%20-%20https://sketch.com%20--%3e%3ctitle%3e3ECBBF50-FD08-412A-943D-2CFC5E025A16%3c/title%3e%3cdesc%3eCreated%20with%20sketchtool.%3c/desc%3e%3cg%20id='Stylesheet'%20stroke='none'%20stroke-width='1'%20fill='none'%20fill-rule='evenodd'%3e%3cg%20transform='translate(-926.000000,%20-668.000000)'%20id='BUTTONS'%3e%3cg%20transform='translate(100.000000,%20597.000000)'%3e%3cg%20id='BTN-/-RIGHTS'%20transform='translate(816.000000,%2065.000000)'%3e%3cg%20id='ICON-/-Rights'%20transform='translate(10.000000,%206.000000)'%3e%3cpath%20d='M13.5752841,0%20L1.34943182,0%20C0.59375,0%200,0.586666667%200,1.33333333%20L0,22.6666667%20C0,23.4133333%200.620738636,24%201.34943182,24%20L17.6505682,24%20C18.40625,24%2019,23.3866667%2019,22.6666667%20L19,5.36%20L13.5752841,0%20Z%20M13.5752841,1.89333333%20L17.0838068,5.36%20L13.5752841,5.36%20L13.5752841,1.89333333%20Z%20M1.34943182,22.6666667%20L1.34943182,1.33333333%20L12.1988636,1.33333333%20L12.1988636,5.36%20C12.1988636,6.10666667%2012.7926136,6.69333333%2013.5482955,6.69333333%20L17.6235795,6.69333333%20L17.6505682,22.6666667%20L1.34943182,22.6666667%20Z'%20id='Shape'%20fill='%23595959'%20fill-rule='nonzero'%3e%3c/path%3e%3ccircle%20id='Oval'%20stroke='%23595959'%20stroke-width='1.5'%20fill='%23595959'%20cx='9.5'%20cy='13.5355339'%20r='5.5'%3e%3c/circle%3e%3cpolyline%20id='Path-2'%20stroke='%23FFFFFF'%20stroke-width='1.5'%20transform='translate(9.500000,%2013.035534)%20rotate(45.000000)%20translate(-9.500000,%20-13.035534)%20'%20points='8%2016.0355339%2011%2016.0355339%2011%2010.0355339'%3e%3c/polyline%3e%3c/g%3e%3c/g%3e%3c/g%3e%3c/g%3e%3c/g%3e%3c/svg%3e)
INTRODUCTION
Since 2008, the phenomenon of ‘land grabbing’ has been increasingly studied, generating an impressive yet still emerging scholarship on the subject. In East Africa, Tanzania has been heavily targeted by investors in this ‘global land rush’. The Tanzanian government has been actively courting investors, arguing that they will ensure agriculture's modernisation and economic development. Yet, the implementation of agri-food projects is highly problematic, as many investors have failed to develop their land and start operations. In response to this issue, Tanzania's central government has recently pledged publicly its commitment to fight these unproductive investors, thereby shifting its discourse from an exclusive focus on welcoming investors to the importance of having operational investors who can effectively foster socioeconomic rural development.
Focusing on Rufiji district – one of the most targeted areas for land investments in Tanzania – this article compares two state agencies’ involvement in land acquisition processes in the context of this new strategy on productive investors.
My discussion is based on nine months of political ethnography that was carried out between June 2016 and September 2017 to study the local impacts of investment in land in two Tanzanian districts, Rufiji and Misennyi. In addition to participant and non-participant observation, I conducted over 250 semi-structured and open interviews with various stakeholders (i.e. international investors, state officials, local leaders and local peasant populations). In Rufiji district specifically, I investigated five distinct investment projects that were targeting land in six different villages.
This article makes two theoretical contributions. First, drawing on the sociology of the state literature, I draw attention to the importance of acknowledging the practical consequences of the contentious and fragmented authority of African states and its impacts on relations among government actors. In addition, I demonstrate that a central monitoring process might have a negative effect. Instead of protecting vulnerable populations, it fosters institutional innovations that aim at protecting local bureaucrats’ opportunities for accumulation with investors. Building on Buur et al. (Reference Buur, Nystrand and Pedersen2017), I argue that to better grasp the complexity of collaborative and competitive interrelations between actors within the state, we need to further unpack bureaucrats as a group of actors.
Second, I contend that the intermediary role played by state officials deserves to be further investigated. In Tanzania, local bureaucratsFootnote 1 often act as brokers between investors and local populations. As a result, interactions between local populations and investors might be almost non-existent, since investors delegate the negotiation process to the bureaucrats. Moreover, the political centralisation of land management entailed by the Tanzanian government's new strategy has nurtured the tendency to exclude village administrations from land negotiations. This exclusion of village authorities is likely to generate conflictual dynamics at the local level between investors and local populations and may impede the investors’ capacity to become operational. Thus, the quasi-systematic presence of local bureaucrats as intermediaries is key to understanding the relationship between investors and local populations.
UNBUNDLING THE EFFECTS OF CONTENTIOUS AND UNEVEN STATE AUTHORITY
After focusing attention on quantifying land deals in developing countries, the initial scholarship on ‘land grabs’ has been criticised by scholars who argued for a shift towards a more qualitative and empirically based standpoint to examine their socio-political and economic local impacts (Baglioni & Gibbon Reference Baglioni and Gibbon2013; Edelman Reference Edelman2013; Edelman et al. Reference Edelman, Oya and Borras2013; Oya Reference Oya2013a; Scoones et al. Reference Scoones, Hall, Borras, White and Wolford2013; Locher & Sulle Reference Locher and Sulle2014).
Recent works in political economy have significantly contributed to our theoretical understanding of land acquisition processes, and of their local outcomes. First, the assumption that global land grabbing necessarily leads local people to engage in resistance activities has been thrown into question (Borras Reference Borras2011; Borras & Franco Reference Borras and Franco2013; Oya Reference Oya2013b; Hall et al. Reference Hall, Edelman, Borras, Scoones, White and Wolford2015a, Reference Hall, Scoones and Tsikata2015b). Local reactions to land grabs are not necessarily organised or constant across time and space. Instead, variance of local reactions appears to be context dependent (Purdon Reference Purdon2013; Scoones et al. Reference Scoones, Hall, Borras, White and Wolford2013; Hall et al. Reference Hall, Edelman, Borras, Scoones, White and Wolford2015a; Castellanos-Navarrete & Jansen Reference Castellanos-Navarrete and Jansen2015; Mamonova Reference Mamonova2015).
Second, recent scholarship has pointed out the need for a broader and systematic analysis of the political economy in which land investments are embedded. Local contexts in the reception of land deals are shaped by historical, institutional and discursive legacies (Baglioni & Gibbon Reference Baglioni and Gibbon2013; Exner et al. Reference Exner, Bartels, Windhaber, Fritz, See, Politti and Hochleithner2015; Haulle Reference Haulle2015; Greco Reference Greco2016; Pedersen Reference Pedersen2016), and processes of accumulation are embedded into historical relations (Wily Reference Wily2012; Martin & Palat Reference Martin and Palat2014).
Moreover, scholars have shown how land deals are intertwined with broader official political objectives of developing the countryside, while covertly enhancing the state's political control in peripheral regions. For instance, Lavers & Boamah (Reference Lavers and Boamah2016) argue that agricultural investments might be used by the state to deploy its infrastructural power over a society. As such, these deals contribute to the transformation of the state itself (Lavers & Boamah Reference Lavers and Boamah2016). As argued by Wolford et al. (Reference Wolford, Borras, Hall, Scoones and White2013: 194), ‘land deals are nothing more or less than transformations in the ground on which states are formed’.
Yet, this scholarship has also stressed the importance of recognizing that the state is not a unitary actor. Unbundling the state is key to examining relations of power that define interactions between state actors and with society (Peluso & Lund Reference Peluso and Lund2011; Cotula Reference Cotula2012; Wolford et al. Reference Wolford, Borras, Hall, Scoones and White2013; Buur et al. Reference Buur, Nystrand and Pedersen2017). Moreover, as argued by Pedersen (Reference Pedersen2016), actors’ agency at all levels should be systematically empirically investigated because governance processes are more contingent than usually acknowledged. Thus, there is a multiplicity of actors involved in land investments, and the role they play and their interests should be investigated more thoroughly to capture how they impact land acquisition and power relations at the local level.
Building on these valuable insights, Buur et al. (Reference Buur, Nystrand and Pedersen2017) have proposed an innovative analytical framework to investigate the implementation process of land investments. It insists on interactions between three sets of actors: ruling elites, investors and local populations. They characterise three types of exchange relations which help in the understanding of divergent actors’ interests and how potential material and non-material exchanges determine the degree of convergence of their positions, which is key to understanding how investment projects unfold (Buur et al. Reference Buur, Nystrand and Pedersen2017: 4).
This analytical framework is useful to understand processes of land acquisitions and how they determine their local outcomes. However, I argue that the practical consequences of the contentious and fragmented authority of many African states and its effect on relations between actors should be more systematically considered. The sociology of the state literature offers us valuable insights that help us to better grasp and consider the complexity of collaborative and competitive interactions between actors.
First, as Lavers & Boamah (Reference Lavers and Boamah2016) argue, insights from Michael Mann and James C. Scott are helpful for understanding the nature of the relations between state and society, which they depict as explicit and purposeful strategies of power. For instance, Mann (Reference Mann1984) was the first to introduce the concept of infrastructural power, i.e. power through society, which he contrasts to despotic power, i.e. power over society. Infrastructural power originated from modern states' prerogatives of territoriality, and political centralisation. Over time, states seek to increase their infrastructural power in order to better penetrate, and efficiently enhance their control over society (Mann Reference Mann1984: 117–19; Soifer Reference Soifer2008). Yet, though there is variation across cases, in a number of African countries, infrastructural power of central governments is relatively weak (Boone Reference Boone2003; Soifer Reference Soifer2008; Soares de Oliveira & Taponier Reference Soares de Oliveira and Taponier2013; Rasmussen & Strøh Varming Reference Rasmussen and Strøh Varming2016; Lavers & Boamah Reference Lavers and Boamah2016). Consequently, they rely on an amalgam of despotic/coercive power and infrastructural power to assert their control over society.
Scott's works complement Mann's thesis in addressing explicitly how the state attempts to consolidate its infrastructural power. He shows that ‘legibility’ was and still is a major mechanism of statecraft. In order to render society and territory more ‘legible’, and thus administrable, states have undertaken large-scale engineering projects privileging a rationality of simplification and standardisation – based on a high modernist ideology, which entails not only a reorganisation of space but also of social relations (Scott Reference Scott1998). Yet, as Scott also adds, many states have failed in achieving these objectives. Accordingly, Herbst has showed that expanding authority for rulers in Africa, from the pre-colonial to the post-colonial period, was heavily constrained by structural factors such as the geography, low people density and availability of land (Herbst Reference Herbst2000). Consequently, an important challenge for African states was and still is to build their authority over their whole territory. Though space is lacking here, it is important to add that this pattern was already at play during the colonial period. Scholars such as Sara Berry (Reference Berry1993) showed the variation in colonial states’ attempts at broadcasting their authority over their entire territory. As Berry termed it, this was often ‘hegemony on a shoestring’, whereby ‘the presence of the state in rural economies has been intrusive rather than hegemonic’ (Berry Reference Berry1993: 22).
An important scholarship on the nature of African states echoes Herbst's argument by arguing for a shift from state-centric perspectives to understanding governance practices (Rasmussen & Strøh Varming Reference Rasmussen and Strøh Varming2016). Scholars call attention to the fragmented and contentious authority of the state, thus arguing that we must unbundle the state's multiple and competing authorities, each endowed with their own agenda, and interests. For instance, Hagmann & Péclard (Reference Hagmann and Péclard2010) propose the notion of ‘negotiated statehood’. This framework argues that departing from the conjoined influence of material resources, which vary in time and space and are distributed unevenly among actors, and symbolic repertoires, existing or invented, is useful to understand the domination and imposition of political authority. Another example is Migdal's state-in-society approach (Migdal Reference Migdal1997, Reference Migdal2001) which contends the crucial importance of disaggregating the state to comprehend the practical consequences of its fragmented and contentious authority. It draws attention to how the state's strategy is always altered or transformed by various actors throughout the implementation process.
This theoretical standpoint draws attention to the gap between central directives and their implementation in local settings in a context of limited and contentious state power. It also insists on the importance of acknowledging how local actors always alter and negotiate central states’ strategies implementation. Moreover, it emphasises the dynamic character of social struggles which determine local practices and outcomes. Thus, considering the fragmented authority of a number of African states is valuable to refine the relational framework proposed by Buur et al. (Reference Buur, Nystrand and Pedersen2017) because it helps us to further specify actors’ interests and the nature of the power struggles they engage in. The remainder of this paper illustrates the utility of these theoretical insights for our comprehension of processes of land acquisitions, and their differentiated local impacts in Tanzania.
LEGAL FRAMEWORK GOVERNING LAND RIGHTS AND OWNERSHIP IN TANZANIA
In Tanzania, legally, all land belongs to the President: ‘all land … is public land vested in the President as trustee on behalf of citizens’ (Government of Tanzania 1995). There are three legal categories of land: general land, reserved land and village land. The Tanzania National Land Policy (1995) lays the general principles of land management, and has been followed by the enactment of several acts which constitute the Tanzanian land legislative framework.Footnote 2
Reserved lands include national parks, forests and marine environments that are parts of national efforts to conserve wildlife. Village land refers to land which is collectively owned by residents of a given village, and is under the authority of the village's council. General land is a residual category: everything that is not reserved or village land falls into general land, including land deemed available for investments.
Acquiring or leasing land usually involves central agencies and the Ministry of Land. In practice, the process is the following: villagers have agreed, willingly or not, to give up their village land rights, and this parcel of land has been transferred permanently to general land.Footnote 3 In addition, Tanzanian investors can obtain rights of occupancy on general land. Foreigners, however, can only be granted derivative rights of occupancy. Consequently, they must pass through the Tanzanian Investment Centre (TIC), and the TIC remains legally the ultimate holder of land rights on their behalf.
THE NEW CENTRAL'S STATE STRATEGY ON ‘PRODUCTIVE’ INVESTORS
Since the election of President John Pombe Magufuli in October 2015, the need to welcome productive investors has been omnipresent in Tanzanian news. Investors who have acquired land and failed to develop their assets are publicly criticised by the central government in the national media. They are depicted by the current Minister for Land, Housing and Human Settlements as Machiavellian ‘tycoons … hoarding big portions of land’ (Daily News 20.8.2016), thereby impeding the economic development of the country. In a strong statement, he also declared:
We will seize all the land pieces, which were legally acquired but had never been developed or are now used other than the prime-purpose … We cannot allow our people to continue to suffer on the expense of investors … There are a few people who are not supporting what I am doing since I annulled their lands. This will not make me stop; I will continue freezing all the land, which have been abandoned by investors regardless of their status and nationality … (Daily News 29.8.2016)
The government's pledge to fight unproductive investors is also closely linked with the President's highly publicised vow to deter rampant corruption. In his first year in office, the President, nicknamed The bulldozer for his style of governance, made clear his position of no mercy on corrupt officials. In addition to investigating numerous state institutions and agencies, he sacked several senior officials accused of embezzling or mismanaging funds (Reuters New Agency 14.12.2016). Clientelistic practices are common in the land sector in Tanzania (Kamuzora et al. Reference Kamuzora, Ngindo and Mutasingwa2009), and several investors have acquired important parcels of land through their political networks for speculative purposes without necessarily having the intention of developing and operationalising their investment. Thus, this discourse against corruption incorporates the denunciation of unproductive investors.
Moreover, this central discursive position fulfils an important political mission of strengthening the government's legitimacy at both national and international levels. First, it carries an image of a strong government committed to protecting its vulnerable local and mostly rural population from unproductive investors. Second, it demonstrates its serious intent to use agricultural Foreign Direct Investments (FDI) to foster socioeconomic rural development. From this official standpoint, corrupt practices are clearly associated with economic inefficiency, and should be hampered. Finally, it creates a climate of fear among investors in the country. The President is perceived as efficient, but also impulsive and unpredictable. Consequently, investors who have undeveloped assets are afraid of losing their land rights.
In sum, this discourse illustrates the government's commitment to increase its control over land investments. To achieve its objective, the central state has recently implemented institutional changes. The next section reviews how one of these changes, the creation of overlapping institutions to manage land investments, is affecting local practices and outcomes of land deals.
REVERSING THE FLOW OF POWER IN A CONTEXT OF LIMITED STATE POWER
A Potemkin state investment agencyFootnote 4
The Tanzanian Investment Centre was created in 1997 by the Tanzania Investment Act. It is largely an investment facilitator and it aims at encouraging investments in the country by attracting foreign investors and promoting Tanzania as a business-friendly country. In addition to offering fiscal advantages to investors, the TIC brands itself as a ‘one stop shop’, offering investors support and facilitating obtainment of all necessary licences and permits. Also, the TIC has created a land bank which in theory allows the agency to match available land to an investor's needs. The parcels of land included in this land bank are generally claimed to be free from local conflicts because the land has already been transferred from village land to general land (TIC Officer 2016, Int.).
However, since the TIC's land bank is incomplete, the process of finding land for investors is in practice reversed. Investors first identify land they want to acquire, and then use the TIC to formalise their access to this land. Therefore, the process is political: the TIC can be used to legitimise politically driven land transactions. For instance, in Rufiji district, the TIC was directly involved with three foreign investors. In these three cases, land acquisitions were characterised by a top-down process involving political interference from Tanzanian top officials.Footnote 5
Finally, the TIC lacks the capacity to play a role in following-up and monitoring land investments. There are only two researchers in charge of evaluating the viability of investment projects and ensure monitoring (TIC Official 2016, Int.). Practically, it is an impossible task given the size of the country and the number of investment projects approved or under evaluation in Tanzania. Unsurprisingly then, the TIC failed to follow-up on Rufiji's investors, and currently all these investments are non-operational.
The new President's election to office brought two changes regarding the TIC. First, he revoked the appointment of the TIC's executive director who had been appointed by the previous President in 2013. Second, the President created a new Land Investment Unit (LIU) under the purview of the Ministry of Land. Officially, the decision was motivated by the desire to remediate the TIC's lack of efficiency.
The Ministry of Land is a key institution regarding land management in Tanzania because it is the issuing authority of land titles. Officially, the relationship between the TIC and the Ministry of Land should be collaborative since the TIC needs to involve the Ministry of Land to get derivative land titles for foreign investors, and the Ministry of Land is used to transfer parcels of land available for investment to the TIC's land bank. Since the Ministry of Land is responsible for taking back land from commercial investors who have failed to develop their assets, it plays a direct role in managing available parcels of land for commercial development, but was delegating this responsibility to the TIC.
However, with the creation of the LIU, the Ministry of Land now intends to directly handle investment projects. In short, the TIC's failure to follow-up and monitor investors is depicted as the cause of the commercial underdevelopment of several pieces of land available for investment throughout the country. Therefore, instead of transferring land confiscated by the Ministry of Land to the TIC, it has been decided that the Ministry of Land will now manage by itself these parcels of land available to investors (Official, Ministry of Land, LIU, 2016 Int.).
Thus, the creation of the LIU implies an important institutional overlap since the new unit has a similar mandate to the TIC. In addition, the LIU allows the President to interfere directly and use his discretionary power for land allocation to investors. For example, in October 2016, he allocated through the LIU 10,000 hectares free of charge to a Tanzanian investor, Bakhresa. This land was confiscated by the Ministry of Land from an unproductive investor in the Coast region. The decision appeared to have been personal and based on the President's perception of Bakhresa as an example of a productive investor. Speaking of the company, he stated: ‘you are the best example of businesspeople we want … the best tax payer and you have created many employment opportunities …’ (Daily News 7.10.2016).
How district officials capitalise on their local knowledge
Since the TIC lacks capacity, it depends on district authorities to assess investment projects because they have a much better understanding of the specificities of the local context. This relation of dependency has significantly empowered district officials, creating opportunities to foster their material interests.
All land acquisitions through the TIC in Rufiji district involved political interference. As explained earlier, it means that instead of the TIC allocating available and suitable parcels of land from its land bank to investors, investors instead used the TIC to formalise their rights on desired land. In this process, the TIC must collaborate with district authorities and it usually mandates them to demarcate the requested land, inform, consult and secure villages’ approval. This process generated revenues for district officials because of their involvement in several steps – land identification, land mapping, organisation of consultations with villagers, etc. – for which they requested a financial contribution from investors. Consequently, district authorities had a direct material interest in satisfying investors by ensuring the concretisation of land deals with the concerned villages.
In practice, it led district officials to disregard village authorities on land rights. Even though they did consult the villages, villages’ approval or refusal was in practice inconsequential. In all three acquisitions that I investigated involving the TIC in Rufiji district, the villages concerned – both village leaders and villagers – were opposed to and refused the land deals, but the land transfers were nonetheless approved and completed. Yet, as my analysis shows, excluding village administrations is likely to create conflictual dynamics at the local level, and potentially may impede investors’ capacity to become operational.
First, this way of proceeding generates significant confusion at the local level. Some villagers are not even aware that their land has been leased to an investor. Even when they are informed, because they were excluded from the negotiation, they lack knowledge about the investor and it prevents them from holding investors accountable and from monitoring expected benefits. For instance, villagers and local leaders are usually denied access to the contract and the terms of the agreement. Even basic information, such as the name of the investor and his contact details, are kept secret. In fact, villagers use surnames to speak about these investors, which illustrates their lack of information.
Moreover, it creates tensions between local populations and investors. An investor that I will name ‘Investor A’, for purposes of anonymity, is an illustrative and representative case in this regard. The company acquired 3,060 ha of land for maize and rice production. Although the village was not consulted before the TIC's approval of the land transfer, the village eventually was convinced that the investor's presence would contribute to village development. The investor promised several things to villagers: to provide material for schools and to build classrooms, to expand the health centre, to ensure access to electricity and to provide employment to villagers (Village leaders, Rufiji 2016 Int.). None of these promises were fulfilled.
A first problem emerged with the compensation process. Since the land acquired was already occupied by farmers, the arrival of the investor required their displacement. According to Tanzanian law, in such cases, farmers are entitled to receive compensation for their land. Displaced farmers were offered a compensation amount in 2005. The company gave the compensation sum to the TIC, and the TIC delegated the responsibility to compensate farmers to the district. Thus, the district on behalf of the company offered farmers TSH45,000 (US$20.14) for five acres without any compensation for the crops they had already planted or their mango trees. This offer was so low that most farmers were insulted and refused. Nothing happened afterwards; apparently, the district kept the money and it was the end of this compensation process.
This led to tension between the company and the villagers. On one side, the villagers resented the company for what they perceived as unfair treatment. On the other side, the company claimed that it was a victim as well since they gave a fair amount to compensate farmers directly to the TIC (Guard investor A, 2016 Int.). Although the role played by the district and the TIC in this compensation is unclear, the fact remains that compensation issues negatively affected the relationship between investors and farmers.
The process of starting agricultural production also generated conflicts. Since the villagers were not consulted and properly informed about the land acquisition, boundaries between the land owned by the investor and the villagers remained blurred. In the case of Investor A, this confusion over boundaries created practical problems. After acquiring the land, the company started its operations and planted a small area with maize – between 300 and 700 hectares. However, their first harvest made a loss because, according to the company, about two-thirds of their harvest was stolen by the villagers. The villagers denied this accusation. For them, the company had pretended to cultivate its land but was in fact cultivating the plots that had already been prepared by the villagers. So, the villagers were not stealing from the company but harvesting their own crops. Although what really happened is uncertain, the conflict escalated to the point where one person working for the company shot some farmers apparently caught stealing crops, and one of them got a bullet in the leg. Subsequently, the company lost all local credibility. These combined tensions drove the investor to stop production, and the land was abandoned one year later, in 2006. Farming equipment was left in the village, but the investor himself never came back.
As this discussion illustrates, the tendency of district officers to exclude villages from land acquisition negotiations tends to generate conflicts between investors and villagers. Villagers, being ignored, lack information about the investor and the location of his investment, thereby creating local conflicts over boundaries and usage of the land. Also, because compensation processes are often problematic, villagers tend to resent the company for not compensating them properly even though the company may not be solely responsible for the insufficient compensation. In the end, it may have important consequences for investors, and impede their capacity to become operational, as it was the case with Investor A.
DISTRICT OFFICIALS: GAMING THE SYSTEM
The process of centralisation of land management with the creation of the Ministry of Land's LIU has not only disempowered the TIC, but it has also weakened the position of district officials. Indeed, this centralisation allows the central state to bypass district authorities and proceed with investors directly in a typical top-down process. Consequently, district officials lost the upper hand in the game, and this institutional change significantly reduced their capacity to benefit materially from new investors. However, as I claim in the next section, district officials were very innovative in finding new ways of fostering their material interests, especially in areas where the state has limited infrastructural power.
A new strategy: the protection of unproductive investors
Even though its new strategy allows the central state to bypass district authorities regarding land acquisitions, it still relies on district officials to follow-up on the productivity of existing investors. This relation of reliance allows district officials to preserve their power with current investors. In addition, the President's pledge to fight unproductive investors has created a climate of fear among investors. Consequently, some investors who have undeveloped assets are afraid of losing their land rights. This situation gives district officials a new and interesting edge with these unproductive investors.
Using two examples in Rufiji district – Investor B and Investor C – I draw attention to how the district officials' discourse on investors’ operationalisation of projects is contradicted by the reality on the ground. I argue that this discrepancy can be explained by considering district officials’ strategy to benefit materially from investors. In fact, partnering and protecting unproductive investors from dispossession by the central state gives them access to new opportunities for accumulation, and compensates for their exclusion from land acquisition processes.
The case of Investor B
Investor B acquired 3,000 ha in Rufiji in 2004 to grow sugarcane. The company is local and owned by Tanzanians. It acquired the land directly from the village and this acquisition forced the displacement of about 50–60 families. The district was involved in a compensation process that was highly contentious.Footnote 6 Furthermore, the company did not respect the development promises it made, never started operations, and eventually disappeared from Rufiji. Consequently, several farmers felt they were treated unfairly and the company now has a bad reputation locally.
However, recently, the owner of the company came back to Rufiji district. According to district officials and local leaders, Investor B has shown a renewed interest and is now committed to starting its commercial operations by clearing its land to prepare for cultivation. In short, the company is depicted as a promising investor who will soon become productive.
Yet, villagers show serious doubt about this official narrative. For them, Investor B is primarily motivated to display an appearance of being active because it is afraid of the President's threat regarding unproductive investors. Thus, they attribute Investor B's motivation to the fear of losing its rights to the land. My field observations tend to support the villagers’ interpretation. Instead of using tractors, as would be expected in this type of commercial agricultural venture, Investor B made an agreement with the village chairman to hire youths to clear its land manually. The least we can say is that this way of proceeding for a such a big parcel, 3,000 hectares (7,400 acres), is far from optimal. In the words of one farmer: ‘they [the youths engaged by the Chairman on behalf of the company] use primitive tools such as panga [machete]. For us, they [Investor B] will do nothing. It is just a show’ (Farmer, Rufiji, 2016 Int.).
Although Investor B's real motivations remain subject to speculation, the significance of this example lies in showing the process through which the company has established partnerships with district officials and local leaders to protect itself from dispossession by the central state. When I questioned district officials about Investor B, they depicted it as being committed and operational. I interpret this discourse as one that aims at protecting the company, and district officials’ interests with it. Without a doubt, officials are aware, as are the villagers, that Investor B's current clearing land operations are seriously contradicting its claim and commitment to become commercially productive.
The case of Investor C
Investor C provides another illustrative example. The company acquired about 5,000 ha in Rufiji for rice cultivation in 2015 through the TIC. The process of this deal was particularly top-down and secretive. At a local level, the process of acquisition was so obscure that the villagers and village leaders do not know how the company acquired this land (Land Rights Monitor, Rufiji, 2016 Int.). The actual location of the land acquired is surprisingly imprecise in official documents: the affected villages are not identified (TIC Officer, 2016 Int.).
Operationalising agricultural operations on this specific piece of land is problematic for many reasons. First of all, the land acquired is in a very remote and undeveloped area of the district. To reach it, which I did, one needs to cross a small artisanal bridge over the Rufiji River and then drive about 28 km on a narrow and sandy road. Clearly, lack of infrastructural development is challenging for any investor who wants to develop commercial agriculture. There is no electricity, no water, and the roads are not sufficiently developed to allow regular transportation.
Moreover, before being able to cultivate this land, it would require a fair amount of investment. As I saw when I visited this investment site, the land is uncleared and wild bushland, except for a tiny section where the investor brought a container, dug a well, and built the guard's house. Since Investor C's land has shared boundaries with the Selous National Park, the presence of wild animals – such as elephants, lions, monkeys, hippos, etc., is likely to disrupt agricultural activities. Unsurprisingly then, the investor has not done much except for enlarging a few kilometres of the road leading to his land, and establishing a precarious base camp. Moreover, it appears that all operations have stopped since February 2016.
Despite these facts, district officials who I interviewed in November 2016 insist that this investor is productively engaged in this agricultural venture. According to them, the company is committed to being operational soon, and is currently active in developing the road to reach its land to start agricultural production (District Official, Rufiji, 2016 Int.).
This contradiction between my field observations and district officials’ discourse is striking. I came to the conclusion that protecting this investor from the central state's dispossession threat may explain this discrepancy. District officials are more concerned about protecting their interests with this investor than by his eventual capacity to become productive. This would explain why they continue praising the investor's commitment despite being aware of major challenges to develop agricultural activities on this land anytime soon.
To conclude, the central state has put in place a new strategy to increase its control over and foster productive land investments. In Rufiji, this has created an institutional overlap which has sidelined district elites and reduced their opportunities for accumulation with investors. However, since the state has weak infrastructural power, it still relies on them to monitor and follow-up on investors. I have demonstrated that district officials have used this power to foster their own material interests.
Besides this, the centralisation strategy of the central state has also led it to monitor closely some local state agencies dealing with land investors in Rufiji. The next section delves into an analysis of the local impacts of this new central monitoring. I argue that it might have a pervasive local effect. Instead of protecting vulnerable populations, it fosters institutional innovations that aim at protecting local officials’ opportunities for accumulation with investors.
THE NEGATIVE EFFECT OF CENTRAL MONITORING: NURTURING LOCAL OFFICIALS TO INNOVATE AND SAFEGUARD THEIR MATERIAL INTERESTS
Ambiguous at best, illegitimate at worst: how an agency can reinvent itself as a land broker
Rubada, a state agency that redefined itself and its mandate in one of the most targeted regions for land investments, provides a fascinating illustration of the consequences of the state's lack of capacity on policy implementation. To comprehend the complexity of the relational prism through which Rubada interacts with other state actors, investors and local populations, historical contextualising is crucial.
Rubada, or the Rufiji Basin Development Authority, is a national agency that was established in 1975 by Parliament Act No. 5. Broadly defined, its mission was initially to promote, regulate, coordinate and facilitate all development activities in the Rufiji Basin. Rubada was created amid an optimistic development buzz in the Rufiji region with the ambitious Stiegler Gorge Project. It is worth noting that this narrative about the tremendous potential of the Rufiji Basin for socio-economic development has been part of the national narrative since the colonial period. Numerous ambitious development projectsFootnote 7 have been undertaken in the region in the hope of taming the Rufiji River for commercial irrigation or generating hydro-electricity. Rubada's initial mandate was thus to coordinate Tanzania's partnerships with international donors and institutions for the development of the Rufiji region (Hoag Reference Hoag2003: 183). However, in the end, all these ambitious development projects failed. The implementation of the structural adjustment programme and economic liberalisation at the end of the 1980s marked the end of this era of large-scale projects with abandonment of the Stiegler Gorge Project.
Consequently, Rubada's mission became unclear, but the agency was not dismantled. The turn towards privatisation at the end of the 1990s, and the focus on attracting investors to promote agricultural development provided new opportunities for Rubada. Echoing what neoinstitutionalists call ‘institutional stickiness’, Rubada's officials found a way of reinventing their agency and justifying its existence. In fact, they broadened its mission (Rubada 2013: 4). A wider conception of Rufiji development was put forward, in which the focus on the Rufiji River was replaced by a focus on land. Fostering land investments and attracting new investors in Rufiji region became, in practice, Rubada's new mandate. Yet, this reorientation was never properly officialised by the central government. Consequently, to this day, the agency lacks legitimacy at local and central government levels. At worst, Rubada is perceived as illegitimate; at best, ambiguous.
At first, Rubada managed to become a very attractive agency for investors since it acts as a broker to deal on their behalf with villages, district officials and the different ministries. The agency was relatively successful in branding itself as the turnkey agency for land investments in Rufiji district until the 2015 corruption scandal with investors. In April 2015, senior officials of Rubada were put under investigation over the disappearance of TSH 3.2 billion of development funds (roughly US$1.5 million). According to the report, investors’ money was not channelled through Rubada accounts: ‘the officials received TSH2.7 billion from investors who wanted to work in various projects in the basin, but only TSH714 million was tracked to Rubada accounts’ (The Citizen, 7.4.2015). Swindled investors never got the land they paid for, and their money simply vanished.
Subsequently, the relationship between the central government and the agency has become arduous as top political officials have politically ostracised Rubada. For instance, its number of employees has diminished drastically: from about 500 employees in the 1990s, the agency has now less than 50. Moreover, it has received no funding at all from the central government since 2015. Rubada's offices in Dar es Salaam have fallen into decrepitude, a situation which is a telling contrast with the TIC's and Ministry of Land's LIU newly renovated offices.
Nonetheless, Rubada is still a very active actor in land investments in Rufiji district. The agency is involved in about 20 investment projects. While most of them are still in the preliminary phases of land acquisition, some are at the Ministry of Land, waiting for the completion of the land transfer process (Rubada Liaison, 2016 Int.). Although the government announced in September 2017 its decision to dismantle Rubada and to transfer the remaining employees under the Ministry of Agriculture, the agency is likely to continue to be active in Rufiji to honour and implement its contractual obligations with investors (Rubada Liaison, 2016 Int.). Therefore, the President and the Ministry of Land must still tolerate Rubada's presence.
To deal with Rubada, the central government has directed a close monitoring of Rubada's activities by the Ministry of Land. For instance, the agency must be accompanied by district officials for every meeting with villagers, and their presence is required during negotiations and villages’ general assemblies. In addition, the Ministry of Land has recently sent delegates into every village which had signed a contract involving Rubada to validate the villages’ acceptance of land investments (District Official, Rufiji, 2016 Int.). This central oversight aims at ensuring that the agency is carrying out its activities with investors properly.
Central monitoring and subversion
Rubada officials have an extensive local knowledge of Rufiji district – the agency has been active for over 40 years in the district – and therefore, they do not rely on district officials as the TIC does. In practice, when it comes to land acquisitions, Rubada arranges meetings directly with village leaders. Village leaders are thus always involved in the negotiation and can establish their conditions for the lease of the land. Yet, while such a process empowers village leaders, we should be careful before claiming that it is more inclusive for villagers.
My field observations led me to believe that each village in Rufiji district has its own power dynamics, and the result depends on the leaders’ relations with the villagers and the nature of consultative mechanisms. I have noticed that some leaders favour very transparent processes and villages assemblies are called to openly discuss the investor, the development conditions to be met, and the possible advantages of this investment for the community. The discussion and the final decision to lease land or not is thus taken in a concerted and relatively democratic way. However, in other villages, I noticed that leaders are using their position of authority to negotiate privately with Rubada and investors. They keep these transactions secret and refuse to answer villagers’ questions during meetings. Therefore, these villagers are often not aware of their land having been sold, have no information about the investor, and don't know the terms of the contract. In practice then, for land acquired in this fashion, it is highly likely that the local outcomes of land deals will be similar to the ones observed with the TIC or LIU.
Before being centrally monitored, Rubada used to collaborate with the district to make sure that land was leased to investors, with or without villages’ approval. For instance, in one specific village that I have visited, the leaders were approached by Rubada and an investor about three years ago. The collective decision was to refuse the investor, for various reasons. First, the land the investor wanted to acquire was larger than the land available for investors in the village's land-use plan. Second, the village was suspicious of Rubada. They distrust the agency: ‘We don't recognise Rubada's authority and we don't want to deal with them. From what we know, Rubada's mandate is to help developing the Rufiji River but it is not what they do. They tend to partner with investors and to not protect people’ rights’ (Local leaders, Rufiji, 2016 Int.). Yet, these leaders learned accidentally while they were at the district office that their land was leased to this investor despite the village's clear opposition. It means that Rubada ignored the village's decision, and colluded with district officials to proceed with the land transaction.
However, the close monitoring of Rubada's activities has led the Ministry of Land to revise all Rubada's land deals in Rufiji district. In some instances, this led to acknowledgement of the illegality of some land transactions and to the recognition of villages’ authority and land rights. The case of this village illustrates the concrete consequences of this central follow-up and monitoring. When the villagers learned that their land was leased despite their refusal, they were shocked. They wrote a letter to the Ministry of Land in 2013 to inform it that their rights were denied by the district and Rubada, and that they, as villagers, consider this transaction illegal. The villagers themselves never got any follow-up on this. Nonetheless, according to my interviews with both Rubada and the district, the village's refusal was acknowledged by the Ministry of Land during its revision of land transactions involving Rubada. Consequently, the transaction was invalidated, and the village finally kept its land rights.
Yet, this outcome appears to be exceptional. In fact, the Ministry of Land's first motivation is to monitor Rubada to avoid a new corruption scandal with investors. Most of its follow-ups on land deals involving Rubada were procedural and conducted only with villages’ executive committees, and selected leaders. Thus, the monitoring process itself does not deal with the issue of local leaders, who have sold village land on villagers’ behalf without properly informing or consulting them. In short, the process overlooks consequences of asymmetrical power within villages. Nevertheless, the new monitoring by the Ministry of Land of Rubada puts pressure on the agency to no longer ignore decisions made by villagers. It has forced Rubada to innovate to protect its interests with investors in a context where the agency must secure villages’ consent before being able to proceed with the land transfer for investment. Therefore, this monitoring process that at first may seem to encourage accountability for the people has in fact pushed the state agency involved to innovate to protect its own interests.
In practice, the agency started to use its position of power to exploit village leaders’ lack of knowledge, and of education regarding their land rights and legal proceedings to advance its own interests.Footnote 8 Since Rubada's main source of revenue comes from its services as a broker for investors, it has a direct interest in satisfying investors and finding the land they request for their investments even though it might involve lying to village leaders. For instance, they will agree with the villagers to certain conditions but they will not include them in the official contract with the company. Numerous villagers have told me that one trick used by Rubada is to write down a bilingual version of the contract. Yet, the content of the Kiswahili contract will include villages’ requests while the English one – which is used as the official and legally binding contract – will not. In this way Rubada can secure villages’ approval by agreeing to all their demands regarding the investor's acquisition, without imposing any constraining development conditions on the investor.
In addition, Rubada is not only tricking villages to secure its interests, but it is also deliberately misleading central authorities on investors to facilitate land acquisition processes. As explained earlier, Rubada is struggling for its financial and political survival. Facilitating land acquisition for companies is its only way to generate new revenues. Therefore, Rubada is inclined to support investors, even though it involves misinforming the Ministry of Land. For example, Rubada is officially working with two different companies that are looking to acquire land for sugar production: Investor D and Investor E. Yet, these two companies are in fact the same. Rubada and the company's director were afraid that they were asking for too much land at once and the Ministry of Land might refuse to allocate such an impressive area to only one investor—it represents over 25,000 hectares. This is why they created two businesses to avoid suspicion and facilitate the process to obtain land titles (Rubada Liaison, Rufiji, 2016 Int.). The district was complicit in this case. Moreover, the fact that the company has been struggling financially, and does not have the financial capacity to develop such an ambitious project does not appear to be a concern for Rubada or the district, although it should be since it might impede the investor's capacity to be productive. This shows that the central state's strategy to foster productive investors has not been internalised by these actors. Their interest lies instead in the prospect of short-term material gains from their partnership with new investors.Footnote 9
Finally, as this discussion shows, central monitoring has direct effects on the way a local state agency conducts its activities regarding land investments. However, empirical investigation of these effects is necessary to understand how bureaucrats transform and alter these policies during the implementation process. As illustrated in this case, local officials innovated to promote their interests with investors, even though it involved misleading and tricking both village and central authorities.
Moreover, this discussion highlights the fact that including villages' administrations within processes of acquisition is not sufficient to avoid conflictual dynamics between villagers and the investors. This finding points out the crucial importance of acknowledging power dynamics within villages – more specifically leaders’ accountability to villagers – to analyse the nature of the potential relationship between investors and villagers.
CONCLUSION
My findings contribute to recent works that seek to explain land acquisition processes, how they unfold and their distinctive local impacts. Building on the relational analytical framework proposed by Buur et al. (Reference Buur, Nystrand and Pedersen2017), I have argued that the fragmented and contentious authority of a number of African states can impact power relations in the field of land deals. More specifically, I investigated the impact of state intermediaries on relations between investors and local populations, contending that unpacking local bureaucrats as a group of actors is key to analysis of flows and relations of power within the state.
Based on my empirical findings in Rufiji district, I first showed that the central state's weak infrastructural power may reverse the flow of power, empowering local bureaucrats over their counterparts. Second, I demonstrated that, rather than protecting vulnerable populations, central monitoring might foster institutional innovations that aim at protecting local bureaucrats’ opportunities for accumulation with investors.
Lastly, my analysis reiterates the importance of recognising that the hegemonic project of many African states is contentious, dynamic and incomplete (Ferguson & Gupta Reference Ferguson and Gupta2002: 982). Furthermore, it shows that investigating empirically how local actors transform central structure processes remains highly relevant to our comprehension of governance practices in Africa (Mandani Reference Mandani1996).
