The Chókwè irrigation scheme, covering approximately 30,000 hectares of land and cultivated by more than 11,000 farmers2, is the largest area of irrigated land in Mozambique. Located in the southern part of the country, within the Limpopo river basin, the scheme and its organisation were an important case study for researchers investigating the consequences of the ‘socialisation of the countryside’ and Frelimo's agricultural policy after independence (see Wardman 1985, Hermele 1988, Roesch 1988, Bowen 1989) and the effects of divestiture of the state farm structures after the signing of the peace agreement (West and Myers 1996). Afterwards, because of poor agricultural performance exacerbated by the very poor state of the irrigation infrastructure, the scheme had almost been forgotten. Since the beginning of 2007, however, the Chókwè irrigation scheme has forcefully re-emerged in Mozambican public debate.
The renewed interest in the country's biggest irrigation scheme stems from the ‘green revolution strategy’, the Mozambican government's current agricultural policy that aims to ‘increase agricultural production and productivity of small farmers’ (Republic of Mozambique [RoM] 2007, p. 7) and to eliminate the need to import rice and potatoes. The foundations of the ‘green revolution strategy’ are to be found in the second Action Plan for the Reduction of Absolute Poverty (Plano de Acção para a Redução da Pobreza Absoluta 2006–2009 [PARPA II]), aimed at ‘achieving higher productivity and intensifying the vertical and horizontal links within the rural and national agricultural economy’ (RoM 2006, p. 132).
This policy has major implications for an aspect that makes the Chókwè irrigation scheme an unusual case study: the high percentage of women in the family sector holding an irrigated parcel in their own name. In the whole family sector – which covers 44% of land and includes 92% of the farmers in the scheme – women are slightly over 30%, and over 80% among the former Farmers’ Cooperatives (known these days as Farmers’ Associations) that were created after Independence. The high number of registered users that are women is a consequence of the historical, political and social dynamics of the area that have pushed the migration of men towards South Africa and that have thus shaped the patterns of gendered access to family plots within the irrigation scheme.3 In the following part I will briefly analyse these dynamics, and the factors that brought about what O'Laughlin (2009) has recently defined as ‘relatively secure’ access to land, before discussing in more detail the implications of the ‘green revolution strategy’ with respect to women's access to and use of land.
The historical dynamics of women's access to irrigated plots
The Mozambican peasant farmers that used to occupy the area when the irrigation infrastructure was built had their land allocated according to customary rules. The construction of the irrigation scheme and the establishment of the Colonato do Limpopo caused a process of land expropriation that affected about 2000 households. Only a small number of them, linked to customary structures, received two hectares of land back, on probation, when irrigated plots were distributed in 1954, while the others had to move to dryland areas. By the end of the 1950s, economic, social and management problems started to undermine the productivity of the scheme, and over one-third of the Portuguese colonos had already left the scheme before 1975, when Mozambique became independent (Hermele 1988, p. 43).
The existing patterns of male migration towards South African mines had created instability in marital and family relations that came to be closely linked with the ongoing processes of social differentiation and the nature of women's work. Remittances from migrant labour were crucial for women, in charge of household agriculture, to rent a plough or to hire seasonal workers for planting or harvesting. Other women continued to ‘wield the hoe’ in their plot or were employed as casual labourers on plantations (O'Laughlin 1995).
As soon as Portuguese settlers left Mozambique, local farmers occupied irrigated lands: Frelimo had yet to decide how to manage the vast irrigation scheme. In 1977, the Third Frelimo Congress set guidelines for the agricultural development of the country that had to be carried out by mobilising people in state farms and cooperatives, and creating communal villages. The Limpopo Valley, with its major irrigation scheme, was designated ‘breadbasket of the nation’. In Chókwè, those who had occupied the irrigated plots were resettled into communal villages and organised in farmers’ cooperatives, and the whole irrigation scheme was put under the control of a state farm, the Limpopo Agro-Industrial Complex (CAIL).
At Independence, 30% of households were headed de jure by women, and but many more women were de facto heads of household. This was reflected in the composition of cooperatives in Chókwè, where two-thirds of the members were women. At that time wages from migrant labour were steadily declining, due to radical change in South African recruitment policy: from 1976, the number of Mozambican miners had decreased by 70% (Newitt 2002, p. 205). This had a marked gendered impact in terms of the division of productive work, with men searching for employment on the state farms and women relegated mainly to subsistence or cooperative production, or recruited as seasonal casual workers.
The Fourth Frelimo Congress of 1983 agreed to halt the expansion of the state farm sector while recognising the relevance of the cooperative and family sector, as compared to large-scale and centralised agricultural development projects that were not achieving the expected results in terms of economic development. In Chókwè, the CAIL was divided into smaller state farms, and part of the land was redistributed among members of the cooperatives for individual farming. The remaining irrigated land was further divided among the family, the ‘private’ sector (i.e. including better-off farmers that owned agricultural inputs) and the entrepreneurial sector – private companies and two joint ventures (Pitcher 2002, p. 110). The divestiture of the state farms was completed after 1986, when a programme of economic rehabilitation was agreed between the Mozambican government, the World Bank and the International Monetary Fund. The land was then further redistributed among individual farmers. Whereas many women members of farmers’ cooperatives were allocated a small irrigated plot in the family sector, they were completely excluded by the category of ‘private’ farmers, who received larger plots closer to the main canals, as they were deemed to be ‘productive farmers’ – which, according to West and Myers (1996, p. 43), related more to wealth, status, local connections and kinship relations than to equipment or specific expertise. Women were thus admitted as individual farmers within the irrigation scheme because of the local patterns of gendered division of labour, but not because they were considered ‘serious actors in the irrigation context’ (Zwarteveen 1997, p. 1346). While the fact that many women were admitted to the scheme as individual farmers – and later registered in their own name as water users – should not be underestimated, their relative security of access is due to their being ‘productive’ and paying the water fee, as further developments in the irrigation scheme show.
The Chókwè irrigation scheme after 1997
The year 1997 seemed to be a landmark for the development of the irrigation scheme, with the establishment of a new management body, the parastatal enterprise Hídraulica de Chókwè-Empresa Publica (HICEP), and of Water Users’ Associations (WUAs) that were charged with providing technical support to HICEP in maintenance activities and with representing all farmers, now officially called ‘water users’ (WUs). The irrigated plots remained divided into three sectors: family (between 0.25 and 3.9 ha. of land), private (4 to 500 ha.) and entrepreneurial (over 500 ha.), with the family sector covering approximately the 44% of land and 92% of the WUs.
According to the HICEP statute, approved in 2002, water users (WUs) are entitled to receive water in a quantity proportional to the size of their plot, and to be represented by their respective Water Users’ Association (WUA). The right to water for irrigation can be suspended if a WU does not pay the water fee. WUAs regulate, manage and keep the accounts of water distribution within the area assigned to them, ensure that infrastructure (secondary canals and ditches) and equipment are properly maintained, represent their members and take part in the administration of the irrigation system. HICEP, for its part, administers the irrigation scheme, regulates and manages the water supply, and is responsible for the maintenance of the main canals as well as for technical support to WUAs.
As concerns land rights, Article 12 of the HICEP statute establishes that the irrigation scheme follows the regulations set out in the 1997 Land Law. Two types of land right apply to WUs: those entering the scheme for the first time are issued a state-granted land title (Título de Uso e Aproveitamento da Terra) once they are approved by their WUA. According to the Land Law, this kind of land title is subject to a legally mandated community consultation process and to the discretion of the state. In the case of this state-owned irrigation scheme, WUAs act as ‘the community’. Non-Mozambican citizens and corporate bodies are only allowed to obtain this type of land title, and their application must include a development plan to be implemented within two years. On the other hand, Mozambican men and women who have occupied irrigated land in good faith for at least ten years are awarded an inheritable land title. Within the scheme administration, both titles can be revoked for abuse of rights, for lack of maintenance of irrigation canals or for not paying taxes. The statute further specifies that land rights within the irrigation scheme should aim to intensify use of land, and therefore WUs are required to maximise use of their plot, provided that water is available. Moreover, land rights are inseparable from rights and duties related to the use of water.
The establishment of the new management structure and of WUAs was accompanied by the emergence of an enduring conflict between them and the Farmers’ Associations. HICEP, in fact, suggested to the Farmers’ Associations (affiliated under the Chókwè Union of Farmers’ Associations) that they stop the associations in order to enable their members to join the respective WUA as individual WUs. The Union strongly disagreed and is currently engaged in a process of establishing a legal persona for its Farmers’ Associations, so that they can join their respective WUAs as a group rather than as individual members. This would enable them to maintain a better bargaining position vis-à-vis HICEP and to advance the water fee for members in temporary difficulty. Indeed, this is often the case for women members of the Farmers’ Associations who – especially when heads of household – are most likely to use the irrigated plots for subsistence rather than for marketing. The recent policy focus on increasing the commercial productivity of the irrigation scheme puts many of these women farmers – especially those who are heads of household and those without alternative or differentiated livelihood strategies – at real risk of being evicted.
Mainstream and conflicting visions over the Chókwè irrigation scheme
The ‘green revolution strategy’ has its foundations in the second Mozambican PARPA: it is aimed at assisting both ‘small family farms during their gradual transition to commercial operation’ and ‘commercial farmers, encouraging them to boost their production, productivity, and competitiveness’ (RoM 2006, p. 129). The development of public–private partnerships is seen as a key strategy to effect this structural transformation in the agricultural sector (ibid., p. 130). While the role of irrigated agriculture is only briefly mentioned in the PARPA, in June 2006 the National Directorate of Water released a proposal for a new National Water Policy (NWP) aiming at: the efficient use of water for economic development; water pricing; direct participation of the stakeholders at basin-level; enhancement of the role of the private sector (RoM 2006a, Art. 2.2–3). As regards irrigation, the policies envisaged aim to ‘promot[e] the full utilisation of those irrigation infrastructures already existing’ including through the promotion of private investment and public–private partnerships. The sustainability of operating, maintenance and management costs of irrigation schemes is guaranteed by the imposition of a water tariff (ibid, Art. 4.2).
The mantras of the international agenda of water management – cost recovery, decentralisation of water management, setting up associations for water users, promotion of commercial agriculture and increased participation of the private sector – resonate in the above documents and fit into the local dimension of a rooted narrative that sees subsistence farmers transformed into commercial farmers as the best answer to the problems of rural poverty, underdevelopment and food insecurity. Irrigation infrastructure is considered an added value to achieve objectives of increased production and must, therefore, be properly and efficiently managed. Efficient management would necessarily include cost recovery through the imposition of water fees, a reduced role for the state, the involvement of the private sector and, an increased role for local users. This, it is claimed, will lead not only to increased production and productivity, but also to a more sustainable use of water that, as a scarce and economically valuable resource, must not be wasted in inefficient subsistence farming.
These measures are currently being implemented in the Chókwè irrigation scheme, and the changes have crucial implications both in terms of gender equality and of dynamics of social differentiation and poverty reduction. Irrigated farming is seen as having a significant potential to enhance the productivity of small farmers, thus reducing rural poverty. However, the fact that the above actions do not take into consideration gender inequalities in accessing and using irrigated land makes one wonder who will benefit from these dynamics of change and who, on the contrary, will be marginalised. Gender mainstreaming, in the key policy documents that are shaping the current state of the Chókwè irrigation scheme, has resulted only in a rather rhetorical acknowledgement of gender inequalities in Mozambican society and in rural contexts in general.
The main concerns of HICEP managers are the completion of the infrastructural rehabilitation, the establishment of joint ventures between WUAs and private investors, and the economic sustainability of the irrigation scheme, which would necessarily imply an increase of the water fee. The fact that this will result in increasing difficulties and a real risk for subsistence farmers of being evicted from their parcel – as the statute sets out for non-payment of the water fee – appears to be of marginal concern for HICEP, who considers that subsistence farmers in the scheme are essentially wasting water, and is planning to have all irrigated parcels allocated to those who can farm ‘efficiently’. There is no mandatory provision in this sense, but the envisaged increase of the water fee will push many small farmers – and particularly women – out of irrigated farming and back to rain-fed plots, and in search of casual jobs in a very insecure regional labour market (O'Laughlin 2009, p. 19). At the same time, powerful actors are entering the irrigation scheme, as HICEP is encouraging private investors to establish partnerships or joint ventures with the WUAs, granting them huge plots of irrigated land. However, private investors are not new to the irrigation scheme: they are only one of the many types of actor with different and competing interests that, over the years, have shaped a historically, politically and socially complex environment.
Reviving the ‘breadbasket of the country’?
Part and parcel of the government discourse about rice production in the Chókwè irrigation scheme is a recall to the Frelimo's almost mythical project of transforming the Limpopo Valley into the ‘breadbasket of the country’. The 2008–11 Food Production Action Plan estimates that the deficit in rice production is around 300,000 tonnes, and the government expects the Chókwè irrigation scheme to contribute 35,000 tonnes to national supply in 2009–10.4 Indeed, HICEP expects farmers in the irrigation scheme to grow rice on their plots, rather than ‘unprofitable’ subsistence crops. However, a brief analysis of the specific patterns of production in the family sector throws some light on the risk that some women farmers are facing, in a situation exacerbated by the increasing competition for casual jobs in the agricultural sector resulting from substantial return migration of males from South Africa (O'Laughlin 2009, p. 19) as a consequence of the economic crisis.
During a series of interviews conducted in January 2008 with three male and 17 female members of a few Farmers’ Associations, supported by a Spanish NGO-funded family farming support project, several interesting factors emerged. While the three men were all married (one of them with two wives), only six of the women were married (one the first wife of a polygamous man), while eight were widowed, and three were single mothers. Their ages ranged from 32 (a single mother) to 63 (a widow), while the three men were 49, 54 and 60 years old.
With respect to land holdings, five out of 20 were farming a plot that was not registered in their name. One woman, for example, had been working the plot of another who died in 2004. She previously had a plot within the same association, but in 1991 she was forced to abandon it because of ‘difficulties’. In 2005 she went back to the association asking for another plot, and she was waiting for it to be registered in her name in the HICEP land register. Her agricultural production was exclusively subsistence-related, and had little growing on it compared to other farmers. This was because her plot was not receiving water due because of a soil problem that was later corrected. At the time of the interview she had just started to irrigate properly, and did not know how much the water fee was, since she had never been asked to pay it. Plot sizes vary from 0.25 ha. to 2 ha., the average being 0.9 ha. It is interesting to note that the three farmers who have a 2-ha. plot (two women and one man) are long-serving presidents of a Farmers’ Association, and that they were granted their individual plots between 1983 and 1987.
As regards crop production, 14 of the farmers were cultivating rice (the three men, and 11 of the women). Of the six women who were not, four were widows and two were single mothers. They mainly produced maize, beans and sweet potatoes for subsistence, and only one sold part of her maize production, from home rather than at market. With only one exception, all the farmers had alternate sources of income, coming mainly from casual labour on other plots, but also from selling charcoal, brick production and laundering clothes. Among the rice producers, the situation was very varied. One man had been able to harvest six tonnes of rice on 1.5 hectares of land in the last season, but most farmers grew rice on only half a hectare of land, even when they had a bigger plot, and average production ranged from approximately 1.5 to 3.5 tonnes of rice. The limited options for selling the rice were considered a major constraint: the majority of farmers were selling the rough rice to Orízicola Inácio de Sousa, a rice-processing factory located in Palmeira, in the district of Manhiça, Maputo province. They received 4.1 to 4.5 meticais (MTn) per kg of rice, from which transport costs, calculated at 1 MTn per kg, had to be deducted. Only three women did not sell their rice to the factory: one claimed that the amount of money received was not worth all the work she did in her field, therefore she was cleaning her own rice and selling it from home, earning up to 15 MTn per kg. Another woman sold rough rice from home, gaining approximately 100 MTn per 15 kg. The third used to sell rice to the factory, but as her production was decreasing due to the poor conditions of her plot, she kept it for family consumption: ‘I am not going to spend all my money on a 25kg bag of rice in the shop!’ she claimed.
All the farmers grew maize and beans, and many also grew vegetables. Some successful rice producers (four) had decided to limit their crops to rice in the warm season and maize and beans in the cold season. Two of them, however, had access to another plot in the dryland where they grew vegetables that were either sold at the local market or used for family consumption.
These patterns of production show that the small farmers in the Farmers’ Associations are trying to comply with HICEP requests to cultivate rice – mainly thanks to the NGO support and because ‘they want to show to the government that small farmers also work hard and can be efficient.’ There is, however, a series of constraints that must be considered. The main problem for these farmers is the marketing of their produce; in fact, having refused to engage in contract farming arrangements with a private company present in the scheme (because it was not considered profitable), their options have remained limited: either they send the rough rice to the processing factory, with a limited profit, or they sell it from home. The woman selling cleaned rice from home is married and has no dependants in her household: her only son lives and works in South Africa. Given that her husband also takes part in the farming work, she has time to clean rice, a time-consuming operation that in any case she prefers to the alternatives of selling in Chókwè market (‘it's a waste of time’) or selling to the factory (‘it's not profitable!’). But for single mothers and widowed or divorced women, rice cultivation is highly demanding in terms of work and time spent in the field, and limits options for accessing other sources of income, which are mainly casual labour on other plots, but also include charcoal selling, brick production and laundering clothes. Constrained by lack of time, money and inputs, these women are not only lagging behind on rice cultivation, but are clearly also facing the risk of eviction once the water fee is increased.
Another important concern is the lack of access to inputs, fertilisers and pesticides. Even though the farmers are entitled to use tools and other inputs from the Associations’ premises, these are not always easily available. With limited access to other forms of stable income, the money they invest in their farming activity is used mainly for land preparation – and for paying the water fee.
A third factor involved the option of employing casual labourers to help with particular activities (mostly rice transplanting). It is interesting to note that five out of the six women not growing rice were only farming for subsistence. The sixth sold only part of her produce from home, depending on the season and on the supplies that she had. They claimed that being by themselves (four of them are widowed, and two single mothers), they do not have enough time to engage with more profitable farming activities, and that they would like to employ some casual labour, but had no money to do so. All the other farmers, more or less regularly, employ other people that are paid 30 to 50 MTn per day. Traditional systems of mutual help or solidarity are not common, except for the amount of work that everyone is supposed to do in the communal field of the cooperative: ‘Money is important. Those who work, want money.’
The Chókwè irrigation scheme has an unusual history that was shaped by male migration to South Africa and resulted in increased options for women to have primary access to irrigated plots, and over the years, women have enjoyed a certain degree of security with respect to access to and use of irrigated land. The implementation of the government ‘green revolution strategy’ – including the HICEP plan to make the scheme economically sustainable by increasing the water fee, the concept of the ‘efficient farmer’ backed by the international agenda of water management, and the increasing insecurity of the regional labour market – will severely undermine this ‘relative security’, given the fact that there is no clear-cut gender policy within the scheme, nor indeed within the key policy documents briefly presented above, that addresses the specific needs and interests of women regarding irrigated land and taking into accounts the problems they face. It appears that expecting small farmers, and in particular women, to be able to compete in the globalised rice market is unrealistic, and can perpetuate unequal gendered access to resources such as land and water.