The Economics and Politics of Local Content in African Extractives: Lessons from Tanzania, Uganda and Mozambique

Abstract

Extractive foreign direct investment (FDI) is heralded as the new development opportunity in Africa. A key precondition for FDI's contribution, however, is that foreign investors create ‘local content’ by linking up to the local economy. Consequently, African host governments are contemplating ways in which they can promote local content. This paper examines local content policies and practices in three African countries – Tanzania, Uganda and Mozambique – all countries with huge expectations for extractive based economic development. It is found that in spite of high ambitions and strong expectations, local content is limited, shallow and inefficient. The paper explores why local content apparently is so difficult to achieve in these African countries. It is argued that conventional economic explanations, focusing on market failures and weak institutions, are partial at best and therefore must be complemented with political explanations. Hence, it is proposed that local content practices in the three countries can be understood partly as the results of ruling elites’ efforts to build and maintain stable political coalitions.

Keywords:

• extractives
• Africa
• FDI and MNCs
• linkages and local content
• political economy

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Michael W. Hansen is an associate professor at the Center for Business and Development Studies, Copenhagen Business School. His research focusses on the strategies and impacts of MNCs in developing countries.

Lars Buur is an associate professor at the Institute for Society and Globalization, University of Roskilde. His research focusses on the political economy of development, in particular as it relates to productive sectors in southern Africa.

Anne Mette Kjær is an associate professor at the Department of Political Science, University of Aarhus, Her research focusses on governance and politics of productive sectors in Africa.

Ole Therkildsen is Researcher Emeritus at the Danish Institute for International Studies, Copenhagen. His research focusses on natural resources and development in Africa.

Notes

¹Focus will be on large-scale investments in energy (oil, gas and coal) and minerals. This is a subset of extractive industries involved in natural resource sectors, which also include production related to land (i.e. forestry, agriculture) and to fisheries. See Morris et al. (2011a, p. 8) for a recent classification.

²We focus on Tanzania, Uganda and Mozambique for three main reasons. Firstly, the three countries formed part of the RECOM study commissioned by Danida and UNU-WIDER that was finalized in 2013 where explorative fieldwork was carried out as part of the study (see Buur et al., 2013 for a summary of the findings). Secondly, the three countries have been at the forefront of the natural resource boom experienced over the last decade in Africa with very high volumes of FDI investments. Finally, the authors have in-depth knowledge about the three countries as they formed part the Elites, Production and Poverty research programme exploring the political economy of productive industrial policies (see Whitfield et al., 2015) and the Successful African Enterprise Project (SAFIC) (see http://www.cbs.dk/en/research/cbs-research-projects/major-research-projects/successful-african-firms-and-institutional-change-safic).

³‘New’ in comparison with other African countries with longer traditions for large scale natural resource extraction, countries such as Nigeria, Angola, Zambia or Botswana.

⁴Rents are payment to a factor of production in excess of its opportunity costs i.e. beyond normal profits. Rents can be obtained by the firms involved in the activity (e.g. if they have monopoly) or they can be obtained by external stakeholders such as governments, employees, politicians and bureaucrats. Rents can be obtained through legal means such as tariffs, taxation, royalties, licenses and contracts, but they can also be obtained through favoritism or corruption (see Altenburg and Melia, 2014; Kaplinsky, 2005; Khan, 2000a; 2000b). Rents can both be pecuniary such as royalties, tax income, license fees or bribes (see, e.g. Barma et al., 2011; Karl, 1997) or they can be non-pecuniary such as entitlements, rights, options or concessions.

The activity by firms or individuals to obtain rents is sometimes referred to as ‘rent-seeking’. It is debated whether rent-seeking activity always is ‘directly un-productive’ for society (so-called DUP rent-seeking (Bhagwati, 1982)) or whether it actually may benefit under certain circumstances, e.g. by providing political stability (North et al., 2012, ch. 1; Khan, 2000a).

⁵Formally, local content can be defined as ‘value addition in a local country (by local staff, local materials, local services and facilities) rather than in terms of ownership of the company performing the value added activities' (Kazzazi and Nouri, 2012, p. 2165).

⁶A recent – allegedly – conservative estimate suggests that local content requirements of all countries affect 2 per cent of global trade across all sectors and reduce it by USD 93 billion annually (Hufbauer and Schott, 2013).

⁷‘Political settlement’ is shorthand for the set of institutions and power relations that characterize social order in a particular country, and that create benefits for different classes and groups in society in line with their relative power (Khan, 2010). At any particular point in time, the political settlement reflects a kind of equilibrium, as the distribution of benefits reinforces the distribution of power in society.

⁸The three countries are relatively rich on resources; however, they are also only in the early stages of resource based development. A commonly used way of describing the importance of endowments is IMF's ‘fiscal dependence on natural resources’ index. For the years 2005–2010, the IMF categorized none of the three countries as ‘fiscally dependent on natural resource exports’. For coming up with this label, the IMF uses two criteria, of which, for the period 2005–10, Tanzania surpassed one (more than 25 per cent of export revenues come from extractives), but not the other (more than 20 per cent of government revenue derived from natural resource extraction). Mozambique and Uganda were significantly below both thresholds, and did not make it onto the runners-up list for the latter category (the list of ‘other’ resource extracting SSA countries) (IMF, 2012, pp. 61–62).

⁹See ‘Mozambique turns 40, high on gas prospects’, Reuters news, June 25, 2015, accessed at http://www.reuters.com/article/2015/06/25/us-mozambique-politics-idUSKBN0P525020150625 in August 17, 2015.

¹⁰‘China's CNOOC wins $2bn Uganda oil field contract, BBC News, September 26, 2013 accessed at http://www.bbc.com/news/business-24279582 in February, 2015.

¹¹In April 2014, the Ministry of Energy and Minerals (MEM) published a draft local content policy (http://www.mem.go.tz/wp-content/uploads/2014/05/07.05.2014local-content-policy-of-tanzania-for-oil-gas-industry.pdf). Compared to the rather non-binding nature of previous local content provisions, the draft local content policy calls for local content throughout the extractive process. MNCs must give preference to local employment, outline succession schemes and local content is becoming compulsory in all procurement. Failure to comply with these provisions will jeopardize MNCs’ ability to participate in concessions and obtain licenses. As of July 2015, the policy is still not completed.

¹²See, e.g. http://www.ey.com/GL/en/Services/Tax/International-Tax/Alert--Mozambique-introduces-new-tax-regime-and-incentives-for-the-Mining-Industry; http://www.vda.pt/xms/files/Newsletters/2014/Petroleum_Law_translation.pdf; http://www.plmj.com/xms/files/newsletters/2015/Marco/Specific_Rules_on_Taxation_and_of_the_Mining_Activity.pdf; http://www.law360.com/articles/645363/understanding-mozambique-s-new-oil-and-gas-regime.

¹³Horizontal linkages are linkages to firms in the same industry; vertical linkages are linkages to firms upstream and downstream in the firm's value chain (e.g. suppliers or distributors).

¹⁴Interviews in Kampala, June 2013 all expressing this opinion: Representative of the Uganda Chamber of Mines; representative of Irish Traidlinks; a member of parliament, an employee with Total; employee with Tullow Oil; Interview with sub-contractors; donor agencies such as NORAD; and the IMF country representative, and World Bank representative, among others.

¹⁵Interview, Total Contract and procurement manager, Kampala, June, 2013.

¹⁶Own interviews with Tanzanian mining suppliers, Dar es Salaam, December 2014.

¹⁷The new President Nyusi who took over from Guebuza from 2015 is expected to create a platform for a more inclusive growth process. But to what extent this will be possible while still maintaining Frelimós control over economic opportunities undermining the political opposition is too early to say anything about yet.

¹⁸This is based on discussions with industry, state and donor representatives and researchers and consultants working with natural resource extraction from Maputo, during January 2015.

¹⁹Rents related to royalties and taxes are still of key importance to governments. In fact, it can be argued that such fiscal rents in some sense are more important in the short term for political settlement, as they provide hard cash at the president's disposal allowing for funding of cash carrots (stipends, travel allowances, new government jobs, etc.) as well as sticks (military & secret service expenditures). There is no doubt that local content offers a new opportunity for accessing and distributing rents, the potential of which is only beginning to be realized by politicians in East Africa. However, harnessing local content for political settlements is a more long-term and arduous process, demanding investments in policy development, infrastructures and institutions.

²⁰It should be noted that the rent issue displays different dynamics in different productive sectors (Whitfield et al., 2015). For instance, rents from extractives are derived to a relatively large extent from foreign firms and firms that operate as enclaves in the host economy. Moreover, rents from extractives typically are very high relative to cost of production (Barma et al., 2011; Fearon, 2005). Finally, a key feature of extractive rents is that they have a high degree of volatility. These specificities of extractive rents may influence the propensity of governments to extract rents, the way rent appropriation is organized and how rents interact with political settlements (Dunning, 2008).