Mining in Africa after the supercycle: New directions and geographies

is structured around examining four themes: (1) new geographies of investment and extraction; (2) new geographies of struggle; (3) national minerals ‐ based development; and (4) labour and livelihoods, for which we identify key trends during the supercycle and post ‐ supercycle and areas for future research and policy development.

regions previously deemed marginal or risky, including many parts of Africa (Humphreys, 2015).Artisanal and small-scale mining (ASM) also proliferated, particularly in the gold sectorestimated to provide direct income to over 10 million people (Hayes, 2008).Optimism about the prospects for minerals-based development grew among African policymakers (African Union, 2009), as economists identified in the supercycle a once in a generation opportunity for growth and poverty reduction through improved fiscal management of mineral rents (Collier, 2015) and minerals-based industrial policies (Morris et al., 2012).
Driven by changing patterns of economic activity and natural resource policy, this supercycle ended in the mid-2010s and a new, more complex period of commodity-market dynamics began.The euphoria of prolonged high prices, which quickly rebounded after the 2008 global financial crisis, spurred excessive investment, creating oversupply in multiple commodities and high levels of corporate indebtedness.In a severe period of economic distress from 2014 onwards, prices plummeted, corporates rapidly cut costs, and revenues dwindled for commodity-dependent governments.Even before the impacts of COVID-19, recovery had been faltering and uneven.Key mineral-energy industriesnotably coal and oilare widely anticipated to face declining long-term demand (International Energy Agency [IEA], 2019).However, there has been increased investor interestboth from private and state actorsin "clean energy" minerals like copper, coltan, and lithium, while gold prices have surged as investors seek a counter-cyclical safe-haven amid increased global economic uncertainty.
Notwithstanding these examples, the mining industry has faced difficult circumstances.Multinational mining companies have become more cautious, focusing on lowering costs and risks: in some cases shedding tens of thousands of jobs and reducing costs through organisational and technological innovation, while repositioning project portfolios away from more "complex" operating environments.This has important consequences for economic development and political change in mineral-rich African countries.Having often borrowed heavilyparticularly on infrastructureand expanded public spending during the boom, lower growth and resource rents have fiscally strained governments.The IMF (2019) observes a dualistic pattern among African economies, with resource-intensive nations experiencing slower growth and increasing fiscal distress relative to non-resource-intensive economies.Mining labour migration (large-scale and ASM) has incited demographic shifts that continue spurring economic diversification and new spatial and occupational patterns (Mususa, 2012).
Inflated expectations have been built around the developmental benefits of an industry whose future is uncertain.Policy mechanisms conceived during a boom have been implemented during a slump.This creates socio-political tension at local and national scales (Bebbington et al., 2018), reconfiguring state-business relations and the dimensions of "resource nationalism" (Andreasson, 2015;Childs, 2016).Perceptions of the limited developmental benefits deriving from extractive industries and the unsatisfactory fiscal contribution of multinational investors are driving new demands for the state to retake control of mineral resources.However, with declining resource rents, the balance of bargaining power between fiscally constrained African states and multinational firms is shifting.
The booming extractive sector spurred a concomitant research boom.Geographers have been quicker to examine the consequences of commodity "booms" than resource "busts" in emerging economies.In academic and policy circles, research has focused on capturing and using the rewards of a supercycle rather than on managing the industry's volatility and busts (exceptions noted below).In this paper we map the terrain of research on the minerals supercycle in Africa in Geography, identify emerging post-supercycle trends in miningsome of which have overtaken researchand establish key questions for geographers and policymakers.The paper combines an analysis of industry trends with critical reflections on literature on extractive industries by geographers from the last two decades, prioritising work by African scholars.Though not a comprehensive formal systematic review, this paper reflects on key trends and absences to identify future directions.We write at a level of generalisation that exceeds country-or mineral-specific research.Industry-or locationspecific dynamics may differ from the more general trends we posit and we examine metals mining, not oil and gas.
The broad trends and new geographies in commodities investment and production of the supercycle are reviewed in section 2, followed by three sections exploring emerging themes from the supercycle and existing research: new geographies of struggle (section 3), national minerals-based development (section 4), and labour and livelihoods (section 5).In each section we sequentially examine key trends during the supercycle, examine the emerging post-supercycle picture, and identify core areas for future research and policy development.

| KEY ECONOMIC TRENDS OF THE SUPERCYCLE AND BEYOND
From the early 2000s, rising emerging-market demand, particularly from China, drove rapid mineral commodity price increases of a kind not seen since the 1960s (Figure 1).Interrupted by the 2007/08 global financial crisis, the trend recommenced immediately thereafter.This reinforced dominant industry narratives that commodity prices were not in an ordinary price cycle, but a "supercycle" that would continue for decades (Humphreys, 2015).These narratives shaped investor, corporate, and government expectations.
Mining is defined by the imperative for continual spatial expansion to secure new reserves, but this took on a new character during the boom (Bridge, 2008).Inflated price expectations made projects previously deemed uneconomic or excessively risky newly attractive to multinationals, while concerns over scarcity compelled strategic investment from statebacked emerging market companies, particularly from China (Carmody, 2011).Industrial mining's geographic scope expanded rapidly, exemplified by increased multinational interest in high-risk post-conflict jurisdictions such as the DRC, Mozambique, and Sierra Leone, as well as large increases in foreign direct investment (FDI) to reach unexploited deposits in more established mining jurisdictions like Zambia, Ghana, and Guinea.There was also a spatial shift to the deep-ocean as a new space of mineral extraction, with diamonds and phosphate now mined from the seabed within the exclusive economic zones of Namibia and South Africa (Childs, 2020).Similarly, the search for new, "unconventional" rare earth mineral deposits focused on land (e.g., Mkango Resources in Malawi), and water, such as the tellurium-rich Tropic Seamount, 500 km off the coast of north-west Africa (Cornwall, 2019).
Neoliberal reforms of previous decades designed to improve investor shares of resource rents (Campbell, 2013) fuelled extractive expansion.Interpretations ranged from a new "scramble for Africa" (Carmody, 2011) entrenching Africa's neocolonial insertion in the global economy (Bush, 2010;Taylor, 2014), to optimistic reappraisals of the role of minerals in Africa's economic development as "Africa Rising" (African Morris et al., 2012;Union, 2009) (see section 4).Scholars and policymakers shared a common assumption of continued high prices and rapid expansion of mining.
Subsequent events confounded these assumptions.Triggered by slowing Chinese growth and excess capacity created by over-investment during the boom, mineral prices fell in near unison in the mid-2010s.As of 2018, most minerals remained well below 2007/08 peaks, a notable exception being battery metals like cobalt.
Many major mining companies were placed in financial difficulty, with excess capacity, over-leveraged balance sheets from increasingly financialised accumulation strategies, and high operating costs from prioritising volume over efficiency, and accessing lower-quality and harder to access ore bodies (Humphreys, 2019b).Market valuations for listed mining companies slumped as returns and investor optimism dwindled (Figure 2).
Post-2016 trends reflect these cost challenges.Revenues for PWC's largest listed mining companiesa global sample that now includes 19 emerging market-headquartered firms, 10 of them Chineseare just below their 2012 peak, but margins remain considerably lower (Figure 3).
Corporate strategy shifted from expansion to cost-cutting and repairing balance sheets.Capital expenditure for PWC's top-40 dropped almost 60% between 2012 and 2018, alongside near identical decreases in exploration expenditure and debt issuance globally (Table 1).In recent years alternative mine financing provided by royalty and streaming companies has become increasingly significant, though aggregate data are not available.
Efforts to improve productivity fuelled interest in automation and mechanisation (Durrant-Whyte et al., 2015), with implications for labour.Major mining companies already show signs of significant shifts in labour intensity (Humphreys, 2000-2018 (2000 = 100).
Beneath these general trends is considerable heterogeneity.Major mining firms in Africa have become more diverse due to emerging market investment, most significantly from Chinese companies, some of which are state-owned or benefit from  forms of state support.Investments often involve unconventional financing arrangements like resource-for-infrastructure deals, and are often motivated by strategic concerns about resource security rather than solely commercial imperatives (Power et al., 2012).Whilst Power et al. (2012, pp. 191-220) highlight differing approaches to environmental governance and labour relations, Lee (2018) suggests that some Chinese mining investors may be more long-term in orientation and thus less sensitive to market and political risks.Deeper discussion of trends relating specifically to emerging market investment in African minerals is beyond this paper's scope, but indicators suggest that the pace of Chinese expansion in African mining may have slowed.The annual value of Chinese investment transactions in sub-Saharan Africa's metals' sector has been falling over recent years, from US $4.4bn in 2014 to US$2.2bn in 2018, compared to a 2011 peak of US$5.6bn. 1 While overall Chinese FDI in Africa continued increasing, Chinese FDI stock in African mining remained stationary between 2016 and 2018. 2  Declining profitability and investor caution has clear implications for mineral-dependent African countries.For the 2013-2017 period, 90% were commodity dependent (commodity exports > 60% total merchandise export value) (Figure 4), with 18 classified as mineral-commodity dependent (UNCTAD, 2019).Altered conditions in mining have had significant macro-economic consequences for many countries' growth prospects.The IMF has observed a dualistic pattern, with resource-intensive countries averaging 2%-3% annual GDP growth since 2013meaning relatively stagnant per capita growthcompared to above 5% in non-resource-intensive countries (IMF, 2019).
This has fiscal consequences.Since the 1980s, African countries have had significantly larger resource rents (oil, gas, mineral, and forest rents) relative to GDP than low and middle income country averages, rents that are important to government revenues given thin tax bases.Resource rents grew spectacularly during the 2000s, averaging just under 15% of GDP compared with 8% during the 1990s, but have now fallen back to 1990s levels (Figure 5).This has created severe fiscal strains as many governments increased public expenditure and borrowing in anticipation of a sustained supercycle.F I G U R E 4 Commodity-dependent countries by region, 2013-2017.

Source: UNCTAD
As the following sections show, these changing economic dynamics prompt a reframing of recent scholarship on mining and development and raise new questions while giving fresh salience to earlier work on the social, political, and economic consequences of boom-bust cycles (e.g., Ferguson, 1999;Karl, 1997).The following sections interrogate how the changing economic context altered the trajectory of mining across different countries and commodity groups.Leaving aside examples such as battery metals, the framing assumption of a "scramble"inexorable rapid expansion of frontiers to secure scarce resourcesno longer appears certain.Instead, a more complex picture is emerging.Major mining companies have generally adopted more conservative approaches, favouring lower-cost, higher-grade deposits in less "risky" jurisdictions.China's growth is slowing and expected to shift to a less mineral-intensive economic model, slowing demand growth, though not necessarily causing absolute declines.These dynamics highlight the need for research that examines what heightened corporate investment caution and withdrawal might mean for (1) policy in commodity-dependent African states such as Zambia, the DRC, and Guinea, (2) relations and bargaining between mining, states, and communities, and (3) the role of artisanal and small-scale mining in exploiting high-risk deposits (Verbrugge & Geenen, 2018).

| NEW GEOGRAPHIES OF STRUGGLE
Recent decades have seen a rise in mining-related conflictsbetween 2002 and 2013 the ICMM documented an eight-fold rise in conflict (ICMM, 2015).This trend has been extensively analysed by geographers, who have documented a rise in levels of struggle and violence exacerbated by the industry entering "new" areas and implementing new forms in previously mined areas across Africa (e.g., from shaft to open pit mining).These struggles were "geographic" because they operated unevenly across both space and time.They occurred locally and were often connected with global activist communities. 3 These struggles were also examples of the mining industry's "politics of time," where the speed environmental impacts were made knowable shapes the ensuing kinds of struggle (Kirsch, 2014).
As the supercycle ended, dynamics changed.Some conflicts associated with expansion lessened, simply because mining companies were less active and not entering new areas.As revenues declined, communities' bargaining power over mining Source: World Bank companies decreased and their "dependence" was exposed (Verbrugge & Geenen, 2018).Other forms of struggle increased, sometimes catalysing in protests against layoffs (see section 5).Where operations continued, expectations of company behaviour sometimes rose, particularly in areas with a history of mining conflicts (e.g., Ghana).Declining mining revenues generated pressure to focus community investment budgets on conflict amelioration.For some industry observers, corporations took an instrumental approach to spending and CSR, further reducing corporate/community relations (Thomson, pers.comm.).Community development became secondary to "shared value," risk management, "social performance," and "measurable results," while "social license" remained, despite its faults (Frederiksen, 2018).
These trends raise important questions for researchers about the new geographies of struggle.How do conflicts over extractive industry change as mining shrinks or retreats?Are new emerging market actors, often with less experience of managing impacts, running "poorer performing" mines?How can "enclave" mining generate benefits (and therefore legitimacy) for local communities?How can changing discourses and pressures on CSR shape development outcomes?How does resurgent CSR instrumentality affect patronage and clientelism in areas of operation?And finally what are the longterm consequences of automation and retrenchment on political volatility?

| NATIONAL MINERALS-BASED DEVELOPMENT
The "supercycle" induced significant shifts in thinking about mining's role in economic development in many Africa countries, and the state's role in governing mineral extraction.The rise of "resource nationalism" broke a period of consensus around neoliberal models of mineral governance that emphasised favourable conditions for FDI, minimising risks and increasing investors' share of resource rents (Campbell, 2013).Simultaneously, dominant thinking in economics around the "resource curse" framed mineral rents as economically and politically distorting (Collier et al., 2011;Siakwah, 2017), a problem to be solved by "good-governance" institutions for mineral extraction and marketing that minimised political discretion and state intervention.Dissatisfaction over the benefits of these policies, despite the supercycle, led many resourcerich governments to consider alternatives (Besada, 2016).Reviews of mining legislation and contracts with a view to increasing state and/or community shares in mineral rents took place in multiple African countries (Tanzania, Ghana, Sierra Leone, South Africa, Liberia, the DRC, Zambia, and Zimbabwe), aimed at higher taxes or royalties, indigenous ownership requirements, and domestic value addition or local content requirements (Humphreys, 2019a).Pejoratively termed "resource nationalism" by critics, this represented a reframing of mining's role in national development (Andreasson, 2015;Childs, 2016).New policies sought to use state intervention to increase domestic economic linkages with large-scale mining, boost domestic industrial capacity with requirements for local content upstream in the value chain, and encouraged mineral beneficiation downstream (Morris et al., 2012).Such ideas were influential in African policy circles, as reflected in the 2009 Africa Mining Vision (African Union, 2009;Busia & Akong, 2017), and saw new legislation, local content policies, and mining licence renegotiations across the continent (Ambe-Uva, 2017; Ayisi, 2015).However, much research into commodity-based development seemed to take for granted that the supercycle, and the increased relative bargaining power of mineral-rich states accompanying it, would continue for decades.Moreover, research focused on material linkages to mining, and paid too little attention either to the relative power of multinational corporations or the political economy that shapes the design and delivery of industrial policy on the ground (Fessehaie & Rustomjee, 2018).Outcomes of local content and linkage development initiatives to date have frequently been poor (Hilson & Ovadia, 2020).
Policies conceived and implemented during the latter stages of a boom have continued into the slump.Many African governments are pursuing a harder line against multinational mining companies in worsening economic conditions.In recent years this has notably included protracted disputes between governments and multinational investors over efforts to increase rates of taxation on mining in Tanzania, Zambia, and the DRC, even as mineral prices and investment levels have remained low.This suggests the durability of "resource nationalist" policy agendas given competition for resources from new emerging market mining investors, increasingly strained fiscal circumstances, and populist political pressures (e.g., see Jacob & Perseson, 2018).The situation emphasises how changing market dynamics intersect with the political economy of minerals-based industrial and fiscal policies, which can create new interest groups, patronage networks, and political claims on the mining industry (Hansen et al., 2016;Jacob & Perseson, 2018).The shift in economic policy towards mining has also been accompanied, in many instances, by changing political rhetoric about natural resources' contribution to development, and rising popular expectations related to public expenditure, employment, and opportunities for domestic business elites (see Barlow, 2020;Frynas et al., 2017).This combination of factors has contributed to an increasingly fraught politics of mining and more adversarial state-business relations in many contexts.
The evolution of national minerals-based development policies in a commodity slump compounded by COVID-19 depends on the balance of forces within society, how new and existing interest groups develop and relate to one another, and what this means for the distribution of rents from mining and commitment to broad-based economic development -including the development of industrial activities.Given the current context, will changing economic conditions continue to entrench "resource nationalism" or spur a return to more "investor-friendly" governance modalities?Will mining companies previously willing to engage with voluntary local content and beneficiation initiatives continue to do so, or will cost-cutting and productivity imperatives spur capital-intensive, "enclave" operating models?Can nascent domestic industries around mining survive continued instability and uncertainty?What forms of counter-cyclical industrial policy support might help protect the limited progress made to date?

| LABOUR AND LIVELIHOODS
The supercycle led to the expansion of formal and informal mining in many African countries with important consequences for labour and livelihoods.Struggles between mining companies and labour expanded with industry growth and increased profits, with companies organising their corporate and operational structure to minimise (organised) labour's power, most notably through the use of subcontracting (Fraser & Lungu, 2007;Negi, 2011;Verbrugge & Geenen, 2018).In areas with accessible deposits, rising prices also made ASM a lucrative livelihood option, leading to booms in areas previously overlooked (e.g., Tanzania, eastern DRC; see Jonsson & Bryceson, 2017;Verbrugge & Geenen, 2018).While augmented local and national state resources enabled expanded welfare programmes and state employment (Bebbington et al., 2018), the impact of mining booms on livelihoods remained complex.For example, in rural areas, increased competition for land and water, negative environmental externalities, and changing rural labour force composition all impacted existing agrarian livelihoods (see section 3), while the corporate priority of securing reserves for future growth generated increased contestation over resource access between the formal and informal sectors (e.g., DRC, Tanzania, Ghana; see Hilson & Potter, 2005;Verbrugge & Geenen, 2018).Areas witnessing both artisanal and formal booms saw influxes of migrant labour, for both employment in mining and the attendant service-sector around operations (Akiwumi, 2011;Bryceson & Geenen, 2016;Nyame et al., 2009).This strained existing natural resource-based livelihoods, infrastructure/social services, and created social tension as communities linked migration to crime and prostitution (Akiwumi, 2011).More widely, increased competition for qualified mining talent drove wage inflation in skilled job categories, exacerbating already stark expat/local wage disparities.
Post-supercycle, this picture changed.As focus shifted to cost control over expansion, emphasis on technological innovation and digital technologies expanded to enable automation and mechanisation and to more efficiently manage resources and reduce costs (Durrant-Whyte et al., 2015).With labour a key cost, labour-intensive extraction techniques were under pressure, creating labour struggles.Redundancies and layoffs (or prospects thereof) from the formal mining sector have political and economic ripple effects.For example, unemployment and constraints on government spending increases pressure on mines as a source of livelihoods and generator of employment (Hilson & Potter, 2005).However, as the formal sector stalls on deposits that are no longer economic, more artisanal producers can occupy space left by the formal sector, absorbing its labour (Verbrugge & Geenen, 2018).The outlook for communities that grew up around now-closed mining operations can be bleak.Historically, many communities have taken decades to recover.For example, South African exgold and coal mining towns have suffered widespread unemployment and deprivation (Binns & Nel, 2009, Siyongwana & Shabalala, 2018).By contrast, company layoffs can spur surges in ASM as unemployed miners resort to artisanal mining at familiar sites (Yankson & Gough, 2019).Particularly gold diggers have proliferated in mineral-rich mining sites, acting inadvertently as prospectors for large-scale mining interests.Artisanal miners, who benefited from the gold price rise of the supercycle, were most likely to move to nearby towns, where they could invest in diversified businesses and improved housing, thereby contributing to urbanisation, while being readily at hand should prospects of mineral investment return (Bryceson & MacKinnon, 2012;Jonsson & Bryceson, 2017).
All this raises questions for future research (Table 3).Automation and mechanisation leads to a smaller labour force, with higher skill requirements.Technology has often been used as a means to control labour, what now?How will these trends interact with expectations for employment creation?If the retreat of the formal sector sees a concomitant growth in the artisanal sector, how best to manage the social and environmental impacts of ASM? How do mining and other livelihoods (particularly, agrarian) interact in changing rural spaces?What do these changing livelihood opportunities mean for patterns of migration in extractive zones?How can mine closure impacts on local communities be best managed?Can deindustrialisation lead to economic innovation and alternative paths to development (Binns & Nel, 2009)?How are former mine sites remembered and what material and emotional geographies do they leave behind with what consequences for justice (Pini et al., 2010)?
In this paper we have explored recent research on the mining sector in Africa and reflected on this considering the supercycle's end.We have explored the changing geographies of investment and sectoral shifts in mining before examining the changes brought by the supercycle, its end, and what questions emerge for future research in three key areas: struggle, national development strategies, and labour and livelihoods.The last half decade has seen a fundamental shift in the possibilities of extraction-led development and the reconfiguring of governance of the extractive sector.New trends have emerged around corporate strategy and actors, including the increasing presence of emerging market-based companies with implications for state-business relations and models for mineral-based development strategies.Equally, the end of the supercycle has perpetuated or intensified pre-existing trends towards mechanisation, automation, and enclavity, while distributive pressures on companies by local communities and host nations increase.While, at the global level at least, it • Will changing economic conditions entrench "resource nationalism" or spur a return to more "investor-friendly" governance modalities?
• Will cost-cutting and productivity imperatives spur capital-intensive, "enclave" operating models?
• What forms of counter-cyclical fiscal policy and industrial policy support might help mitigate the effects of slumps and protect nascent domestic industries around mining?

Labour and livelihoods
• Growth of employmentboth formal and ASM • Increased labour struggles and use of subcontracting • Migration and service sector growth in areas of mining growth • Skilled wage inflation • Focus on cost control, accelerated automation, and mechanisation leading to retrenchment • ASM remains largeformal sector withdrawal transitions some deposits to ASM • Challenges for former mining towns and areas • How will retrenchment, automation, and mechanisation interact with expectations for employment, and other rural livelihoods?
• How to manage the social and environmental impacts of a larger ASM sector?
• What do these changing livelihood opportunities mean for migration in extractive zones?
• How can mine closure impacts on local communities be best managed, both culturally and economically?remains that Africa is not that important to mining, mining is important to Africa.As scholars we need to grapple with the changes afoot in Africa since the end of the supercycle, not least those spurred by the COVID-19 pandemic.The research directions suggested here offer initial ways to chart how mining can offer more constructive pathways towards development in both boom and bust.

F
I G U R E 2 FTSE 350 mining index 2000-2018.Source: Bloomberg Data F I G U R E 3 Revenue and profitability of PWC top-40 largest mining companies, 2006-2018.Source: PWC T A B L E 1 Global mining industry investment, 2012-2018 (US$bn, 2018 prices).

F
I G U R E 5 Resource rents as a percentage ofGDP, 1970GDP,  -2017.   .
Employment and capital intensity of production for Rio Tinto, AngloAmerican, and BHP, 2008-2018.
T A B L E 2Source: company reports.
T A B L E 3 Summary of key arguments.Revised minerals legislation to secure greater share of rents for national development and counter enclave tendencies, e.g., local content • Increased bargaining power of states over international mining companies • Rise of "resource nationalism" •