Large-scale investments in natural resources (extractives as well as agriculture) can help transforming African economies by accelerating economic growth, creating jobs and strengthening links between local economies and the global economy more broadly. However, they often end up violating rights. This, in turn, may lead to social protests and political instability. This paper develops an analytical framework for how to analyse the implementation of large-scale investments into natural resources. It focuses on the triangular relations between investors; local populations; and ruling elites. The framework points to the outcomes of these triangular relationships as 'reciprocal exchange deals' between investors and local populations, 'compatible interests' between ruling elites and investors, and 'productive social relations' between local populations and ruling elites. We show that in order to understand why some investments are more successfully implemented than others, it is necessary to grasp the politics behind an investment. The paper furthermore explores the conditions under which investments can be implemented without violating the rights of local populations. The paper is based on a review of relevant literatures and is linked to ongoing empirical studies of large-scale natural resource investment cases in Mozambique, Tanzania and Uganda related to gas/oil, mining and agriculture.

Tanzania has been very actively courting investors, arguing that they will ensure the country's agriculture modernization and economic development. Yet, the implementation of agri-food projects has revealed highly problematic, as many investors have failed to develop their land and start operations, thereby effectively dispossessing local communities from their rights to land without fulfilling any promises of economic development.

Based on a seven-months political ethnography conducted mostly in Tanzanian villages, this paper addresses this issue by asking: what effects do recent agri-food projects have on local political dynamics in Tanzania, and inversely, how do local dynamics influence the implementation of these projects?

Focusing on Rufiji District—one of the most targeted area for new agri-food investments in Tanzania, I analyze the political reconfiguration of power relations associated to recent agri-food projects, highlighting power dynamics between government, ruling elites, investors and local populations. I contend that the acquisition process of land for agri-business ventures is key in understanding their local impacts, and their failure to become operational. I show that central institutions' knowledge and control over investors is fragmented and incomplete, and I argue that this lack of capacity allows government authorities and leaders to play an active role in subverting central policies, partnering and covering-up for unproductive investors in order to keep 'eating'. Moreover, I argue that top-down directives and governance are undermining villages' capacity and power to hold investors accountable. Finally, by comparing land acquisitions in nine different villages, I draw attention to the importance of differentiating local political dynamics.

Although countries in Southern Africa have traditionally been large exporters of tobacco leaf, important changes have taken place in the past two decades: exports and the geographic footprint of tobacco are expanding and non-equity forms of production (mainly in the form of contract farming) are displacing direct forms of investment. Taken as a region Southern Africa is the third largest exporter of tobacco now and the tobacco contract farming schemes are among the largest in the continent.

This paper explores how contract farming is allowing for intense land and labour mobilization and how it has facilitated the integration of direct producers to commodity chains, without requiring the transfer of property over land. By comparing Southern African experiences and focusing on the Mozambican case, the paper argues that non-equity forms of production are more adaptable to contexts in which direct agricultural producers have access to land and were corporate investors in land face political contestation and the risks associated with informal land markets. Non-equity forms of production raise fewer red flag and less attention from the region's states. However, while contract farming seems to unleash an extensification of production it does entail the risk of reinforcing exploitative land and labour practices and in the absence of targeted state involvement it has limited capacity for enhancing productivity and effective spillover effects on the rest of the economy.

Tanzania has emerged as one of the leading gold mining countries in Africa. New urban areas develop around the mines. Still mining give a small contribution to GDP. Huge natural gas deposits are discovered, and have attracted the largest foreign investment ever in Tanzania. But so far little of inclusive development have been generated. The paper analyses how and to what extent reforms for deepening democratic governance are leveraging the extractive sector, for inclusive growth, through transforming and shaping the institutional structure (formal and informal) and organizational capacities to deal with competing interests (personal versus public; intra elite conflicts; party versus collective; local versus national; and national versus international). In the analysis of the underlying power structures on national and local level, we link to the expanding literature on "political settlement theory". This lens allows an analysis of the contending interests which constrain and facilitate institutional and developmental change.

Amongst the reasons attributed to the limited outcome in terms of inclusive development and local content, include inadequate political leadership capacity and political systems for reconciliation and coordination of competing local, national and international claims on the natural resources. Political measures to ensure inclusive benefits for current and future generations are not guaranteed through political settlements. At the technocratic level, weak technical leadership capacity and corruption are factors associated with the failure at the executive level to intervene and to link investments to local development plans, financing and implementation.