Social Differentiation and the Politics of Land: Sugar Cane Outgrowing in Kilombero, Tanzania

The privatisation of formerly state-owned sugar cane estates and mills in the late 1990s led to the reconfiguration of the ownership structure and production of the Tanzanian sugar industry. This included the decentralisation of sugar cane production operations to outgrowers and their associations. Based on an investigation of sugar cane outgrowers of the Kilombero Sugar Company, this article explores the outcomes of this transformation and the dynamics of social differentiation, and the ways in which this is shaped by Tanzania’s institutional and legal framework governing the sugar industry. While increased sugar cane production has created jobs and income for some outgrower households, it has also been associated with several adverse impacts. Large-scale outgrowers have rapidly captured the most lucrative business opportunities and the land they require, marginalising smaller outgrowers. This is further exacerbated by heightened competition among farmers and patronage relations affecting the distribution of harvest quotas and cheap sugar imports. The outgrower model is central to national development initiatives such as the Southern Agriculture Growth Corridor of Tanzania and ‘Big Results Now’. Despite the grand ambitions of large-scale agricultural commercialisation, who wins and who loses out depends more on the local political economy, where the sugar industry, local business, political elites and local communities compete for the benefits of expanded sugar production.


Introduction
Outgrowing is a type of contract farming 1 with a long and complex history in Africa, but has recently emerged at the centre of global debates about agricultural commercialisation. Contract farming, it is argued, provides an opportunity for the rural poor to participate in production in commercial value chains and access markets for their produce. 2 Contract farming is widely associated with 'land grabbing' in Africa. 11 However, as Vermeulen and Cotula further observe, contract farming needs to be scrutinised on its 'specific terms' because it can either improve smallholders' access to markets or extend exploitative relationships in which smallholders carry a disproportionate share of production and market risks. While outgrowing needs to be studied on a case-by-case basis, a focus needs to be on the impacts of class dynamics and social differentiation on outgrowers. This is because the terms of inclusion of outgrowers in large-scale agriculture such as sugar-growing schemes are uneven, and are affected by a range of factors.
This article examines these dynamics in Kilombero, asking who the winners and losers are, and why. It explores social differentiation and how it is shaped by the existing institutional and legal framework governing the sugar industry, foreign capital investments and actual practices on the ground. The article draws on extensive field research conducted between 2010 and 2014. 12 Interviews were conducted between July 2013 and September 2014 with 64 respondents from KSCL, government institutions, outgrowers and leaders of outgrower associations.
The next section describes the current large-scale agricultural commercialisation initiatives in which sugar cane is being promoted. This is followed by a brief historical overview of the Tanzanian sugar industry, its current state, and a synopsis of the institutional and legal framework that governs the sugar industry. The case of KSCL is then discussed and analysed. This is followed by an analysis of the social differentiation of outgrowers and how these communities have been affected by planned large-scale initiatives. In conclusion, the article examines the factors that affect how smallholders are incorporated into large-scale sugar production schemes, and with what results.

Sugar Cane: A Central Crop under Large-Scale Initiatives
Currently, sugar occupies centre stage in Tanzania's national political economy. It is among the three priority crops under 'Big Results Now' (BRN), the presidential initiative modelled on Malaysia's Big Fast Results programme. The BRN programme aims to transform the country into a middle-income economy by 2025 through investments in six key sectors, of which agriculture is one. 13 Contrary to previous investments in agriculture, which focused on smallholder farmers, BRN aims to attract private sector investment in medium-and largescale agriculture, notably maize, rice and sugar cane. The government aims to conclude 25 commercial farming deals for paddy and sugar cane by 2015 as part of BRN plans. 14 According to the Ministry of Agriculture, Food Security and Cooperatives, sugar cane is prioritised in order to fill the persistent domestic sugar deficit of 300,000 tonnes per year and to export surplus into regional and global markets. 15 The rationale for the prioritisation by the Tanzanian government of sugar cane as one of three cash crops is not entirely clear. Sugar production is frequently subsidised in major sugar-producing countries, leading to overproduction and dumping of cheap sugar onto the world market. 16 This situation disproportionally affects smaller sugarproducing countries like Tanzania. However, despite this, the government argues in favour of ensuring a reliable and secure supply of (domestically produced) high-quality sugar, and job creation in the country. 17 Alongside BRN is another national development initiative, the Southern Agriculture Growth Corridor of Tanzania (SAGCOT), which is a multi-billion-dollar initiative to commercialise farming across a large swathe of central to southern Tanzania, with the support of bilateral agencies, international financial institutions and private agribusiness corporations. SAGCOT is being supported by the G8 as part of its New Alliance for Food Security and Nutrition, and aims by 2030 'to bring 350,000 hectares of land into profitable production, transition 100,000 smallscale farmers into commercial farming, create 420,000 new employment opportunities, lift two million people out of poverty, and generate $1.2 billion in annual farming revenue'. 18 So far, the Tanzania Investment Centre has put up for auction large areas of up to 63,000 hectares (ha), 19 and 20 more sugar cane estates and mills are to be established through public-private investment. This will include nucleus estates alongside small and medium-scale farmers supplying sugar cane as outgrowers. 20 Outgrowing of sugar by smallholders is central to all these plans.
Despite the rhetoric of these grand plans, they largely remain on paper, 21 and they bear a resemblance to past grand agricultural schemes in Tanzania and elsewhere that have failed. 22 Although BRN and SAGCOT seem to have attracted large sums of funding support from a wide range of major donors 23 (including the UK Department for International Development, the US Agency for International Development, and the Norwegian Agency for Development Cooperation, to mention a few), their predecessors were failures because they did not garner the requisite levels of investment and, critically, they suffered from inadequate planning and top-down approaches -precisely the same maladies as under the current initiatives. The 2015 review of the G8's New Alliance implementation in Africa found that companies are 'mostly submitting existing investment plans to garner favour with governments, secure a seat in policy dialogue or to win good publicity'. 24 A challenge facing all these initiatives is the absence of easily accessible land for large-scale commercialisation. Land deemed to be available for investment is mostly village lands that cannot be secured without displacing villagers. Second, water remains an obstacle, even in areas already identified as suitable for sugar cane production. As Andrew Coulson, formerly a staff member at Tanzania's Ministry of Agriculture, writes: 'the large-scale farms have yet to prove themselves -and those who advocate them should remember that water resources are often less than anticipated, rains unreliable, the soils easily exhaustible'. 25 The central place of the sugar industry in these large-scale agricultural commercialisation initiatives provides an opportunity for exploring what is happening on the ground as a result of the restructuring of the sugar industry in the 1990s, and for looking beyond the ambitious objectives and impressive targets to the reality of local contexts and practices.

The Sugar Industry and its Regulatory Framework
The history of sugar cane production in Tanzania dates back to the early 1920s, during the time of the colonial administration of then Tanganyika. Sugar production was based on smallholder cane jaggery 26 in the Kilombero and Mtibwa valleys. It was only in the late 1960s that the first commercial sugar factory was established at Msolwa village in Kilombero District, and several others followed. 27 In 1967, the government nationalised private firms, including sugar estates and mills. These were placed under the National Food Corporation (NAFCO) until the government established the Sugar Development Corporation (SUDECO) in 1974 to deal with the development of the national sugar industry, including distribution, exports and imports. These efforts led to the establishment of two new state-run estates: Mtibwa Sugar in 1973 and Ruembe Sugar Factory (Kilombero II) in 1976. In 1982 the small Kagera sugar plant, damaged during the Uganda-Tanzania War in 1978, was rehabilitated. 28 Currently, there are five active sugar companies in the country, of which four are producing sugar commercially: Kilombero Sugar Company (KSCL) and Mtibwa Sugar Estates in Morogoro, Tanganyika Planting Company (TPC) in Kilimanjaro and Kagera Sugar in Kagera. KSCL has two medium-sized mills, while the other companies have one mill each. Although the imports and exports of sugar remain regulated by the Sugar Board of Tanzania, the sugar industry was liberalised in the early 1990s and prior to the companies' privatisation at the turn of the millennium. The government retained a 25 per cent stake in two companies, KSCL and TPC, and the rest are fully owned by private firms. Among these four companies, only TPC depends wholly on its estate for sugar cane, and the rest have varying numbers of outgrowers supplying sugar cane to the mills on contract. A fifth sugar company, Bagamoyo EcoEnergy Ltd (BEE), aims to develop an 8,000-ha sugar cane estate in Bagamoyo District and to recruit outgrowers on a further 3,000 ha. BEE plans to produce sugar and ethanol for the domestic market.
Tanzania's sugar industry employs about 18,000 people in direct jobs and supports over 300,000 people employed in related businesses. 29 Sugar cane outgrowers earn more than TZS45 billion (US$28 million) 30 annually and through all its value chain activities the industry generates about TZ12.4billion (US$8 million) in revenue for the government, or 1.7 per cent of total tax revenue. 31 Since privatisation, sugar producers in the country claim to have invested over TZS813.5 billion (US$500 million) 32 in the rehabilitation of factories, expansion of estates and mobilisation of outgrowers. Between 1998 and 2013, the area under sugar cane estates grew from 16,000 to 36,000 ha and the area under outgrowers expanded from 5,000 to 19,000 ha. 33 Despite this expansion of the industry, Tanzania continues to be a net importer of sugar, with an estimated annual sugar deficit of around 300,000 tonnes. 34 To fill this gap, the government annually issues permits to traders to import sugar at a reduced or zero tariff, and it has repeatedly appealed to existing producers to increase their production, while promoting available investment opportunities in the sugar sector. 35 The importation of cheap sugar has been a major problem for outgrowers and millers alike for several years now. These imports are made by either illegal traders or those with legal permits issued by the government. 36 Experience shows frequent excessive imports. For instance, in 2012, domestic sugar production shortfalls were estimated at about 80,000 tonnes, 37 but zero tariff imports permitted during the year amounted to nearly 200,000 tonnes (see Figure 1). 38 The problem of imports resurfaced in 2013-2014, badly hurting KSCL and its outgrowers, who jointly responded by advocating strict measures against cheap sugar imports, and opening up alternative approaches towards imports. In 2014 KSCL and outgrower representatives held several meetings with the members of parliament dealing with industrial and agricultural sectors in the country, as well as with the prime minister.
Yet, as a net importer of sugar, Tanzania's sugar industry is continuously affected by unstable sugar prices on the international sugar market. The international market and 'preferential markets' 39 for sugar are affected by the dumping of excess sugar from main producers and weather changes in these countries. 40 Tanzania, like many other governments, intervenes to stabilise sugar prices, either through import subsidies or waiving import taxes. 41 The government's efforts to stabilise prices are influenced more by pressure from urban consumers than rural producers. 42 But the importation of cheap sugar, mostly by a few politically connected businesspeople, undermines local production, and in recent years Tanzanian producers have called on the government to control such imports. 43 Recent calls for scrutiny of import practices have come from mostly black outgrowers, while the import trade has been controlled largely, since independence, by traders with mostly Asian and Arab origins. It is this same group that is alleged to have deep-rooted connections to the ruling elites and who 'finance the ruling party in exchange for tax exemptions and reduced or waived custom duties'. 44 In an attempt to regulate the liberalised sugar industry, the government enacted the Sugar Industry Act No. 26 of 2001, which established the Sugar Board of Tanzania (SBT). The Board's directors are appointed from the Tanzania Sugar Producers' Association, the Tanzania Sugarcane Growers' Association, consumers, and the Ministry of Agriculture. While the Sugarcane Growers' Association used to represent the interests of all cane outgrowers, a number of outgrowers associations are no longer allied to it after tensions between large and small growers' associations arose. In response, the small growers' associations have formed the Council of Cane Growers' Association, which they want to represent them on the Sugar Board. 45 Under the current regulations, an outgrower requires a minimum farm size of 0.4 ha within a 40-km radius of a registered miller to which the outgrower has the obligation to sell sugar cane. All outgrowers need to register with the Sugar Board in their own name, or their association must enter into a commercial agreement with a sugar factory to which they are to sell cane. 46 The procurement and distribution business of locally produced sugar is concentrated in a handful of private companies. 47 They collect sugar from producers and store it in their warehouses, mostly in Dar es Salaam for sugar produced in Morogoro. They then distribute to wholesalers around the country except in regions supplied by Kagera Sugar and TPC. Given their monopsonistic access to producers who also favour this system, distributors decide to whom and where they sell. This arrangement adds to transportation costs incurred by traders as they have to purchase sugar from Dar es Salaam and then bring it back to consumers.  This analysis of the sugar industry and its regulatory framework provides us with the context within which to assess the production of sugar by KSCL and its outgrowers scheme in the next sections.

The History and Evolution of Kilombero Sugar Company
In 1961, KSCL established its first estate, with 1,600 ha and a sugar-processing mill named K1 in Msolwa village near Kidatu Township (see Figure 2). In 1998, KSCL was privatised and the South African-based company, Illovo Sugar Ltd, bought a majority share. Illovo itself is fully owned by Associated British Foods Plc, which completed the full acquisition of Illovo in June 2016. 49 The company leases a total of 12,000 ha of land from the government of Tanzania on which it runs two mills and two estates under cane covering 9,562 ha. 50 The company accesses water from the Great Ruaha River that flows across its two estates. It uses and controls large assets built by the state: offices, staff houses, schools and health facilities, irrigation, road infrastructure and a small airport, and it has access to the Tanzania  and Zambia Railway for transporting sugar. Since privatisation, the company has improved the factory facilities, but has not been able to expand the size of its estates due to land scarcity in the area. The company has encouraged the growth of independent registered outgrowers, raising their number from 2,400 in 1998 to 8,500 in 2014. 51 Outgrowers' land under cane is estimated at 15,000 ha (see Figure 3). 52 Although the inclusion of outgrowers was part of the government strategy to expand production before privatisation, Monson 53 argues that for Illovo this was a strategy to reduce operating costs by transferring production risks to outgrowers. In addition to sugar, the company also produces molasses and bagasse 54 for its own electricity generation.
In 2013, the company installed a distillery to produce ethanol targeted at the domestic market.
In 2012-13 KSCL production reached 129,700 tonnes of sugar, representing about seven per cent of the total sugar produced by Illovo Group. 55 This output is however below the company's agreement with the government of Tanzania to produce 200,000 tonnes per annum. Having acquired KSCL, Illovo Group introduced its corporate production and business practices. 56 These include new labour practices, outsourcing of services and the use of outgrowers in its expansion strategy. Illovo's investment in Tanzania is an example of the tendency for many sub-Saharan African states to rely on foreign capital taking over state-owned 'parastatal' companies. 57 Consequently companies such as Illovo have been able to dominate sugar production in Malawi, Zambia and Tanzania. Illovo also operates mills and estates in Mozambique, South Africa and Swaziland. 58 However, despite its initial success in increasing sugar cane production and the number of outgrowers in Kilombero Valley, the image of Illovo and other similar companies remains tainted by the stigma of apartheid. 59 As discussed in the next section, outgrowers have been central to the production of sugar cane processed by KSCL since the 1960s, but the terms of outgrower contracts have changed markedly under Illovo.

Sugar Cane Outgrowers in the Kilombero Valley
Sugar cane outgrowing in the Kilombero Valley started in the early 1960s when smallholder farmers began to grow sugar cane alongside their major food crops such as maize, millet and rice. 60 After acquiring KSCL, Illovo encouraged villagers in the nearby areas to grow sugar cane by assuring them of high prices, and the provision of inputs, infrastructure, and managerial and technical support. 61 In addition, the EU has invested in the upgrading of roads within the sugar-producing areas, strengthening outgrowers' associations and the establishment of sugar cane block farms. 62 In addition, Solidaridad -a Dutch non-governmental organisation (NGO)is collaborating with the Kilombero Charitable Trust founded by KSCL to build the capacity of outgrowers' associations. 63 Outgrowers' production share steadily increased, reaching 45 per cent of the cane used by the company in 2013-2014 season. 64 They supply sugar cane based on a cane supply agreement (CSA) that the company signs with outgrowers' associations every three years. The CSA specifies the terms of partnership, including the division of proceeds (DP) between the two parties. In May and June 2013, 14 associations of cane outgrowers and KSCL renegotiated the CSA and reached an agreement that the proportion of the outgrowers' proceeds would increase by 0.5 per cent every year for the next three years, with the target ratio to be reached of 60 per cent of revenues for the outgrowers and 40 per cent for the company. 65 In April 2014, the DP was 57 per cent to outgrowers and 43 per cent for the company. The DP is expected to rise to 59 for outgrowers in the 2015-2016 harvesting season. 66 Outgrowers, nonetheless, do not receive full payment on delivery of their sugar cane. Under this arrangement, outgrowers receive 90 per cent of their pay on the 15th of the next month after they have delivered sugar cane to the mill. The company holds 10 per cent of the outgrowers' expected proceeds until the end of the season to compensate for any price difference in domestic and international markets. Any balance from this amount is paid to farmers at the end of the sales season, 67 meaning that outgrowers earn more benefits when the sugar price is greater than forecast, or carry the market risk for their sugar cane from the time of delivery to wholesale, when the sugar earns a lower price.
Overall, outgrowers' incomes from sugar cane remain volatile, affected by the imports of cheap sugar, agronomic factors and their terms of contract with the buying company. Their earnings from sugar increased from TZS1 billion (US$614,628) in 1998 to TZS38 billion (US$23.4 million) in 2013. The price of outgrowers' sugar cane per tonne increased to TZS69,000 (US$41.50) during the 2013 season. However, due to importation problems, prices fell to TZS62,889 per tonne (US$37.80/t), and for the 2013-14 season, the price fell further to TZS58,000/t (US$34.90). 68 Apart from cheap imports, outgrowers' earnings are affected by levels of sucrose -the cane's crystalline sugar content -and tonnage recorded at the mill. Outgrowers interviewed during this study have complained about the lack of transparency in the existing weighing procedures and measurement of sucrose levels. Outgrowers' complaints about KSCL employees providing inaccurate or false sucrose levels for their cane have been documented since 1987 by Ellen Sprenger, 69 before KSCL was privatised. However, it is important to note that sucrose levels are affected by many other issues. Smalley, Sulle and Malale (2014) attribute the fall in the sucrose levels of outgrowers' sugar cane to factors such as lack of effective extension services and a poor regulatory framework for monitoring the weighing and measuring of sucrose levels when sugar cane is delivered to the mill. 70 The terms of partnership between the company and outgrowers, as described in the CSA, has changed over time. While at the beginning of its operation Illovo offered some direct support services such as providing seeds, extension services and transportation, most of these activities are now carried out by the outgrowers' associations. This is part of the reshaping of contract farming by the private sector, especially after acquiring the state-owned company. 71 For example, the CSA states that the company 'has no obligation or responsibility to make 71 Oya, 'Contract Farming in Sub-Saharan Africa'. available to the associations, or to the association growers ... any management, operational, extension or administrative services, other than as may be mutually agreed in writing from time to time'. 72 In contrast to the research-based argument that contract farming may enable outgrowers in different contexts 73 to access inputs and extension services from agribusinesses, in this case farmers receive little or no help from the company. 74 However, a collective voice among outgrowers is limited, due to the fragmentation over time of their associations. Sugar cane outgrowers' associations have grown from two in 1998 to 15 in 2014. 75 This growth is largely associated with the failures of the first two associations to address outgrowers' challenges that include access to equitable and timely harvesting quotas and allegations of patronage. Massimba, Malaki and Waized argue that the mushrooming of numbers of outgrowers' associations weakens their bargaining powers with the company. 76 Outgrowers pay a one-time membership fee of TZS5,000 (US$2.6) that is used by the association to run the office and other outgrowers' affairs. 77 They also pay TZS200/t (US$0.10) for services like transportation, loading and road maintenance, the cost of which is deducted from their earnings after the associations have supplied sugar cane to the company. Unlike KSCL, which depends on migrant labour for cane planting, weeding and fertilising, many outgrowers depend on family members to do their farm work. The use and availability of family labour varies between richer and poorer farmers, with those better off hiring workers. 78 The next section explores these dynamics among outgrowers and their associations.

Sugar Cane Expansion: Class Dynamics and Social Differentiation
The boom in the sugar business in the Kilombero Valley has attracted great interest from local elites and the general public. The area under outgrower production increased from a mere 3,500 ha in the 1990s to 15,000 ha in 2014. 79 But these outgrowers are highly differentiated: they have different sizes of farm, and their produce and the income they earn from it vary widely. According to Illovo reports, in the 2002-2003 season the company sourced cane from 3,384 small-scale, eight medium-scale and only three large-scale outgrowers, 80 but by 2013-2014 season these numbers had increased substantially for each category (see Table 1). 81 The role of medium-and large-scale outgrowers has also been growing over time, expanding accumulation by this class of producer. 82 In 2006, outgrowers collectively produced more sugar cane than the company's core estates (see Figure 4).
Within the outgrowers' associations there is a high level of class dynamism, social differentiation, and patronage. Small-scale farmers that were interviewed claimed that local elites ensure that outgrowers' associations largely serve their own interests. For example, most of the associations' constitutions require leaders such as the chairperson to have a farm of 50 ha or more, and only a few new associations require that the chairperson have at least 10 ha of land under cane. In practice, this means that elected leaders are medium-or large -scale farmers, several of whom also own equipment and vehicles such as tractors, loaders and trucks. They provide these on tenders to their own associations to cut, load and transport members' sugar cane to the mill. Large-and medium-scale outgrowers come from a range of backgrounds, and include former civil servants, who often hold positions in the outgrowers' associations. 83 Association leaders are also able to influence the allocation of quotas for the quantity of sugar   Small-scale  6,320  Less than 5  70  Medium-scale  1,667  Between 5 and 50  19  Large-scale  13 Over 50 11 cane their members can deliver to the factory each season, a process that has allegedly been manipulated in favour of the larger farmers. These practices reduce small outgrowers' chances of having their sugar cane harvested in time, as aptly stated by one respondent: 'we live in a state of fear, we do not know if our sugar cane will be harvested this year around, so my wife and I have decided to register in different associations; hopefully, one of us will get a quota'. 84 These patterns of class formation in Kilombero and Tanzania in general are not new and started before privatisation of KSCL, although they have been heightened since 2000. While the largest producers among outgrowers are small-scale farmers with less than five hectares, the company is not certain that these kinds of farmers are suitable for its business in the long run. This is clear in one of its reports: 'outgrower farms at Kilombero are highly fragmented; with sizes as small as 0.02 ha […]. Such a setting requires a unique way of organizing cane-growing to achieve [a] sustainable and dependable cane supply system'. 85 The company is also concerned with the growing number of outgrowers' associations. According to a KSCL senior employee, the company would prefer to have as few as two associations, as it had before it was privatised, in order to reduce transactions costs related to management and communication. 86 As sugar cane production expands, the availability of land for both the company and outgrowers remains a major constraining factor. The company acknowledges that about '60 per cent of the land within the 40 km radius from KSCL's is under cane, with much of the remainder used for other crops'. 87 This, however, ignores the fact that some of the remaining land is used for communal settlement, including areas for social services and national parks. The lucrative prices for sugar cane from the company and the previously reliable market have encouraged outgrowers to put most of their farmland into sugar cane, leaving little space or none for food crop production. It is also difficult to leave sugar cane farming and opt for the production of food crops, such as rice and maize, because birds nesting in sugar cane eat these crops. 88 This means that families either look for alternative food-producing land in distant areas 89 or depend on buying foodstuffs from the market.

Category of outgrowers Number of outgrowers Land cultivated (ha) % of sugar cane supplied to KSCL
Among poor families, producing food crops in distant areas means that one or both parents must commute from one location to the other in different seasons, leaving children on their own or with one heavily occupied parent at home. This mobility may be associated with significant social problems, such as early pregnancies and girls dropping out of school. 90 On the other hand, rich farmers are able to rent or purchase enough land and hire equipment and labour for the production of food and cash crops. Yet, while these faraway villages now act as a buffer zone for land-constrained families, large-scale initiatives like SAGCOT are likely to absorb most of these lands, turning the majority of these poor farmers into wage labourers in estates or unemployed urban dwellers. Of course this will not be without resistance 'from below', particularly from those communities to be displaced. 91 The expansion of sugar cane farming by the company, including that which involves outgrowers' schemes, is therefore not a simple task. Any plan that involves the eviction or relocation of the local community is subject to all sorts of contestation over land ownership at the local and national levels. For instance, after securing KSCL in 1998, Illovo wanted to get full control of all estate land as indicated in the documents it received from SUDECO. However, since the previous management of the company did not cultivate all land, the nearby communities occupied uncultivated land, and it became difficult to remove all of them because the central government was reluctant to compensate the occupants. 92 As a result the company asked for extra land for the estate and the factory, and the SBT offered its previously surveyed land in Ruipa several kilometres away from the current KSCL location. District officials supported this proposal, 93 but the communities, who had occupied the land since it was first earmarked for sugar cane development back in the 1970s, refused to leave uncompensated, leading to a long-lasting land dispute. 94 Community resistance against the corporate takeover of their land is critical for understanding power relations between state bureaucrats and local communities. The first element of this is the contestation between local and central government authorities over the allocation of communal land (or village land) to the investor. Second, despite their wishes to allocate village land, the pro-investment state bureaucrats have been unable to evict communities because of the safeguards enshrined in the Village Land Act of 1999. 95 The law requires villagers' consent before their land is transferred to a general land category and used for investment purposes, although villagers have fewer powers to resist the allocation of more than 250 ha by the Commissioner of Lands. Moreover, the land acquisitions by both local and foreign investors have recently attracted the attention of local and international NGOs and parliamentariansmostly from the opposition camp. As a result, local communities have strengthened their resistance to land acquisitions for large-scale investments that displace them. 96 This is why the promotion of large-scale farming under SAGCOT and BRN has been criticised by a number of local and international farmers' organisations as misguided. The executive director of the national network of smallholder farmers in Tanzania, commonly known as Mviwata (Mtandao wa Vikundi vya Wakulima Tanzania) put it succinctly: Frankly speaking, most initiatives like the SAGCOT target large-scale farmers and probably profitoriented foreign companies operating in the country. Smallholder farmers are always losers in most government initiatives." 97 This resistance from the local communities and farmers associations underscores the need to understand local politics and its influence in the policy implications of large-scale investments that involve acquisition of land.
The current farming models promoted through these initiatives radically shift the direction of agrarian change towards large-scale agriculture, despite the rhetoric about contract farming being a 'win-win' option. Such initiatives are particularly driven by global capital interested in the use of capital-intensive production systems, which may displace smaller-scale farmers for larger, capital-rich firms and farms to be able to produce at scale for global markets. 98 SAGCOT implementation plans involve the introduction of new settlement schemes to allow for the establishment of nucleus estates. As Michael Bergius argues, the current SAGCOT project 'represents an expansion of the corporate food regime in the country'. 99 This is particularly of concern, because with the poor investments in public infrastructure and technologies, smallscale producers are unlikely to compete with the multinational corporations.

Conclusion
This article reveals the highly differentiated nature of sugar cane outgrowers and their associations. Access to, control over, and ownership of resources and assets varies widely between the small-, medium-and large-scale farmers, and between those wholly reliant on farming for their livelihoods and those with other business enterprises or off-farm jobs. Since 2000, the number of medium-and large-scale farmers entering into sugar outgrowing who have accumulated assets and taken advantage of sugar cane business opportunities that include sugar cane-cutting, loading and transportation has increased massively. The company describes them as 'commercial farmers', 100 and indeed they are the clear winners in the process of differentiation under way at Kilombero.
Within these processes of transformation, the government of Tanzania has often failed to safeguard the interests of outgrowers, while allowing the privatised sugar industry to set the terms of contracts in ways that externalise costs and risks 101 to outgrowers and their associations. This is what underpins the process of class differentiation, whereby the larger outgrowers are able to accumulate within such conditions, by providing goods and services to the smaller outgrowers through local patronage relationships. They thus profit from this business model due to their ability to diversify. These processes of social differentiation among sugar cane growers, in the context of the lack of state policy supporting smallholder outgrowers, are well documented elsewhere. 102 In addition, findings support observations made in earlier studies that outgrowing has potential negative gender impacts. 103 These need further attention in future research, singling out such impacts particularly for children and women, and how to address them. 104 The KSCL business model encourages outgrowing, but in the longer term sees this being focused on medium and larger-scale farms. The planned expansion of sugar cane land will require expropriation of land, and further social differentiation. Resistance to such changes occurs, but the power of the state-sugar alliance in national politics, backed by aid donors and international capital, is substantial. For smallholder outgrowers especially, the contractual arrangements remain problematic. Prices for outgrowers' sugar cane is affected by market externalities and there is growing distrust between the company and outgrowers. Regulations fail to stop frequent distortion of domestic sugar prices by cheap imports. Ambitions to expand local production of sugar therefore need to take into account the impacts of global overproduction that lead into the dumping of cheap sugar into the world markets. Combining sugar cane with other crops is limited by land area and pest attacks, and the trade-off between household food security and sugar cane production is significant for small-scale producers without large land areas or other income sources. For all growers, the sugar industry needs an effective legal and institutional framework that spells out a transparent and effective mechanism to protect outgrowers against production and market risks. Better company policies and practices should include a transparent means of weighing cane and measuring sucrose levels.
The 'outgrower model' is central to national development initiatives such as the Southern Agriculture Growth Corridor of Tanzania and 'Big Results Now'. This paper has shown how outgrowers are increasingly differentiated, posing the question of which 'outgrowers' will benefit from these initiatives? Despite the grand ambitions of large-scale agricultural commercialisation and the rhetoric of 'pro-poor development', who wins and who loses out depends on the local political economy, where the sugar industry, local business, political elites and local communities compete for the benefits of expanded sugar production. In this, smallholder outgrowers are clearly losing out.