although women reported more pressure on time to do housework than the men. For instance, in some cashew industries, the workers who are paid in a task-based system, mostly women, start working at 6–6.30am but the end of the working day is uncertain, depending on how quickly they can finish their daily workload. Their working day can last up to 10–12 hours, although the contract states 8 hours a day. This is a concern for most workers as, besides tiredness, it creates difficulties in returning home when it is already dark and transport is less frequent and not affordable for many. Notably, due to the gendered composition of the workforce and division of labour, there are specific barriers for female workers, whose time to take care of children and to do domestic work is significantly squeezed.
Second, equipment and tools are inadequate. In the cashew industries, due to the corrosive substance contained in the shell of the cashew nut, it is essential that workers who deal with the shell (e.g. de-shelling, scooping) wear gloves. Workers reported employers’ failure to replace the gloves (which last for 4–5 days) when needed. In addition, the uniforms should protect workers from the hazards of their job, but in some cases this does not happen, and some workers are left with wrists and ankles exposed to contact with the corrosive substance, which produces burns. Moreover, in forest plantations, workers reported problems with tools and extra effort needed – for example:
This work here is very tough … the wages aren’t enough for anything … it’s difficult to meet the target … it’s hard to use a pickaxe … . This is suffering, it’s not a job, it even seems like Xibalo! [the term for forced labour under colonial rule] … . When I tried to complain about the pickaxe, boss X replied, ‘I don’t talk with raw material’ … . There are others who also don’t like the conditions, but because of life’s suffering, we have to accept. (Interview with casual worker C, male, Chimbonila, 13 December 2014 [also cited in Ali 2020b, 592])
Fourth, there is a general lack of clarity over social security entitlements. However, in both of the agro-industries analysed, while most of the workers have their wages deducted monthly to make contributions to the National Institute of Social Security (INSS), the majority of them do not have access to any social benefits, or at least they are not clear on the reasons why their salaries are subject to this deduction. This is the case for both females and males working in the forest plantation fields with temporary contracts, and those in the cashew-processing factories with ‘uncertain time’ contracts. These workers are unclear on how benefits are calculated in their contractual system.
Fifth, workers’ organisation and collective bargaining are fragmented. Socially differentiated agro-industrial workers have weak bargaining power. Relations between workers and unions are weak; some workers do not even know their union representative. Casualisation and high turnover limit the organisation of heterogeneous groups of workers, although such organisation is central to the possibilities of negotiation. Also, trade unions are confined to labour issues within the point of production/conventional ‘productive unit’ or the struggles over the nominal minimum wage, neglecting the wider organisation and social conditions of work and reproduction outside the ‘production unit’ which are inter-related with/dependent on these spheres, as discussed in the following subsection.
Finally, transforming agro-industrial productive structures to change the low-quality employment for stable, secure and dignified jobs is crucial to enhance living conditions through employment creation and to enable some channels for wealth redistribution. The following measures are imperative: ensuring that an effective day of work is registered and actually paid; establishing improved work conditions (e.g. reducing long working hours, strengthening contractual systems, providing appropriate equipment and tools, and doing so promptly, ensuring health and safety and promoting a dignified work environment); matching social security contributions with benefits; and enhancing workers’ representation and collective organising.
The interdependence between productive and reproductive work and its intersections
The irregularity, instability and poor working conditions push workers to engage in complementary types of work, both paid and unpaid, on-farm and off-farm. This practice enables the sustenance of the labour force and its availability at a low cost for capital. Socially differentiated workers in the cashew-processing industries and forest plantations reported pursuing, heterogeneously, other forms of work such as biscate (casual piece-rate wage work), negócio (‘business’, cash-earning activity), and farming for own consumption or sale.
The interdependent nature of various forms of work, savings practices and debt management, in which differentiated wage work in agro-industry is embedded, is outlined in Figure 5 (Ali and Stevano 2019). From these inter-relations, two important perceptions emerge: (i) wage work in agro-industry cannot be understood in isolation from other types of work and money flows; and (ii) a crisis in one of these spheres triggers effects on others, impacting individual and household well-being. An example is provided by the crises that unfolded in the early years after Mozambique’s independence (1975–77), when the crisis of wage labour led to a crisis in family food production which, in turn, exacerbated the crisis in wage work through the weak purchasing power of salaried workers (O’Laughlin 1981).
Interdependence of wage and reproductive work through money flows. Source: Ali and Stevano (2019), based on interviews with forest plantations and cashew processing factories workers in Mozambique.
Wage work and unpaid work (including household agricultural production and domestic work) continue to be interdependent, as they ensure cheap labour. Furthermore, income from agro-industry wage work may enable workers to: (i) acquire wage goods; (ii) have an investment base (in alternative activities, including their own farm); (iii) partly release food production to the market and (iv) respond to shocks. Family production subsidises the low wages by feeding and nurturing the workforce, especially in periods of employment shortages. Similarly, wages allow for the funding of small businesses, which in turn subsidise low and irregular pay. Informal savings groups, such as Xitique, provide a social safety net in case of unexpected events (such as accidents, fires, deaths or floods).1
A striking aspect is that many workers interviewed in both sectors, especially those with temporary contracts, pointed out that, given the low real wages, they resort to debt or borrowing money from neighbours, friends, colleagues, local shops and their employer to buy basic goods and services. Some workers appear to be trapped in debt cycles, and the wage earned serves the primary purpose of repaying the debts accumulated previously. Some forest plantation field workers in Naconda and in Mussa (Niassa) expressed frustration and preoccupation at the necessity to repay their debts: ‘we are working to pay debts’2 and ‘I am tired of working to pay debts.’3 Similarly, a worker in the cashew-processing factory with an ‘uncertain time’ contract explained that, in the month before the interview, she had to buy two cooking pans, clothing for her toddler and a capulana (garment used by women) on credit and then used her salary to repay the debts incurred.4 Other workers who said that they did not have any debt were fearful of accumulating debt as they did not know if they would be able to repay it. These debt dynamics compound and are compounded by the lack of stable jobs and high living costs.
Moreover, low-paid and long work hours amid weak public provision of basic goods and services give rise to resistance mechanisms such as absenteeism (which is high in many agro-industries) or sabotage of production to meet targets. This has implications for productivity and quality of production. For example, in forest plantations, several workers face an overlap between wage work and household agricultural production, especially in the rainy season, which is the peak period in forest plantations and also the time when seeds are released on workers’ fields. Workers’ lack of time is exacerbated by insufficient income for subcontracting other day-workers for their fields.
Furthermore, there is another tension between jobs created and the destruction of livelihoods. For example, some areas of household agricultural production in Niassa were granted to agro-industries but, on the one hand, without providing stable and secure employment and, on the other hand, having destroyed the existing food production base, complementary income and consumption. Expropriation of land and (under)unemployment inhibit workers and their families from developing complementary activities to compensate for precarious working conditions, impacting on the social reproduction of labour.
Thus, the interconnected nature of work and family-centred social reproduction in the context of precarious livelihoods and minimal public provisioning make the separation between production and reproduction almost invisible. Crucially, the imperatives of social reproduction shape the rhythms of employment in agro-industry and the broad organisation of work, and vice versa.
(High costs of) social reproduction of labour and socioeconomic transformation: some questions and policy implications
Employment conditions are part and parcel of the broader patterns of Mozambique’s mode of production. Capital still has low-cost access/control over resources and a semi-proletarianised, low-paid, disposable, poorly organised labour force with low bargaining power.
The inability of the economy to provide quality employment and affordable basic needs, indispensable for the reproduction of human life and the workforce, aggravates the precarity of work and the fragmentation of livelihoods. The Covid-19 pandemic has been exacerbating the existing crises of the organisation of work and un/underemployment embedded in the structural crisis of capitalism. Unemployment has been on the rise: around 75% of agricultural companies were affected and around 10% of them did not pay full wages to their workers and expected to reduce their number of workers by around 27% (INE 2020b).
Redistributive responses are necessary and socioeconomic transformation is crucial to reduce the high costs of social reproduction endured by workers and to revert Mozambique’s residual position in global value chains, which perpetuates social relations of exploitation, alienation and expropriation. The provision of basic consumer goods and services is central, especially food, of quality and at low cost, which can increase real wages in the society while creating a workforce that is competitive and with better living conditions. This implies, at least, rethinking two interconnected components. The first is the type of economy, re-evaluating what the economy produces, how and for what; who produces; who benefits; and how the value produced is distributed. The second is the provision system, which transcends the production system and involves rethinking circulation, access, and productive social relations. Besides the provision of public services, it is necessary to ensure pension payments, unemployment benefits, contributory and non-contributory disability benefits, and healthcare social protection benefits (e.g. paid sick leave), complemented with a universal basic income for socially differentiated groups (redistributive options are discussed further by Ghosh 2019; Castel-Branco 2020; Standing 2020). In other words, it is necessary to act to address the ongoing crisis of social reproduction in Mozambique.
Conclusions
Following a political economy approach informed by a social reproduction lens, we analysed the contradictions between job creation and employment quality in the social system of accumulation in Mozambique. The prevailing extractive economic structure has mostly failed to generate stable work, pay and social conditions for a dignified life. A social reproduction approach allows us to see the interconnections and interdependence between wage and unwaged work that are particularly important in the context of low-wage casualised employment, high turnover of the labour force and poor public provisioning, of which the rural and peri-urban agro-industry is a prime example. These patterns and dynamics are neglected in the mainstream view of labour markets, which is dominated by dualisms and is limited by its blind spot on social reproduction. Unpaid work is excluded from this dominant view, despite its historically central role in social reproduction, subsidising low-paid work, mostly in informal/casual settings.
We argue that there is a contradiction between job creation and employment quality in the social system of accumulation in Mozambique, as evidenced in the social conditions of work in agro-industry. The type of employment reflects the prevailing mode of organisation of production and work, in which the companies’ profitability depends on available, disposable, low-paid and insecure work. The availability of labour power is overall reliant on workers’ ability to sustain their social reproduction through a range of interlinked, and mostly precarious, paid and unpaid work.
The (re)organisation of work and collective bargaining for better working conditions in the country rely on the conflictual relationship between the workers, state and capital. Socioeconomic transformation, in particular addressing the elements of that conflictual relationship, is critical to redistribute the burden of the high social reproduction costs endured by workers.