TY - JOUR
AU1 - Bowman,, Andrew
AB - ABSTRACT This article analyses the causes, outcomes, and political significance of the inter-connected operational, financial, and governance crises afflicting Eskom, South Africa’s electricity parastatal. These crises emerged in the context of African National Congress initiatives to turn Eskom and other key parastatals into instruments of an envisaged South African developmental state, through increased investment and strategic procurement to support economic transformation goals. Instead, Eskom’s spiralling costs, procurement irregularities and inability to translate increased investment into functional new infrastructure meant it impeded these goals. Its indebtedness became a severe macro-economic risk, making Eskom a precarious nexus for the circulation of public funds, while the cost and unreliability of electricity has undermined South Africa’s energy-intensive industrial core. Intertwined with this were multiple high-profile corruption scandals associated with the ‘state-capture’ controversies of the latter stages of Jacob Zuma’s presidency. The article argues that Eskom’s extreme dysfunctionality results from long-running, and as yet unresolved, contestation of the parastatal and electricity policy more broadly by various interest groups, in a context of an increasingly fragmented political and business elite. This created a range of incoherent distributional pressures and institutional constraints. Rather than a straightforward outcome of corruption and ‘state capture’, this reflects deeper tensions in the post-apartheid political economy. Over the past decade, South Africa has been afflicted by a series of severe crises at the state electricity provider, Eskom. The vertically-integrated generation, transmission, and distribution company is South Africa’s largest parastatal, producing nine-tenths of its electricity and selling to seven neighbouring countries. Its crises have been simultaneously operational, financial, and political. Most prominently, they have included widespread power-outages—‘load-shedding’—in 2008, 2014–2015, and 2018–2020, which have brought far-reaching disruption to economic activity. Eskom’s worsening financial difficulties have also had far-reaching consequences. With its costs soaring, a decade of above-inflation tariff hikes raised Eskom’s nominal average electricity prices by over 300 percent in the decade to 2017/2018.1 Multiple state bailouts have nonetheless been required to prevent bankruptcy, with the company’s 440bn South African Rand (ZAR) debt now a leading source of macroeconomic risk.2 Accompanying these operational and financial crises have been intense political conflicts over procurement spending, control of Eskom’s board, and the broader direction of energy policy, linked to South Africa’s ‘state capture’ controversies.3 This destabilized Eskom’s corporate governance—there have been 12 chief executives since 20074—and made it a key battleground in the ruling Tripartite Alliance’s5 escalating factional struggles. This article contributes to analyses of the causes of Eskom’s crises and their implications for understanding the political economy of South Africa, in particular regarding the role of parastatals.6 While situating events in their deeper historical context, it focuses on the period after 2004 in which, having abandoned privatization, the African National Congress (ANC) sought to turn South Africa’s key parastatals from commercialized public utilities into instruments of broader economic transformation policy.7 Eskom was foundational to an envisaged South African ‘developmental state’ that became a central imaginary of ANC economic policy discourse in this period, with increased parastatal investment and strategic procurement expected to revive growth, support re-industrialization, and accelerate black economic empowerment (BEE).8 Outcomes have been far removed from this imaginary. Transformation and industrial policy initiatives in Eskom’s procurement achieved limited success, but inefficient investment, cost increases, and widespread contracting irregularities created huge financial losses that have been socialized9 with significant political and economic consequences. Eskom’s mounting debt burden has constrained fiscal capacity in a wider context of prolonged economic stagnation, while the cost and unreliability of electricity has damaged the competitiveness of South Africa’s pivotal energy-intensive manufacturing and mining industries. Historically, Eskom’s provision of cheap electricity underpinned South Africa’s Minerals-Energy Complex (MEC), the system of accumulation defined by close, synergistic relations between mining, heavy industry, and finance.10 Eskom increasingly came to undermine this system. With consumption from major industrial and mining consumers declining, the company faces a death spiral remediable only through major restructuring or further socialization of losses. The political spill-overs have been as significant. Factional infighting over Eskom contributed significantly to the end of the Jacob Zuma presidency, and continue to blight Cyril Ramaphosa’s administration, with Eskom’s financial and operational problems far from resolved.11 The article draws on the political settlements literature on industrial policy to interpret these processes.12 This literature understands the performance of state industrial policy agencies—which Eskom and other parastatals were envisaged as—as dependent on the configuration of organizational power, institutions, and rents. Successful industrial policy requires the emergence of ‘pockets of effectiveness’ within the state bureaucracy with high levels of capability and a degree of autonomy, but also for powerful groups beyond the state, in political leadership, business elites, and popular-political formations, to cohere around mutual interests in outcomes.13 Tracing specific instances of contestation over public resources and industrial policy rents can therefore also generate broader political insights.14 The article argues that Eskom’s extreme dysfunctionality derives from long-running contestation of the organization and the rents it controls through its monopoly position in the electricity system. This created critical delays to key policymaking processes, and a range of conflicting distributional pressures and incoherent institutional constraints. These included balancing mounting commercial pressures from reliance on capital markets to fund a huge and hurried investment programme, a utility role in a context of high energy-poverty and an energy-intensive industrial core, and a radical re-orientation of procurement towards political goals in a context of disjointed governance and accountability mechanisms. These circumstances created conditions of organizational chaos in which predatory accumulation associated with ‘state capture’ could flourish and further exacerbate Eskom’s dysfunctionality. The article further argues that explanations of Eskom’s crises as straightforward outcomes of corruption by a small elite network, or a generic failure of state-ownership and interventionist industrial policy, risk being reductionist. Instead, Eskom’s situation reflects deeper tensions in the post-apartheid state. These have arisen from historical processes producing entrenched racialized inequalities in economic ownership, factional fragmentation in politics and misalignment within and between bureaucratic, political, and business elites.15 Related to this are strong external pressures on state organizations as means to enable accumulation, and sharply differing ideological positions on the state’s economic role.16 These political conditions proved inimical to the realization of the envisaged ‘developmental state’, of which parastatals were an intended cornerstone, with associated industrial policy initiatives subverted or undermined. The article’s empirical material is presented through a qualitative case study method, which, following historical institutionalist approaches, analyses large firms as socially embedded with strategy, structure, and performance shaped by dynamic interaction with the external environment.17 Rather than a nexus of contracts or a production function, large firms are understood as ‘contested organizational devices for the accumulation, mobilization, and deployment of resources’, and accordingly as sites of multiple power struggles.18 Tracing complex processes of institutional and inter-organizational change, and their manifestations in a specific organization, requires careful attention to contextual specificity and a wide range of data sources. The article uses grey literature from Eskom, relevant government ministries, statutory bodies, private companies and trade associations, triangulated with 11 in-depth key-informant interviews with former Eskom executives and employees, government officials, and private sector organizations, as well as observation at industry conferences. This was conducted between 2014 and 2018 while undertaking research in South Africa. Several interviews have been anonymized due to the sensitivity of the material. Further evidence was gathered from an analysis of testimony to parliamentary enquiries and official investigations into Eskom. Quantitative time-series data were compiled from Eskom financial statements and analysed with descriptive statistics and accounting ratios to examine changing financial and operational trends. The article proceeds in the next section to elaborate the theoretical framework, critiquing developmental state theory via the political settlements literature. The subsequent two sections situate the Eskom crisis in its historical and institutional context, from Eskom’s role in the apartheid MEC, through the aborted privatization initiatives of the Mbeki period, to more recent ANC ‘developmental state’ initiatives. After that, the article analyses Eskom’s evolving crises since 2008 and their causes and consequences, while the final section discusses their wider implications for the political economy of South Africa. State-led industrialization, parastatals, and political settlements This section situates the article in relevant literature and elaborates the theoretical framework for explaining outcomes in state-led economic development initiatives. Across much recent political economy scholarship on this subject in an African context, which has proliferated alongside revived interest in state-led development models, developmental state theory has provided this framework.19 Developmental state theory has also manifested in policy discourse in several contexts, notably South Africa (ANC developmental state discourse is an object of analysis in the next section). A middle-range theory of state-led industrialization, the ‘developmental state’ derives from studies of East Asian newly-industrialized countries, but is increasingly employed as a framework for understanding the varying outcomes of industrial policy efforts in contemporary Africa. Unlike ideal-type ‘market rational’ states, archetypal developmental states intervened extensively in markets and strategically allocated rents to coordinate industrialization.20 Explanations in classic literature for their success in doing so while avoiding unproductive rent-seeking point to autonomous, capable, and highly-centralized state bureaucracies disciplining firms—rent allocations being accompanied by ‘reciprocal control mechanisms’—combined with cooperative state-business relations founded on ‘institutionalized channels for the continual negotiation and renegotiation of goals and policies’.21 Parastatals typically played important roles in these archetypal ‘developmental states’. As hybrid public-commercial entities they are unlike other elements of the state bureaucracy, both acting on and participating in the sphere of private enterprise.22 Their role commonly involved, firstly, lowering strategic businesses’ costs by providing cheap access to network infrastructures or intermediate goods like steel, predicated on parastatals’ ability to endure low returns or absorb losses. Secondly, it involved initiating high-risk knowledge or capital-intensive activities, like electronics and petrochemicals, and subsequently crowding-in private investment.23 Finance was also often state-owned as a means of directing credit to priority industries.24 Parastatals were particularly prevalent where governments doubted the loyalties, capabilities, or developmental orientation of domestic business elites, or where foreign or ethnic minority-owned firms predominated.25 Such concerns contributed to the prevalence of parastatals across post-colonial Africa. Appraisals of parastatals’ developmental role in Africa, however, have tended to be more negative, generally presenting them as inefficient patronage vehicles.26 Alongside orthodox economic critiques of state-ownership, this bolstered sweeping structural adjustment privatization programmes, and contributed to arguments that African developmental states were prevented by intrinsically neo-patrimonial politics and capture-prone bureaucracies.27 Recently, these arguments have been challenged by scholarship on state-led development in fast-growing African economies, with Ethiopia and Rwanda in particular drawing tentative comparisons with archetypal developmental states.28 Frequently highlighted in this literature is the use of state or party-owned enterprises to coordinate or intervene in strategic sectors.29 This is in a wider context of renewed enthusiasm for parastatals in emerging economies given the perceived successes of Chinese state-capitalism.30 However, as political settlements scholars argue, developmental state theory has limitations as a means to analyse the revival of state-led development initiatives. Importantly, bureaucracies in archetypal developmental states were not as autonomous, coherent, and free of rent-seeking and clientelism as commonly suggested.31 Using the developmental state as a means to analyse industrial policy in Africa is therefore to risk searching ‘for an ideal type of governance that never existed in the first place’.32 Evidence from other successful late-industrializers also shows instances of states exhibiting neo-patrimonial tendencies while also implementing successful industrial policy.33 As Hazel Gray puts it, that ‘[t]he fact that rents are used both to achieve political and economic goals does not…render them ineffective’ in terms of advancing industrialization.34 Developmental state theory therefore has key limitations in explaining the variety of industrial policy outcomes across different contexts. Alongside the questionable empirical foundations, political settlements scholars view this limitation as stemming from an overly narrow focus on the formal characteristics of state institutions, such as levels of autonomy or centralization.35 They instead argue that understanding why widely adopted industrial policies succeeded in some instances but not others requires extending the analytical frame to encompass the wider distribution of, and struggle for, organizational power and rents among social groups within and beyond the state.36 This configuration of organizational power, rents, and institutions that constitutes the political settlement evolves with social struggle, and different configurations may be more or less amenable to the successful implementation of industrial policies.37 Industrial policy strategies that succeed in some contexts, as with the case of the archetypal developmental states, may fail in others not because of inherent flaws in the policies, or a lack of autonomy or capability among implementation agencies, but due to incompatibility with the political settlement.38 Where policies and formal institutions governing rent distribution misalign with the distribution of power, they may be obstructed, subverted, or create forms of contestation that limit effectiveness.39 To create necessary pockets of effectiveness among state industrial policy implementation agencies, bureaucrats must have a degree of autonomy to ‘fend off distributional demands or conflicting interests’ from powerful actors that might subvert them.40 This, however, is a necessary but insufficient condition, with successful industrial policy, and the emergence of such agencies, contingent on political elites, business elites, relevant aspects of the state bureaucracy, and popular-political demands cohering around mutual interests in its outcomes.41 The emergence of pockets of effectiveness is therefore a reflection of the political context, rather than an ‘autonomous’ extraction from it.42 States otherwise defined by widespread corruption are capable of producing such agencies where powerful interests align around their functioning as such. Oil parastatals provide interesting examples, notably Angola’s Sonangol, a relatively stable, protected technocratic pocket of effectiveness interfacing with international oil capital in a context of pervasive bureaucratic dysfunction.43 Dispersal of power and competition over state agencies, in contrast, is inimical to the emergence of pockets of effectiveness.44 A political settlements approach enables a ‘more nuanced understanding of the dynamics of industrial policy and outcomes for economic transformation’, but relies on a more careful consideration of historical specificity rather than the transfer of a general model for industrial policy success as in the case of developmental state theory.45 This requires analyzing evolving relationships and processes of contestation within and between the political groupings, the bureaucracy and businesses, relationships which are shaped not simply by material interests and struggles over power and resources, but also by ideology.46 In the following sections, the paper traces these evolving relationships and processes of contestation concerning Eskom. Eskom and the changing political economy of parastatals in South Africa This section situates Eskom’s recent crises in their historical context. The account is inevitably abbreviated,47 but illustrates path dependencies, critical junctures and forms of institutional and inter-organizational embeddedness that have shaped Eskom’s present situation. Parastatals were pivotally important to South Africa’s developmental trajectory during colonialism and apartheid. Eskom’s48 formation in 1923 enabled provision of abundant cheap energy to spur inter-war industrialization and eventual consolidation of the electricity supply industry.49 It became a key component of the minerals–energy complex (MEC), the nexus of finance, mining, and heavy-industry constituting the apartheid economy’s ‘core site of accumulation’.50 Eskom, alongside parastatals in steel, petrochemicals, and logistics, underpinned the MEC, creating a capital-intensive, highly concentrated, energy-intensive industrial structure centred around mineral extraction and beneficiation.51 Parastatals were also crucial to the formation of an Afrikaner middle class through employment, and an Afrikaner capitalist class through preferential procurement.52 Eskom generated among the world’s cheapest electricity—half the Western economies’ 1970s unit-price—to indirectly subsidize heavy industry and deep-level mining.53 A symbiotic relationship with coal mining enabled this. Generation plants were built adjacent to ‘tied’ co-financed collieries, tailored to low-grade coal—while mining companies exported high-grade coal—fed by conveyors, and providing long-term ‘cost-plus’ contracts that provided a steady income stream.54 South Africa thereby developed as a peculiarly energy-intensive economy (Figure 1). As explored in the fourth section, tightly-coupled, mutually-dependent relationships between Eskom and large-scale mining and industrial suppliers and customers proved enduring in subsequent decades, despite far-reaching institutional changes. Energy-intensive industry has diminished proportional to economic activity, with mining declining from 14 to 8 percent of GDP from 1994 to 2017. But with almost 60 percent of goods export value deriving from minerals, metals, and chemicals, these industries, and Eskom’s ability to supply them with cheap and reliable electricity, remain centrally important to the South African economy.55 Eskom, correspondingly, remains dependent on a small number of major industrial consumers—now organized through the 30-member Energy Intensive Users Group (EIUG)—that account for around 40 percent of electricity sales. With coal-fired plants constituting over 80 percent of Eskom’s nominal generation capacity, it also remains entwined with powerful coal industry interest groups.56 Figure 1 Open in new tabDownload slide Energy production per-capita (kWh), 1971–2011. Source: World Bank Word Development Indicators. Figure 1 Open in new tabDownload slide Energy production per-capita (kWh), 1971–2011. Source: World Bank Word Development Indicators. During much of the apartheid period, Eskom had high levels of independence from regulatory and market disciplines, and almost exclusively served white residential and industrial consumers to the exclusion of the majority black population.57 However, a 1980s financial crisis arising from over-investment in generation capacity compelled reforms, whereby Eskom became quasi-commercialized and more closely regulated.58 The democratic transition broadened the consumer base, and embedded Eskom in popular politics. Mass-electrification began in earnest in the late-1980s, but by 1994 fewer than half of households had grid connections.59 Rapidly expanding access was prioritized in the ANC’s Reconstruction and Development Programme, envisaging a social welfare role for parastatals in basic infrastructure provision.60 Eskom aligned, widening its remit and committing to double household connectivity by 2000. Serendipitously, previous over-investment meant it had excess capacity and, temporarily freed from generation capital expenditure, it could co-finance rapid electrification, endure sub-inflation tariff increases, and implement Special Pricing Agreements for major industrial customers.61 This both cosseted energy-intensive industry, and shaped enduring expectations around low electricity costs. Along with other basic services occupying a grey area between commodity and socio-economic right, a particular moral economy of electricity affordability and access remain important to the ANC’s electoral support, embedding Eskom in popular politics. While post-apartheid advances in electricity access have featured prominently in election campaigning—with around 85 percent of households now connected—continued shortcomings are reflected in the high-usage rates of wood and paraffin fuels, electricity’s prominence in many so-called ‘service delivery protests’, and high levels of illegal connection and non-payment.62 The adoption of privatization policies in the late-1990s was a critical juncture for Eskom. Their adoption reflected the ANC’s ideological repositioning on economic policy, the opportunity to advance black capitalist class formation through BEE investors’ purchase of discounted parastatal shares, and energy sector experts’ calls to follow international reform trends.63 Following a 1998 White Paper plan to unbundle Eskom and facilitate entry of private generation competitors, generation investment was suspended, against the advice of Eskom management.64 Eskom was corporatized in 2001, and began pursuing commercial returns through more assertive cost-recovery to incentivize would-be investors.65 However, with ideological schisms in the Tripartite Alliance intensifying, opposition from trade unions, civil society mobilizations, and resistance from Eskom management, privatization was abandoned in 2004.66 By the time generation investment recommenced after 2005, capacity was becoming severely constrained, necessitating haste and—because low tariffs had prevented building cash reserves—large-scale debt issuance. The implications of this are examined in the fourth section. ‘Levers that the government should pull’: reconceptualizing parastatals and economic transformation The factionalized contestation of privatization policies foreshadowed conflicts to come over the role of parastatals, and indeed the wider role of the state in economic development. Broader shifts in ANC economic thinking were reflected in the adoption of ‘developmental state’ discourse in the late-Mbeki period. Amid mounting anger over the disappointing outcomes of market-orientated policies, leftist and nationalist factions propelled Zuma to ANC president at the 2007 National Conference, which resolved to build a ‘democratic developmental state…which leads and guides [the] economy’.67 ANC translation of developmental state discourse into policy practice has been partial and inconsistent,68 but, crucially, has been marked by an expanded role for parastatals that became a defining feature of the Zuma presidency.69 The reasons for this relate to deepening tensions in the evolving post-apartheid political settlement during this period. ANC hegemony has rested on broad coalitions within and between the party and other elements of the Tripartite Alliance, and other social groups, notably the powerful labour movement, the working class electoral-base, and an emerging black capitalist class.70 Maintaining these coalitions in a context of persistent racialized inequalities, consolidated white control of the private sector and corresponding misalignment between economic and political power, necessitates elision of redistribution, deracialization and nation-building imperatives with other dimensions of economic policy.71 For the electoral-base, this entailed extension of welfare provision and social grants, improved basic infrastructure, and public sector employment, all in turn contingent on sustained growth. For the emerging black capitalist class, deracialization of private sector ownership was institutionalized under Mbeki as BEE policy. This mandated black equity-ownership targets alongside affirmative-action in procurement and senior employment for larger companies seeking state tenders or operating in licenced sectors, the aim being to create a ‘patriotic bourgeoisie’ aligning big-business with government’s developmental agenda.72 This was enacted within the orthodoxly liberal macro-economic policy programme and market-enhancing institutional framework derived from post-1994 bargains with big business.73 This framework produced moderate growth to the end of the commodities supercycle in 2011, but unemployment remained high, racialized inequalities wide, manufacturing in decline, and investment levels low relative to middle-income peers. Subsequently, growth has been effectively stagnant, less than 1 percent on a per-capita basis between 2011 and 2017.74 BEE accumulation also lagged behind expectations, with Treasury-commissioned estimates suggesting direct BEE-ownership of just 9 percent of the Johannesburg Stock Exchange’s market capitalization in 2016, against government’s 25-percent aspiration.75 Consolidated ownership and high entry-barriers across major industries have impeded independent black-owned firms, with BEE ownership proceeding primarily through leveraged purchases of minority shareholdings in established enterprises by politically-connected holding companies.76 This created a small and somewhat precarious group of super-wealthy BEE-entrepreneurs, but not the envisaged patriotic bourgeoisie.77 Reflecting these trends, the Zuma presidency saw the ascendency of leftist and nationalist political factions advocating greater exercise of state power to accelerate re-industrialization and deracialization. This was notably the case, as explained below, in public procurement. This so-called ‘developmental state agenda’ was later accompanied by a more far-reaching questioning of the liberal post-apartheid institutional order that became encapsulated in the discourse of ‘radical economic transformation’.78 In contrast to the centralized, coordinated operations of the archetypal developmental state, the South African state has, however, been defined by a fragmented approach to economic strategy.79 Deep ANC divisions over economic policy are reflected in state institutions due to the politicization of senior appointments, commonly referred to as cadre deployment. This creates a complex form of ‘party-state’ arrangement within key parts of the bureaucracy.80 Some agencies, notably the Department of Trade and Industry (DTI), Economic Development Department (EDD), and the Department of Public Enterprises (DPE) adopted more interventionist industrial policy and transformation measures. However, the Treasury, the superordinate economic policy agency, upheld a more orthodox approach and the maintenance of market-rational institutions.81 This was defended by liberal factions in the ANC, and incumbent big business, which has demonstrated remarkable holding power. Given the accompanying deep misalignments among state economic policy agencies, and between political leadership and the highly organized, rapidly internationalizing incumbent big-businesses,82 parastatals represented an obvious remaining means to mobilize and allocate significant resources in pursuit of more radical transformation policies. State-controlled through the DPE but quasi-independent operationally and financially, parastatals are not subject to the same oversight and constraints as other parts of the bureaucracy.83 Privatization programmes left negligible state-ownership in industry, and South Africa’s development banks are small compared to private banking. Infrastructure parastatals, though, remained large and economically influential, particularly Eskom and Transnet, the rail-freight and ports company, with a combined asset value equivalent of over 20 percent of GDP.84 A variety of initiatives proposed transforming these parastatals into ‘instruments of the developmental state’.85 As a 2012 Presidential Review Committee articulated, this involved reconceptualizing parastatals as ‘levers that the government should pull’, shifting beyond primary service-delivery roles to align with wider transformation policies.86 Reflecting incoherent parastatal governance arrangements, these were not encapsulated in over-arching legislation, but manifested in overlapping policies and proposals from a range of agencies. Synthesizing from official documents, two key roles were envisaged. Firstly, increased parastatal investment could stimulate growth, compensating for stalling private investment, and overcoming large infrastructure investment backlogs.87 Parastatals were, as one DPE official put it, ‘instructed to establish aggressive investment programs … to support the needs of a growing economy, rather than what their balance-sheets could comfortably accommodate’.88 Given parastatals’ independent capital raising capacities through the bond markets, this would function as de-facto off-balance sheet state development financing.89 Secondly, parastatal procurement, properly reformed, would provide a mechanism to support both domestic manufacturing through local content requirements and black capitalist class formation through raised BEE requirements.90 Added to this, parastatal job-creation would provide a means of boosting employment and skill development. Table 1 Eskom procurement, 1998–2017/2018 . Estimated Eskom procurement1 . Procurement as a proportion of GDP . Procurement from B-BBEE compliant suppliers2 . Procurement from black-owned (BO) suppliers2 . . Rbn, real . Percent . Percent . Percent . 1998–2007/2008 (average) 46 0.8 NA NA 2008/2009 145 2.3 63 NA 2009/2010 133 2.3 29 NA 2010/2011 139 2.4 52 NA 2011/2012 168 2.9 73 15 2012/2013 183 3.2 86 22 2013/2014 175 3.2 92 35 2014/2015 174 3.3 89 34 2015/2016 176 3.5 82 34 2016/2017 172 3.6 98 41 2017/2018 153 3.3 81 45 . Estimated Eskom procurement1 . Procurement as a proportion of GDP . Procurement from B-BBEE compliant suppliers2 . Procurement from black-owned (BO) suppliers2 . . Rbn, real . Percent . Percent . Percent . 1998–2007/2008 (average) 46 0.8 NA NA 2008/2009 145 2.3 63 NA 2009/2010 133 2.3 29 NA 2010/2011 139 2.4 52 NA 2011/2012 168 2.9 73 15 2012/2013 183 3.2 86 22 2013/2014 175 3.2 92 35 2014/2015 174 3.3 89 34 2015/2016 176 3.5 82 34 2016/2017 172 3.6 98 41 2017/2018 153 3.3 81 45 Source: Eskom financial statements. 1Estimated from company annual financial statements by subtracting EBITDA and labour costs from revenue, and adding capital expenditure. 2Stated in company annual reports as a proportion of Total Managed Procurement Spend. Open in new tab Table 1 Eskom procurement, 1998–2017/2018 . Estimated Eskom procurement1 . Procurement as a proportion of GDP . Procurement from B-BBEE compliant suppliers2 . Procurement from black-owned (BO) suppliers2 . . Rbn, real . Percent . Percent . Percent . 1998–2007/2008 (average) 46 0.8 NA NA 2008/2009 145 2.3 63 NA 2009/2010 133 2.3 29 NA 2010/2011 139 2.4 52 NA 2011/2012 168 2.9 73 15 2012/2013 183 3.2 86 22 2013/2014 175 3.2 92 35 2014/2015 174 3.3 89 34 2015/2016 176 3.5 82 34 2016/2017 172 3.6 98 41 2017/2018 153 3.3 81 45 . Estimated Eskom procurement1 . Procurement as a proportion of GDP . Procurement from B-BBEE compliant suppliers2 . Procurement from black-owned (BO) suppliers2 . . Rbn, real . Percent . Percent . Percent . 1998–2007/2008 (average) 46 0.8 NA NA 2008/2009 145 2.3 63 NA 2009/2010 133 2.3 29 NA 2010/2011 139 2.4 52 NA 2011/2012 168 2.9 73 15 2012/2013 183 3.2 86 22 2013/2014 175 3.2 92 35 2014/2015 174 3.3 89 34 2015/2016 176 3.5 82 34 2016/2017 172 3.6 98 41 2017/2018 153 3.3 81 45 Source: Eskom financial statements. 1Estimated from company annual financial statements by subtracting EBITDA and labour costs from revenue, and adding capital expenditure. 2Stated in company annual reports as a proportion of Total Managed Procurement Spend. Open in new tab Eskom’s procurement budget afforded huge potential influence. It rose from an estimated 1998–2007/08 average of around 0.8 percent of GDP to over 3 percent of GDP after 2012 (Table 1).91 The DTI described parastatal investment as the ‘single largest opportunity to stimulate industry’, with selective procurement enabling local manufacturers to grow and black-owned firms to overcome entry-barriers.92 Achieving this would, proponents argued, require corresponding revisions to parastatal governance, with executives’ viewed as harbouring ‘resistance’ to transformation initiatives, adhering instead to ‘narrow commercial considerations’ in procurement decisions.93 Increased political intervention was necessary, with parastatal capital investment being, as one former Minister of Public Enterprises put it, ‘our money and we have the right to leverage it to achieve what we want to achieve’.94 This coincided with a wider challenging of the ‘detached formalism’ of public procurement regulations, which critics argued implicitly favoured established white businesses and provided insufficient support to developmental aims.95 However, initiatives to instrumentalize Eskom took place in a context of fragmented governance arrangements and institutional constraints, from which a range of entities with competing agendas could exert influence.96 Corporatization made Eskom a public company subject to the Companies Act, with a conventional corporate governance structure with a board of directors. Its single shareholder, the DPE, sets its mandate via shareholder compacts that, notionally, balance financial sustainability with universal access, affordability and an unspecified ‘developmental role’.97 The Department of Mineral Resources and Energy (DME, previously Department of Energy) creates energy policy, determining Eskom’s generation technology. Tariffs are controlled by the National Electricity Regulator (NERSA), balancing Eskom’s commercial needs with its interpretation of the social interest.98 The Treasury overseas compliance with financial governance regulations for public entities, and the DTI overseas BEE codes.99 Added to these formal institutional constraints, the context of extreme inequality, consolidated private ownership and low economic dynamism places significant pressure on state agencies as means for facilitating accumulation, particularly around employment and public procurement. Given the effective party-state, this has created ‘informal patronage-based political networks working in parallel with, sometimes in opposition to, the impersonal political institutions of the state’.100 Fierce contestation of state institutions occurs at multiple levels, with corruption frequently intersecting with political factionalism.101 As the following section shows, this has been particularly acute in Eskom’s case. From developmental to captured state? Eskom’s crises and the contestation of parastatal procurement In its anticipated role as an instrument of a developmental state, Eskom was envisaged to provide an economic stimulus through increased investment and improved electricity access, while leveraging procurement for economic transformation. The outcomes were very different, as Eskom descended into a series of severe crises, driven by interwoven technical, economic and political processes. As the preceding section showed, these processes have deep historical roots, but the key trigger was the aforementioned delay to building new generation capacity as a result of the privatization saga of the early 2000s. Electricity consumption growth averaged 2.5 percent annually 1994–2008,102 and with Eskom’s spare capacity eroded, a coal supply crisis in 2008 caused unprecedented load-shedding.103 The economic impacts were devastating and under political pressure, particularly ahead of the 2010 World Cup, Eskom adopted a ‘keep-the-lights-on’ strategy, deferring plant maintenance and running generators ‘despite them having known defects’.104 Relief hinged on two 4.8GW coal-fired power station megaprojects, Medupi and Kusile. Both encountered catastrophic delays. In 2008, Eskom anticipated completion of Medupi’s six generating units by 2015, and Kusile’s the following year.105 By 2014 neither was producing any electricity. They are still incomplete at the time of writing. Rather than a single ‘turn-key’ contract, Eskom opted to coordinate construction through multiple contract-packages to increase influence over local supplier development.106 This proved beyond Eskom’s managerial capabilities. Construction was impeded by design flaws, strikes, skill shortages, financing shortfalls, contractor failures, and massive cost inflation around which there have been multiple corruption allegations—most notably relating to delayed boiler-work by Hitachi Power Africa, which was quarter-owned by its BEE partner, the ANC’s funding vehicle, Chancellor House, between 2005 and 2014.107 Faults, meanwhile, have proliferated in the ageing generation fleet, in which coal-fired plants now average 37-years old. Eskom’s energy availability factor—a measure of power plant availability—fell precipitously, to as low as around 70 percent in 2015/2016, down from 90 percent in the early 2000s. Outages returned in 2014/2015, with further equipment breakdowns and an energy availability factor below 70 percent in 2018/2019 bringing more load-shedding. Alongside becoming unreliable, electricity became dramatically more expensive. After 2007/2008, real electricity prices soared as NERSA granted successive above-inflation tariff increases, doubling in under a decade to cover Eskom’s increased costs (Figure 2). Figure 2 Open in new tabDownload slide Eskom tariff increases and electricity prices, 1994–2016/2017. Source: NERSA data; Eskom annual reports; International Monetary Fund. Figure 2 Open in new tabDownload slide Eskom tariff increases and electricity prices, 1994–2016/2017. Source: NERSA data; Eskom annual reports; International Monetary Fund. Cost overruns meant these massive revenue increases could not prevent a financial crisis. During 2013/2014–2017/2018 Eskom’s ZAR143bn project capital expenditure was over double the amount approved by NERSA, and Medupi and Kusile’s final costs are estimated between double and treble the R160bn initially budgeted.108 This necessitated dramatically increased borrowing. Debt quadrupled in a decade to ZAR389bn in 2017/2018, equivalent to 8 percent of GDP (Table 2). Concurrently, operating costs exploded. Primary energy costs, the largest operating expense, quadrupled per-unit of electricity sold quintupled from a 1997 to 2007 average of 8 cents per kilowatt-hour (c/kWh) to over 40c/kWh (Table 3). Coal costs are the largest contributor to this, for reasons discussed further below. Massive expenditure on diesel for open-cycle gas turbines to cover plant shutdowns, and new power-purchase agreements for private renewable energy producers, also contributed significantly. Labour costs per-unit doubled to 15c/kWh, as employee numbers increased around 50 percent to over 47,000 in the decade to 2016. This eroded profitability, and after 2014 Eskom struggled to generate sufficient income simply to cover its escalating interest costs (Tables 2 and 3). Table 2 Eskom debt and leverage, 1997–2017/2018 . Total debt . Debt-equity ratio1 . Interest cover ratio2 . Eskom debt to South African GDP . . ZARbn (real) . . . Percent . 1997–2006/2007 (average) 79 1.5 2.3 3 2007/2008 91 1.7 0.3 2.4 2008/2009 121 2.3 -3 3.1 2009/2010 161 2.5 1.3 4.2 2010/2011 234 2.8 2 5.8 2011/2012 254 2.7 2.6 6 2012/2013 267 3 1 6.2 2013/2014 317 3.2 1 7.2 2014/2015 349 3.7 0.7 7.8 2015/2016 362 2.6 0.7 8 2016/2017 374 3 0.5 8.2 2017/2018 389 3.3 0.6 8.4 . Total debt . Debt-equity ratio1 . Interest cover ratio2 . Eskom debt to South African GDP . . ZARbn (real) . . . Percent . 1997–2006/2007 (average) 79 1.5 2.3 3 2007/2008 91 1.7 0.3 2.4 2008/2009 121 2.3 -3 3.1 2009/2010 161 2.5 1.3 4.2 2010/2011 234 2.8 2 5.8 2011/2012 254 2.7 2.6 6 2012/2013 267 3 1 6.2 2013/2014 317 3.2 1 7.2 2014/2015 349 3.7 0.7 7.8 2015/2016 362 2.6 0.7 8 2016/2017 374 3 0.5 8.2 2017/2018 389 3.3 0.6 8.4 Source: Eskom financial statements; South African Reserve Bank (SARB) Quarterly Bulletin. 1Calculated by total liabilities/equity, representing levels of leverage. That is, the extent to which the company finances its assets by debt or equity capital. A higher ratio indicates a higher level of leverage and therefore financial risk. 2Earnings before interest and taxation/interest expenses, representing the company’s ability to cover interest expenses through the earnings it generates. A ratio > 1 indicates earnings are sufficient to cover interest. Open in new tab Table 2 Eskom debt and leverage, 1997–2017/2018 . Total debt . Debt-equity ratio1 . Interest cover ratio2 . Eskom debt to South African GDP . . ZARbn (real) . . . Percent . 1997–2006/2007 (average) 79 1.5 2.3 3 2007/2008 91 1.7 0.3 2.4 2008/2009 121 2.3 -3 3.1 2009/2010 161 2.5 1.3 4.2 2010/2011 234 2.8 2 5.8 2011/2012 254 2.7 2.6 6 2012/2013 267 3 1 6.2 2013/2014 317 3.2 1 7.2 2014/2015 349 3.7 0.7 7.8 2015/2016 362 2.6 0.7 8 2016/2017 374 3 0.5 8.2 2017/2018 389 3.3 0.6 8.4 . Total debt . Debt-equity ratio1 . Interest cover ratio2 . Eskom debt to South African GDP . . ZARbn (real) . . . Percent . 1997–2006/2007 (average) 79 1.5 2.3 3 2007/2008 91 1.7 0.3 2.4 2008/2009 121 2.3 -3 3.1 2009/2010 161 2.5 1.3 4.2 2010/2011 234 2.8 2 5.8 2011/2012 254 2.7 2.6 6 2012/2013 267 3 1 6.2 2013/2014 317 3.2 1 7.2 2014/2015 349 3.7 0.7 7.8 2015/2016 362 2.6 0.7 8 2016/2017 374 3 0.5 8.2 2017/2018 389 3.3 0.6 8.4 Source: Eskom financial statements; South African Reserve Bank (SARB) Quarterly Bulletin. 1Calculated by total liabilities/equity, representing levels of leverage. That is, the extent to which the company finances its assets by debt or equity capital. A higher ratio indicates a higher level of leverage and therefore financial risk. 2Earnings before interest and taxation/interest expenses, representing the company’s ability to cover interest expenses through the earnings it generates. A ratio > 1 indicates earnings are sufficient to cover interest. Open in new tab Table 3 Eskom costs and profitability, 1997–2017/2018 . Primary energy costs1 . Labour costs . Net profit margin . Return on assets2 . . c/kWh (real) . c/kWh (real) . Percent . Percent . 1997–2006/2007 (average) 8 8 12 4 2007/2008 15 9 0 0 2008/2009 19 11 −17 −5 2009/2010 20 10 5 2 2010/2011 23 11 9 3 2011/2012 29 12 12 3 2012/2013 37 14 4 1 2013/2014 40 15 5 1 2014/2015 45 14 2 1 2015/2016 44 15 3 1 2016/2017 41 16 1 0 2017/2018 40 14 −1 0 . Primary energy costs1 . Labour costs . Net profit margin . Return on assets2 . . c/kWh (real) . c/kWh (real) . Percent . Percent . 1997–2006/2007 (average) 8 8 12 4 2007/2008 15 9 0 0 2008/2009 19 11 −17 −5 2009/2010 20 10 5 2 2010/2011 23 11 9 3 2011/2012 29 12 12 3 2012/2013 37 14 4 1 2013/2014 40 15 5 1 2014/2015 45 14 2 1 2015/2016 44 15 3 1 2016/2017 41 16 1 0 2017/2018 40 14 −1 0 Source: Eskom financial statements. 1Fuel consumed in the generation of energy, alongside water, environmental levies, power buy-backs, co-generation, and international purchases. 2Net profit margin/total assets. Open in new tab Table 3 Eskom costs and profitability, 1997–2017/2018 . Primary energy costs1 . Labour costs . Net profit margin . Return on assets2 . . c/kWh (real) . c/kWh (real) . Percent . Percent . 1997–2006/2007 (average) 8 8 12 4 2007/2008 15 9 0 0 2008/2009 19 11 −17 −5 2009/2010 20 10 5 2 2010/2011 23 11 9 3 2011/2012 29 12 12 3 2012/2013 37 14 4 1 2013/2014 40 15 5 1 2014/2015 45 14 2 1 2015/2016 44 15 3 1 2016/2017 41 16 1 0 2017/2018 40 14 −1 0 . Primary energy costs1 . Labour costs . Net profit margin . Return on assets2 . . c/kWh (real) . c/kWh (real) . Percent . Percent . 1997–2006/2007 (average) 8 8 12 4 2007/2008 15 9 0 0 2008/2009 19 11 −17 −5 2009/2010 20 10 5 2 2010/2011 23 11 9 3 2011/2012 29 12 12 3 2012/2013 37 14 4 1 2013/2014 40 15 5 1 2014/2015 45 14 2 1 2015/2016 44 15 3 1 2016/2017 41 16 1 0 2017/2018 40 14 −1 0 Source: Eskom financial statements. 1Fuel consumed in the generation of energy, alongside water, environmental levies, power buy-backs, co-generation, and international purchases. 2Net profit margin/total assets. Open in new tab A vicious cycle emerged of rising costs, rising indebtedness, and rising interest payments feeding back into rising costs. Eskom shifted the burden onto other social actors. Firstly, to consumers through further tariff increases. This, however, drew opposition from powerful interest groups, including business associations for energy-intensive consumers like the EIUG and Chamber of Mines (now Minerals Council of South Africa), and trade unions. After initial largesse, NERSA increasingly sought to moderate tariff increases, granting increments well below Eskom’s requests. It cites Eskom management’s ‘unwillingness’ to control costs, and alarm at the socio-economic repercussions of high electricity prices given stalling growth and persistent energy poverty.109 Indeed, with free basic electricity provision limited, tariff increases have disproportionately impacted low-income households.110 They also exacerbated the MEC’s difficulties in a context of low mineral prices, with South Africa’s industrial energy costs rising above other emerging economies.111 As the DTI emphasized, this has undermined ‘developmental state’ initiatives by diminishing the competitiveness of strategic energy-intensive industries in which South Africa has strong capabilities.112 Compounding these problems, higher tariffs appear to have weakened electricity demand. Consumption declined after 2011/2012, particularly in mining and industrial consumer categories that comprise 40 percent of electricity sales (Figure 3), where large firms are increasingly seeking independent off-grid generation. Municipal sales have also slumped. Municipal distributors buy roughly half of Eskom’s domestic output, and re-sell at a mark-up to generate around 30 percent of their revenue.113 They have become increasingly dysfunctional intermediaries, adding to Eskom’s financial strains. A growing number are in default, with arrear debt from the top-10 defaulting municipalities and Soweto at ZAR36.5bn in mid-2019 greater than the ZAR20.7bn loss Eskom made that year.114 Figure 3 Open in new tabDownload slide Index of electricity sales 1994–2017/2018, GWh, 1994 = 100.115 Source: Eskom annual reports. Figure 3 Open in new tabDownload slide Index of electricity sales 1994–2017/2018, GWh, 1994 = 100.115 Source: Eskom annual reports. These factors combined have created a ‘utility death spiral’ where declining sales necessitate further price increases to sustain revenues, precipitating further sales declines.116 This combined with the fraught political economy of tariff increases has necessitated multiple state bailouts to delay bankruptcy. A ZAR60bn 2008 Treasury loan was in 2015 converted to equity, alongside a ZAR23bn injection funded by selling state shares in Vodacom.117 ZAR176bn in debt guarantees were granted in 2009, raised to ZAR350bn in 2010. Further ZAR69bn and ZAR59bn support packages were agreed in 2019. These contingent liabilities and the likelihood of further bailouts have made Eskom a leading source of fiscal risk, bringing Eskom firmly onto the public sector balance sheet in the eyes of financial markets. The company featured prominently in credit rating agency rationales for downgrading South Africa’s sovereign debt to junk-status in 2017.118 Eskom correspondingly became dependent on public creditors as commercial financiers grew wary.119 Loans from development finance institutions grew from 7 percent of total debt in 2010/11 to 31 percent in 2017/18.120 In bond markets, which account for half of Eskom’s borrowing, it increasingly leant on the Public Investment Corporation (PIC), which invests the Government Employees Pension Fund (GEPF). It owned approximately 60 percent of Eskom’s bonds, by value, in 2016/2017, with its R84bn exposure up from R21bn in 2008/2009.121 Citing its developmental mandate, the PIC compensated for faltering private investor confidence by supporting private bond placements and granting emergency loans.122 Through this, Eskom became a precarious nexus for the circulation of public funds to address its out-of-control cost base. Underlying this was a governance crisis arising from intense, and often politicized contestation of Eskom procurement. As discussed above, policymakers proposed using parastatal procurement essentially as a rent allocation mechanism supporting industrial policy and BEE. Judged against stated aims, the outcomes were disappointing. Considering industrial policy, while Eskom annual reports typically claimed success in this arena in terms of local content levels achieved in the Medupi and Kusile construction programme,123 various official analyses criticized performance in implementing its supplier development programmes, with continued high import penetration for both key capital goods and basic inputs, and serious difficulties encountered by local contractors.124 The DTI’s assessment in 2018 was that: Effectively, South Africa’s largest rail and energy infrastructure expansion programmes have not translated into the development of robust local supply chains. Despite government intervention frameworks … local suppliers are yet to be significantly integrated into key OEMs’ [original equipment manufacturers] global supply chains.125 Economy-wide, industrial decline continued alongside massive parastatal investment increases. After 2006, parastatal investment doubled as a proportion of overall fixed investment to 20 percent (Figure 4), driven by Eskom and Transnet. However, after the 2008 global financial crisis manufacturing investment declined in absolute and relative terms, surpassed by parastatal investment for the first time since the mid-1980s (Figure 4). Figure 4 Open in new tabDownload slide SOE and manufacturing gross-fixed capital formation. Source: SARB Quarterly Bulletin. Figure 4 Open in new tabDownload slide SOE and manufacturing gross-fixed capital formation. Source: SARB Quarterly Bulletin. BEE outcomes have been highly contentious. Eskom raised targets for procurement from Broad-Based BEE legislation (B-BBEE) compliant firms and raised those for majority black-owned firms, with the latter introduced in 2012/2013 at 10 percent and increased to 40 percent after 2015/2016, and the former raised above 50 percent in 2012/2013 to 70 percent, and then to 80 percent in 2015/2016.126 Performance on both measures increased significantly (Table 3) but the headline figures conceal complex underlying processes. Critics of B-BBEE legislation contend corporate compliance with ownership requirements often merely incorporates BEE partners as passive, minority equity owners, dependent on established businesses.127 Some policymakers correspondingly criticized parastatals for favouring formally-compliant large-scale or foreign firms over smaller, independent domestic companies.128 Coal was the main focus of efforts to instrumentalize Eskom’s procurement for BEE. Eskom purchases around half the national coal output, including low-quality coal that is harder to export. This gives it considerable power to shape the supply chain in what has been the largest mining sub-sector by sales value during much of the past decade. Mining, as part of the economy’s ‘commanding heights’, has been a key target for BEE policymaking, but accumulation by independent black-owned firms has been hindered by high entry-barriers. Even with unbundling of apartheid-era conglomerates, consolidation in the post-apartheid coal industry remained high, with four to five companies consistently accounting for around three-quarters of output.129 Eskom procurement presented a means to lower coal mining entry barriers and accelerate ownership transfer. Concerns in the late-2000s of rapid export growth undermining Eskom’s supply security, and difficulties agreeing new tied-mine investments, added to the impetus for diversification.130 Particularly after the 2008 load-shedding crisis, which was blamed in part on coal supply problems, Eskom increasingly used short-to-medium term supply contracts (SMTSCs), while allowing tied-mine supply to decrease. Negligible in the early 2000s, SMTSCs reached 37 percent of Eskom supply by 2015, lowering entry barriers for smaller mining and trading companies.131 The ‘black emerging miners’ (BEM) strategy targeted procurement of half Eskom’s coal from majority black-owned companies by 2018, and, subsequently, introduced majority black-ownership in new coal supply agreements.132 This took Eskom beyond the Mining Charter, which requires 26 percent black ownership. It spurred growth among smaller-scale coal mining, trading, and transport companies, with Eskom reporting BEM procurement—small scale emerging miners—increasing from 6 percent in 2012 to 18 percent in 2015.133 Meanwhile, major companies were compelled to raise black ownership levels, pause investment, or consider selling operations outright following difficulties obtaining new long-term contracts. The multinationals Anglo American and South32 have recently divested their Eskom-linked coal assets to the majority black-owned Seriti. However, higher logistics expenses and organizational complexity arising from multiple, dispersed small-scale producers delivering by road exacerbated difficulties controlling primary energy costs and managing coal quality.134 Leaked internal reports and testimony from a Parliamentary enquiry suggests competitive tendering was routinely discarded for SMTSCs, with high levels of buyer discretion.135 Subsequent analyses found multiple coal contracts to be excessively priced, with seven suppliers achieving profit margins of around 100 percent.136 BEE ownership requirements were, reportedly, frequently disregarded in practice, with BEM targets missed.137 Given coal mining’s limited potential for upgrading industrial capabilities, and its sunset industry status given climate change mitigation imperatives and falling renewables costs, clean energy would have been a more appropriate target for rent allocation from an industrial policy perspective. However, aligning with vested interests in coal, Eskom played an obstructive role in the Department of Energy’s Renewable Energy Independent Power Producer Procurement Programme (REIPPPP).138 This included stalling South Africa’s nascent renewables industry in 2016 by refusing to sign new REIPPPP power-purchase agreements. Accompanying energy sector reforms that would erode Eskom’s monopoly position in energy production stalled, such as the Independent Systems and Markets Operator Bill, or liberalization of licencing for small-scale independent generation. In recreating coal interest groups and stymying growth in renewables, Eskom ultimately indirectly hampered structural transformation and divergence from South Africa’s carbon-intensive, MEC-centred economic trajectory.139 Alongside these various failures against stated aims for instrumentalizing Eskom’s procurement, the company became embroiled in multiple high-profile corruption scandals. Most notably these involved the Gupta family—the aforementioned influential businessmen and key Zuma allies140—and their business associates, outlined in an explosive 2016 Public Protector investigation and a subsequent Parliamentary Enquiry.141 The investigation found Eskom ‘severely prejudiced’ Optimum Coal Holdings (OCH), owned by Swiss multinational Glencore, through fines and lowered coal prices ‘for the purposes of forcing [them] into business rescue and financial distress’.142 Eskom subsequently supplied a company part-owned by Zuma’s son and the Guptas an irregular ZAR659m prepayment for the ‘sole purpose’ of funding OCH’s takeover.143 This led to then-chief executive Brian Molefe’s November 2016 resignation. The range of controversies over the Zuma-Gupta network’s relationship with Eskom (and other parastatals) are too extensive to detail here, but extended beyond coal, encompassing payments to the Gupta-owned New Age newspaper, and consultancy firm Trillian, which was involved in the award of an irregular boiler tender to Chinese electrical contractor, Dongfang.144 The rhetoric of ‘radical economic transformation’ was frequently employed as a rationale or defence for Eskom’s actions, but in practice BEE appeared a lower-order priority. For example, Dongfang was incompliant with the B-BBEE Act, and Tegata only became majority black-owned retrospectively.145 Haroon Bhorat et al. argue commonalities among these events at Eskom and controversies at other parastatals, notably Transnet and Denel, represented systematic state capture by a ‘Zuma-centred power elite’, in which the Guptas played a pivotal brokerage role and parastatal procurement was the principle target.146 Nevertheless, Eskom’s procurement irregularities pre-dated and extended well beyond this specific elite network, notably including a range of overseas multinational enterprises.147 Leaked internal investigations outlined multiple irregular procurement practices, particularly in primary energy, warning some managers ‘have followed an agenda which ostensibly serves to favour suppliers at the expense of Eskom’, with unwillingness to reduce procurement costs due to corrupt relationships.148 This was formally acknowledged in Eskom’s 2016/2017 and 2017/2018 Financial Statements, released with qualified audits due to ZAR23bn reportable irregular expenditure. As of late-2018, 11 criminal cases and eight regulatory investigations had opened around corruption allegations, with 12 senior executives among 97 employees exiting the company in consequence, alongside an ongoing Special Investigations Unit investigation into large-scale corruption in the Medupi and Kusile construction programme procurement.149 Rather than a pocket of effectiveness strategically and systematically allocating rents through procurement to accelerate BEE and re-industrialization, there was instead growing disorder and informalization, with ambiguous objectives and accountability measures. This was exploited by a variety of internal and external interest groups, the Zuma-Gupta network being the most prominent, as Eskom became the object of fierce, destabilizing contestation. One former executive highlighted divisions between politically-appointed board members and managements’ operational concerns: There are two conversations at the board, on one level the board are engaged in transformation, and the executives are trying to run a power system: the mismatch became very difficult to manage.150 Procurement, they said: …gets a lot of attention from outside, you are not able to make a decision any more as an executive about what is good for the company. Like, you try two suppliers. You choose. Actually, there is someone out there doing all sorts of shenanigans to get their contract higher on the list. You are being manipulated by major suppliers outside. The [procurement] monster cannot be controlled. You as an executive, to put it bluntly, are absolutely like the ham in the sandwich.151 Another former executive discussing the matter described how Ministers: …insisted on approvals that quite frankly went quite deep into the operations of the organization...there was an element of fear that came into the organization. People were concerned about losing their jobs if they did not do what the Minister said they must do. The Chief Executive could do nothing without checking with the minister. And political ties just got very, very long.152 Similar perspectives are relayed in testimony to the Eskom Parliamentary Enquiry.153 This contestation of procurement contributed to the difficulties in controlling costs of production inputs and the Medupi and Kusile construction programme discussed above. It also created rapid churn of senior management through resignations, suspensions and dismissals. Eskom had two chief executives between 1994 and 2007, but between that point and the time of writing there have been 12, with the average tenure dropping to around one year. Executive appointments after 2010 became highly politicized, with board reshuffles in 2011 and 2014, and executive appointments in 2015/2016, controversial for deployment of politically-connected individuals.154 Former ministers and executives report presidential-level engagement with board appointments.155 Eskom’s financial and operational difficulties were intensified by this organizational instability. From 2016, major investors increasingly viewed Eskom’s dysfunctional governance arrangements as an additional credit risk, contributing to further credit-ratings downgrades and undersubscribed bond auctions.156 Potential covenant breaches for failure to produce clean audits risked precipitating a cascade of bond defaults, necessitating a bailout on a scale catastrophic to public finances. Eskom’s governance, operational and financial crises hereby became interwoven and mutually reinforcing. The scale of the dysfunction and severity of its consequences ultimately drew countervailing responses from other powerful actors in both the state and the private sector, enabled by the complex institutional arrangements for Eskom governance. Parastatal procurement—and the energy sector specifically—became a key battleground in the factional power struggles across the party-state during the late-Zuma period.157 Reflecting mounting concerns for public finances, its Public Financial Management Act mandate, and disposition towards conventional good governance in procurement, the Treasury took increasingly forthright attempts to intervene in parastatals.158 Their position was supported by more liberal ANC politicians. Acrimony centred around Treasury efforts to investigate parastatal procurement, and the proposed ZAR1 trillion nuclear energy procurement programme that would have formed the heart of Eskom’s future generation fleet. Divisions within and between Eskom, the ANC and other state agencies contributed to an effective paralysis of energy policymaking, reflected in uncertainty and delay around the Integrated Resource Plan, the superordinate electricity planning document, caused by dispute over the relative roles of coal, nuclear and renewables, and delays to the REIPPPP programme. The nuclear programme was beset by controversy over non-compliance with procurement legislation, unconstitutional cooperation agreement with the Russian government, and questions over demand forecasting and cost-competitiveness relative to renewables.159 Aggressively endorsed by Zuma’s allies, nuclear procurement was opposed by Zuma’s political opponents, who pointed to the Gupta family’s recently acquired interests in the uranium mining industry. It was halted by the Treasury as unaffordable, leading to Pravin Gordhan’s early-2017 removal as finance minister. This proved a watershed moment, as long-running contestation of the energy system transitioned into much more overt, public factional ANC struggle, fuelled by growing concern over the potential electoral impacts of persistent corruption scandals.160 Zuma’s opponents—led by Ramaphosa and Gordhan—organized around claims of ‘state capture’ and calls for parastatal reform, with mobilizations comprising the ANC’s Tripartite Alliance partners, the SACP and COSATU, and openly supported by established big-business through Business Unity South Africa and Business Leadership South Africa.161 These processes contributed to Ramaphosa’s victory at the 2017 ANC Electoral Conference, though at the time of writing are far from resolved. Eskom as a microcosm of tensions in the post-apartheid political economy As the preceding sections showed, efforts to instrumentalize Eskom failed against stated aims, with operational, financial, and governance crises intertwining such that Eskom actively undermined government’s declared developmental state agenda. This agenda was a sincere aspiration at its conception, at least for the policy intellectuals that developed it, but in implementation proved incompatible with the political settlement and was undermined and co-opted in a manner which undermined stated aims. In particular, Eskom’s crises have undermined South Africa’s energy-intensive industrial core and placed grave strains on the fiscus at a time when stimulus is needed to overcome prolonged economic stagnation. The contestation of Eskom’s procurement rents and the organizational instability it created, was central to this dysfunction. Within both public and academic discourse there has been widespread resort to the concepts of ‘neopatrimonialism’ and ‘state capture’ to explain these crises. This risks being reductionist, rendering the causes of Eskom’s dysfunction to the self-interested actions of a small elite network, generic in cause, and remediable through expulsion of implicated individuals to produce the autonomous ‘un-captured’ state. As the preceding material showed, Eskom’s failings result from a more complex confluence of historical processes. This included technical and economic factors as well as political ones. The company’s particular forms of institutional and inter-organizational embeddedness leading to a multiplicity of competing and sometimes contradictory objectives and distributional pressures. As discussed above, electricity production is embedded in both popular politics and the interests of established big business in a context of sustained inequalities in basic infrastructure provision and the energy-intensive path dependencies of the South African economy. Eskom has been unable to reconcile demands for cheaper electricity from major industrial users and ordinary citizens, the formal requirements for cost-recovery from users in the context of a massive and complex capital investment programme, and the requirements of creditors for a commercial return on investment. The contestation of Eskom’s procurement and of energy policy took place over a much longer period, and with a much wider range of actors than the clientelist networks associated with the Zuma-Gupta faction. This network exploited Eskom’s chaotic condition, and certainly contributed influentially to its intensification in the process. But this does not in itself adequately explain the scale and severity of Eskom’s dysfunction, and the broader failure of state-led development initiatives of the Zuma era that this has come to exemplify. As discussed in the first section, the success or otherwise of state-led economic development strategies derives not from the presence or absence of rent-seeking and clientelism, which is an inherent feature of capitalist development.162 Indeed, widespread rent seeking and clientelism are not in all instances inherently anti-developmental, nor indeed antithetical to the emergence of pockets of effectiveness in the state bureaucracy.163 As political settlements literature argues, more systemic clientelism and corruption of state organizations derives from imbalances and misalignments between the distribution of power and rents in society and its institutions.164 In South Africa, the highly concentrated economic structure, extreme inequalities, imbalances between political and economic power, and underlying tensions of the post-apartheid political settlement have meant the state faces strong pressure to facilitate accumulation in a context of limited opportunities, a pressure that has heightened in the context of prolonged economic stagnation.165 Combined with the failure of conventional BEE policies and the limited scope of redistributive measures, the situation has produced, as Karl von Holdt puts it, ‘an emerging informal political-economic system’ in which factional adherents view established financial regulations, governance procedures, and the Constitutional settlement itself, as illegitimate impediments sustaining an unjust order.166 The discourse of ‘radical economic transformation’ became a legitimation for some forms of un-developmental accumulation—in the sense that they did not advance the acquisition of new industrial capabilities—but ultimately, as Ivor Chipkin et al. argue, its resonance reflected, to a degree, economic realities and political convictions.167 While industrial policy initiatives variously labelled as part of a ‘developmental state’ or ‘radical economic transformation’ agenda were on one level co-opted and subverted for corrupt or clientelistic ends, at another the meaning of these terms is blurred by the disagreements over the means, ends and pace of economic transformation and the role of state power within this. As Alexander Beresford notes, in a South African context, state corruption is about more than simple greed and criminality, rather it is an intersection of accumulation, class formation, and factional political struggles.168 This is exemplified in the case of Eskom and other parastatals, but present in multiple aspects of the state, not least at the municipal level. This has also created growing incoherence in the South African state as regards economic policy, manifested prominently in parastatals. In particular, concerning contradictions between nationalist imperatives for accelerated redistribution and black capitalist class formation via asset transfers from the existing industrial base, and the ‘market-enhancing’ institutional framework and orthodox good governance of the post-apartheid transition.169 As Nimrod Zalk argues, industrial policy has been caught in a ‘pincer’ between these tendencies.170 Factional and ideological divisions within the Tripartite Alliance and party-state widened during the Zuma era, resulting in a range of contradictory agendas and dispersed coalitions of interest. This contestation of the South African state is inimical to the foundation of anything resembling the ‘developmental state’ imaginary articulated by ANC policy intellectuals since the mid-2000s, and so too for the emergence of pockets of effectiveness within the bureaucracy that would effectively guide industrial policy. Nonetheless, while far from the ANC’s envisaged developmental state, it is also far from approximating an ideal-type neo-patrimonial regime with ‘weakly centralized and barely legitimate authority structures, personalistic leaders unconstrained by norm or institutions, and bureaucracies of poor quality’.171 Indeed, as the events analysed in the fourth section and the accompanying fall of the Zuma administration show, despite the obvious weaknesses in oversight mechanisms for Eskom, key formal institutions, business actors and norms around conduct in public office remain extremely powerful checks on political power. Reading the Eskom crisis in this context, South Africa more closely approximates what Atul Kohli described as a ‘fragmented multi-class’ state in which the fragmentation of public authority means states ‘are not in a position to define their goals as narrowly or to pursue them as effectively [and] must typically pursue several goals simultaneously, as they seek to satisfy multiple constituencies’.172 Rather than a straightforward process of capture the events at Eskom represent a microcosm of these broader tensions in the post-apartheid state, the fierce and often contradictory distributional pressures accompanying them, and the opportunities for predatory activity this dysfunction creates. Conclusion The article has analysed the causes of Eskom’s inter-connected operational, financial, and governance crises, which have unfolded in the context of aspirations to transform parastatals into instruments of a South African developmental state. It has shown how, through these crises, Eskom undermined such aspirations, and became both a destabilizing force in the South African economy—notably for the electricity-intensive minerals-energy complex—and a focal point for factional battles within the Tripartite Alliance. Using a political settlements framework, it has argued that these crises reflect deeper tensions in the South African political economy, which have produced intense and long-running contestation of the parastatal and the rents it controls through its position in the electricity system. The success of industrial policy initiatives requires bureaucratic capability and autonomy of state implementation agencies, but also compatibility with the political settlement.173 This was not the case in this instance. Instead, South Africa’s persistent racialized inequalities and increasing factional and ideological fragmentation in politics generated incoherence within the state, delays in critical policy processes, and conflicting formal and informal distributional pressures and institutional constraints acting upon Eskom, which ultimately undermined effective governance. For this reason, and the important technical and economic path-dependencies reflected in the crises, the article has correspondingly argued that it would be reductionist to see the company’s crises as solely an outcome of ‘state capture’ corruption scandals, significant though these were.174 These crises remain unresolved at the time of writing. The early phases of Ramaphosa’s administration were marked by bold commitments to transform parastatal governance and prosecute high-level corruption within Eskom and other wayward parastatals. Two years after his rise to president however, Eskom’s crises are far from over. Load-shedding continues to severely disrupt day-to-day economic activity due to a proliferation of plant breakdowns, continued delays and faults at Medupi-Kusile and hold-ups to licencing and procurement of new impendent power production. The sustainability of the company’s financial position remains uncertain in spite of new bailouts, with the DPE calling it ‘the single largest risk to the South African economy’.175 Eskom’s disputes with the energy regulator over efforts to secure higher tariffs have escalated to litigation.176 With its balance sheet carrying ZAR441bn of debt, up from ZAR389bn the previous year, and interest payments consuming cash flows, radical financial and organizational restructurings are required.177 Plans for the organizational restructuring were released in October 2019, with Eskom’s vertically integrated model to be unbundled into three legally separate subsidiaries for generation, transmission, and distribution, by the end of 2022.178 The proposals effectively reincarnate the abandoned 1998 Energy Policy White Paper, and represent a pendulum swing towards more market-orientated electricity governance. The formation of the independent transmission company is expected to ‘foster a competitive market and … encourage the use of diverse sources of energy’ through competition between different Eskom generation units and new private producers that can ease capacity constraints.179 However, deep divisions within the Tripartite Alliance over the reforms—in particular disagreements over the role of private ownership, and the relative positions of coal, nuclear, and renewables in South Africa’s energy mix—raise serious questions about the implementation timelines. Most notably, key trade unions have firmly opposed unbundling and threatened industrial action over the issue, claiming it to be a path to full privatization that will create redundancies—a claim that President Ramaphosa denies. If a pathway through these challenges is not rapidly reached, the prolongation of Eskom’s dire condition will have important implications not only for the stagnant South African economy, but for the ANC’s increasingly fragile hegemony. Footnotes Andrew Bowman (andrew.bowman@ed.ac.uk) is a Lecturer in Africa and International Development at the Centre of African Studies, University of Edinburgh, and a Research Associate at the Society, Work and Politics Institute, University of the Witwatersrand. The author gratefully acknowledges the guidance and insights provided by peer reviewers and editors, and feedback from various colleagues and seminar attendees during the drafting of the article, particularly the detailed comments from Karl von Holdt, Gavin Capps, and Hazel Gray. 1 Author’s calculations, Eskom annual reports, various years. 2 E.g. Fitch Ratings’ ‘Fitch revises outlook on South Africa to negative’ 26 July 2019, <https://www.fitchratings.com/site/pr/10084089> (27 July 2019). 3 Meaning, in broad terms, capture of the state apparatus by private interests seeking to utilize state powers or resources to their advantage. The concept features commonly in South African political discourse, and is used to describe mutually-beneficial corrupt relations between former-President Jacob Zuma and his political allies, and the Gupta family, businesspersons with interests across a range of industries. There have been multiple high-profile corruption scandals involving this network, with parastatal companies Eskom, Transnet and Denel featuring prominently. These are typically alleged to have involved high-level political interference in the operations of state entities to the benefit of Gupta-owned or linked companies. As argued by Chipkin et al., this was systemic rather than a series of isolated instances, and formed the economic basis for a ‘Zuma-centred power elite’. This has been a key subject of a major public enquiry, the Judicial Commission of Inquiry into Allegations of State Capture, Corruption and Fraud in the Public Sector (the Zondo Commission), which is ongoing at the time of writing. See Ivor Chipkin, Mark Swilling, Haroon Bhorat, Mzukisi Qobo, Sikhulekile Duma, Lumkile Mondi, Camaren Peter, Mbongiseni Buthelezi, Hannah Friedenstein, and Nicky Prins, Shadow state: The politics of state capture (Wits University Press, Johannesburg, 2018). 4 This figure includes appointments of acting chief executives. 5 Comprising the ANC, South African Communist Party (SACP), and Congress of South African Trade Unions (COSATU). 6 See also Bill Freund, Twentieth century South Africa: A developmental history (Cambridge University Press, Cambridge, 2018); Nancy Clark, Manufacturing apartheid: State corporations in South Africa (Yale University Press, New Haven, 1994); Chipkin et al., Shadow state; Anne Pitcher, ‘Was privatization necessary and did it work? The case of South Africa’, Review of African Political Economy 39, 132 (2012), pp. 243–260. 7 Roger Southall, ‘The ANC, Black Economic Empowerment and state-owned enterprises: a recycling of history?’, in Sakhela Buhlungu, John Daniel, Roger Southall, Jessica Lutchman (eds), State of the nation: South Africa 2007 (HSRC Press, Cape Town, 2007), pp. 201–225. 8 BEE legislation defines ‘black’ as ‘a generic term which means Africans, Coloureds and Indians’, see Republic of South Africa (RSA), B-BBEE Amendment Act (46/2013). This definition is employed here. 9 Meaning that responsibility for the company’s losses has been passed on to wider society, in this case through bailouts and other forms of state support. 10 Ben Fine and Zavareh Rustomjee, The political economy of South Africa: From minerals energy complex to industrialisation (Hurst, London, 1996). 11 Karl von Holdt, ‘The political economy of corruption: Elite formation, factions and violence’ (Working paper, Society, Work and Politics Institute, University of the Witwatersrand, 2019); Chipkin et al., Shadow state. 12 Pritish Behuria, Lars Buur and Hazel Gray, ‘Studying political settlements in Africa’, African Affairs 116, 464 (2017), pp. 508–525. 13 Lindsay Whitfield, Ole Therkildsen, Laars Buur and Anne Mette Kjaer, The politics of African industrial Policy: A comparative perspective (Cambridge University Press, Cambridge, 2015); Hazel Gray, ‘Industrial policy and the political settlement in Tanzania’, Review of African Political Economy 40, 136 (2013), pp. 185–201. 14 Hazel Gray ‘The political economy of grand corruption in Tanzania’, African Affairs 114, 456 (2015), pp. 382–403. 15 von Holdt, ‘The political economy of corruption’; Roger Southall, ‘The coming crisis of Zuma’s ANC: the party state confronts fiscal crisis’, Review of African Political Economy 43, 147 (2016), pp. 73–88. 16 Ibid. 17 Glenn Morgan and Peer Hull-Kristensen, ‘The contested space of multinationals: Varieties of institutionalism, varieties of capitalism’, Human Relations 59, 11 (2006), pp. 1467–149. 18 Henry Wai-Chung Yeung, ‘The firm as social network: An organizational perspective’, Growth and Change 36, 3 (2005), pp. 307–328; Neil Coe and Henry Wai-Chung Yeung, Global production networks: Theorizing economic development in an interconnected world (Oxford University Press, Oxford, 2015), p. 58. 19 From a large literature on the topic, see Jewellord Ne Singh and Jesse Salah-Ovadia (eds), Developmental states beyond East Asia (Routledge, London, 2019); Laura Routley, ‘Developmental states in Africa? A review of ongoing debates and buzzwords’, Development Policy Review 32, 2 (2014), pp. 159–177. 20 Chalmers Johnson, MITI and the Japanese miracle: The growth of industrial policy, 1925–1975 (Stanford University Press, Stanford, CA, 1982). 21 E.g. Alice Amsden, The rise of the rest: Challenges to the West from late-industrializing economies (Oxford University Press, Oxford, 2001); Peter Evans, Embedded autonomy: States and industrial transformation (Princeton University Press, Princeton, NJ, 1995), p. 12. 22 Clark, Manufacturing apartheid. 23 Ha Joon Chang, ‘State-owned enterprise reform’ (United Nations Department for Economic and Social Affairs Policy Notes, New York, 2007); Evans, Embedded autonomy; Amsden, Rise of the rest. 24 Ibid. 25 Ibid. 26 Roger Tangri, The politics of patronage in Africa: Parastatals, privatization and private enterprise (James Currey, Oxford, 1999). 27 Thandika Mkandawire, ‘Thinking about developmental states in Africa’, Cambridge Journal of Economics 25, 3 (2001), pp. 289–313. 28 Singh and Salah-Ovadia, Developmental states. 29 Tim Kelsall, Business, politics, and the state in Africa: Challenging the orthodoxies on growth and transformation (Zed, London, 2013); Pritish Behuria, ‘Between party capitalism and market reforms: understanding sector differences in Rwanda’, Journal of Modern African Studies 53, 3 (2015), pp. 415–450; Berhanu Abegaz, ‘Political parties in business: Rent seekers, developmentalists, or both?’, Journal of Development Studies 49, 11 (2013), pp. 1467–1483; Jesse Salah Ovadia, The petro-developmental state in Africa: Making oil work in Angola, Nigeria and the Gulf of Guinea (Hurst, London, 2016). 30 Aldo Musacchio and Sergio Lazzarini, Reinventing state capitalism: Leviathan in business, Brazil and beyond (Harvard University Press, Cambridge, MA, 2014). 31 David Kang, Crony capitalism: Corruption and development in South Korea and the Philippines (Cambridge University Press, Cambridge, 2002); Hazel Gray, Turbulence and order in economic development: Institutions and economic transformation in Tanzania and Vietnam (Oxford University Press, Oxford, 2018); Whitfield et al., The politics of African industrial policy. 32 Whitfield et al., The politics of African industrial policy, p. 3. 33 See also Kelsall, Business, politics, and the state in Africa; Mushtaq Khan, Political settlements and the governance of growth-enhancing institutions (Mimeo, SOAS, London, 2010); Thandika Mkandawire, ‘Neopatrimonialism and the political economy of economic performance in Africa: Critical reflections’, World Politics 67, 3 (2015), pp. 563–612. 34 Gray, Turbulence and order, p. 65. 35 See Gray, ‘Industrial policy’; Gray Turbulence and order; Whitfield et al., Politics of African industrial policy. 36 Ibid; Hazel Gray and Lindsay Whitfield, ‘Reframing African political economy: Clientelism, rents and accumulation as drivers of capitalist transformation’ (International Development Working Paper 14–159, London School of Economics, 2014). 37 Gray, ‘The political economy of grand corruption’. 38 Khan, Political settlements. 39 Ibid. 40 Whitfield et al, The politics of African industrial policy, p. 289. 41 Whitfield et al., The politics of African industrial policy; Behuria et al., ‘Studying political settlements’. 42 Sam Hickey, ‘The politics of state capacity and development in Africa: reframing “pockets of effectiveness”’ (Effective States and Inclusive Development Working Paper 117, University of Manchester, 2019). 43 Ricardo Soares de Oliveira. Magnificent and beggar land: Angola since the civil war (Oxford University Press, Oxford, 2015), pp. 25–49; Steffen Hertog, ‘Defying the resource curse: explaining successful state-owned enterprises in rentier states’, World Politics 62, 2 (2010), pp. 261–301. 44 Hickey, ‘The politics of state capacity’; Whitfield et al., The politics of African industrial policy; Gray and Whitfield, ‘Reframing African political economy’, p. 18. 45 Gray, Turbulence and order, p. 65. 46 Whitfield et al., The politics of African industrial policy, p. 12; Behuria et al., ‘Studying political settlements’. 47 For more comprehensive historical discussions, see Freund, Twentieth century South Africa, Andrew Marquard, The origins and development of South African energy policy (University of Cape Town, unpublished PhD dissertation, 2006); Renfrew Christie, Electricity, industry and class in South Africa (Macmillan, London, 1984); Alain Dubresson and Sylvy Jaglin, Eskom: Electricity and technopolitics in South Africa (UCT Press, Cape Town, 2016). 48 The company was named Escom originally but renamed ‘Eskom’ in 1987. 49 Christie, Electricity, industry and class. 50 Fine and Rustomjee, The political economy of South Africa. 51 Ibid; Clark, Manufacturing apartheid. 52 Freund, Twentieth century South Africa. 53 Fine and Rustomjee, The political economy of South Africa, pp. 183–184; Christie, Electricity, industry and class, pp. 150–172. 54 Ibid. 55 Author’s calculations from South African Reserve Bank (SARB) Quarterly Bulletin and United Nations Comtrade Database. 56 Author’s calculations, Eskom annual report data. 57 Freund, Twentieth century South Africa; Marquard, South African energy policy. 58 Ibid. 59 Anton Eberhard and Clive van Horen, Poverty and power: Energy and the South African state (Pluto, London, 1995). 60 Government of South Africa, ‘White paper on reconstruction and development’ (1994). 61 Bernard Bekker, Anton Eberhard, Trevor Gaunt and Andrew Marquard, ‘South Africa’s rapid electrification programme’, Energy Policy 36, 8 (2008), pp. 3125–3137. 62 Jackie Dugard, ‘Urban basic services: Rights, reality, and resistance’, in Malcolm Langford (ed.), Socio-economic rights in South Africa: Symbols or substance? (Cambridge University Press, Cambridge, 2013), pp. 275–309. 63 Pitcher, ‘Was privatization necessary’; Anton Eberhard, ‘From state to market and back again: South Africa’s power sector reforms’, Economic and Political Weekly 40, 50 (2005), pp. 5309–5317. 64 Department of Minerals and Energy (DME), ‘White paper on the energy policy of South Africa’ (DME, Pretoria, 1998). Interview, former Eskom executive 1, conducted remotely, 5 September 2016. 65 David McDonald (ed.), Electric capitalism: Recolonising Africa on the power grid (HSRC Press, Cape Town, 2009). 66 Eberhard, ‘From state to market’. 67 ANC, ‘52nd National Conference: Resolutions’ (ANC, Johannesburg, 2007). 68 For discussion, see Samantha Ashman, Ben Fine and Susan Newman, ‘Systems of accumulation and the evolving South African MEC’, in Ben Fine, Jyoti Saraswati and Daniela Tavasci (eds), Beyond the developmental state: Industrial policy into the 21st century (Pluto, London, 2013), pp. 245–267; Ben Fine (2016) ‘Across developmental state and social compacting: The peculiar case of South Africa’ (Working paper, Institute of Social and Economic Research, Stellenbosch University, 2016). 69 Pitcher, ‘Was privatization necessary’; Southall, ‘The coming crisis’. 70 Susan Booysen ‘Hegemonic struggles of the African National Congress: From cacophony of morbid symptoms to strained renewal’, Africa Spectrum 53, 2 (2018), pp. 5–35. 71 William Gumede, ‘Delivering a democratic developmental state in South Africa’ (Working Paper, Development Bank of Southern Africa, Development Planning Division, 2009); Alan Hirsch, ‘Aspirations to an elusive developmental state’, in Busani Ngcaweni (ed.), The Future we chose: emerging perspectives on the centenary of the ANC (Africa Institute of South Africa, Pretoria, 2013), pp. 297–310. 72 Okechukwu Iheduru, ‘Black economic power and nation-building in post-apartheid South Africa’, Journal of Modern African Studies 42, 1 (2004), pp. 1–30. 73 Shaukat Ansari, ‘The neoliberal incentive structure and the absence of the developmental state in post-apartheid South Africa’, African Affairs 116, 463 (2017), pp. 206–232. 74 Author’s calculations, World Bank World Development Indicators. 75 Lynne Thomas, ‘Ownership of JSE-listed companies: Research report for National Treasury’ (National Treasury, Pretoria, 2017). 76 Jason Bell, Sumayya Goga, Pamela Mondliwa, and Simon Roberts ‘Structural transformation in South Africa: Moving towards a smart, open economy for all’ (Industrial Development Think Tank, Johannesburg, 2018). 77 Roger Southall, Liberation movements in power: Party and state in Southern Africa (James Currey, Oxford, 2013), pp. 212–246 78 Chipkin, Shadow state; von Holdt, ‘The political economy of corruption’. 79 Bell et al., ‘Structural transformation’. 80 Southall, ‘The coming crisis’. 81 Ansari, ‘The neoliberal incentive structure’. 82 Hirsch, ‘Aspirations to an elusive developmental state’. 83 In particular, as Chipkin et al., Shadow state, note, since they are not required to submit budgets to parliament. See also Southall, ‘The coming crisis’, p. 76. 84 Author’s calculations, Eskom annual reports and South African Reserve Bank Quarterly Bulletin. 85 ANC, ‘53rd National Conference: Resolutions’ (ANC, Johannesburg, 2012), p. 24; Edwin Ritchken, ‘The evolution of state-owned enterprises in South Africa’ (Background Paper, Organization for Economic Cooperation and Development, Paris, 2014). 86 The Presidency of the Republic of South Africa, ‘Green paper: Role of state-owned entities in the developmental state’ (The Presidency, Pretoria, 2013), p. 35; Presidential Review Committee on State-owned Enterprises (PRCSOE), ‘Growing the economy, bridging the gap’ (The Presidency, Pretoria, 2013). 87 Ibid; EDD, ‘The new growth path’ (EDD, Pretoria, 2011); DPE, ‘Strategic plan 2010–13’ (DPE, Pretoria, 2010); DPE, ‘Strategic plan 2016/17–2020/21’ (DPE, Pretoria, 2016). 88 Ritchken, ‘Evolution’, pp. 2–3; PRCSOE, ‘Growing the economy’; Presidency, ‘Green paper’. 89 Ibid. 90 Ibid; Presidential Infrastructure Coordinating Commission, ‘A summary of the South African National Infrastructure Plan’ (The Presidency, Pretoria, 2012); EDD, ‘New growth path’; Key initiatives included the Competitive Supplier Development Programme and the Public Preferential Procurement Framework Act (PPPFA) amendment enabling designation of local-content requirements for parastatal procurement, see DTI, ‘Industrial policy action plan 2017/18–2019/20’ (DTI, Pretoria, 2017). 91 Authors calculations, Eskom annual financial statements. 92 DTI, ‘Industrial policy action plan 2012/13–2014/15’ (DTI, Pretoria, 2012), p. 33; DTI, ‘Black industrialists policy’ (DTI, Pretoria, 2015). 93 PRCSOE, ‘Balancing of social, political and economic imperatives in delivering objectives for SOEs’ (The Presidency, Pretoria, 2012). 94 DPE, ‘Public lecture by Malusi Gigaba, Minister of Public Enterprises, at Wits University on 7th June 2011’ (DPE, Pretoria, 2011); ANC, ‘National General Council discussion document: Transformation of state and governance’ (ANC, Johannesburg, 2010). 95 Ryan Brunette, Jonathan Klaaren and Patronella Nqaba, ‘Reform in the contract state: Embedded directions in public procurement regulation in South Africa’, Development Southern Africa 36, 4 (2019), pp. 537–554. 96 For a fuller discussion than is possible here, see Dubresson and Jaglin, Eskom, pp. 33–55; Lucy Baker, Jesse Burton, Catrina Godinho, Hilton Trollip, The political economy of decarbonization: Exploring the dynamics of South Africa’s electricity sector (Energy Research Centre, University of Cape Town, Cape Town, 2015). 97 RSA, Eskom Conversion Act (2001). 98 RSA, Electricity Regulation Act (2006). 99 Brunette et al., ‘Reform in the contract state’. 100 For further discussion, see Alexander Beresford, ‘Power, patronage, and gatekeeper politics in South Africa’, African Affairs 114, 455 (2015), pp. 226–248; Karl von Holdt, ‘Nationalism, Bureaucracy and the developmental state: The South African case’, South African Review of Sociology 41, 1 (2010), pp. 4–27; von Holdt, ‘The political economy of corruption’; Chipkin, Shadow state. 101 Ibid. 102 World Bank World Development Indicators. 103 Eskom, ‘Annual report 2007/08’ (Eskom, Johannesburg, 2008). 104 Eskom, ‘Annual report 2012/13’ (Eskom, Johannesburg, 2013). 105 Eskom, ‘Annual report 2007/08’. 106 Parliamentary Energy Committee, ‘Update on the Eskom-led new build programme’ 20 August 2012 <https://pmg.org.za/committee-meeting/14713/> (04 January 2020). 107 Dentons, ‘Report in respect of the investigation into the status of the business and challenges experienced by Eskom’ (Dentons, Cape Town, 2015); Parliament of South Africa, ‘Report of Standing Committee on Public Accounts on oversight visit to ESKOM and its projects of Medupi and Kusile’ (Parliament, Cape Town, 2019); U.S.-Securities and Exchange Commission, ‘SEC charges Hitachi with FCPA violations’, 28 September 2015 <https://www.sec.gov/news/pressrelease/2015-212.html> (4 January 2020). 108 NERSA, ‘Decision: Eskom’s revenue application for 2018/19’ (NERSA, Pretoria, 2018). 109 Interview, Thembani Bukula, executive manager, electricity, NERSA, Pretoria, 21 July 2016; NERSA, ‘Decisions’. 110 Ibid. 111 Ferro-Alloy Producers Association, ‘Impact of RCA adjustment on MYPD3’ (NERSA, Pretoria, 2016). 112 DTI, ‘Industrial policy action plan 2016/17–2018/19’ (DTI, Pretoria, 2016). 113 StatsSA, ‘Statistical release P9114: Financial census of municipalities’ (StatsSA, Pretoria, 2018). 114 Eskom, ‘Annual financial statements 2018/19’ (Eskom, Johannesburg, 2019). 115 Sales to metro and municipal redistributors, major industrial customers and the mining industry represent around three quarters of Eskom’s electricity sales. Other customer categories include international, agriculture, rail, and direct residential and commercial sales. 116 NERSA, ‘Decisions’, p. 29. 117 DPE, ‘Annual report 2015/16’ (DPE, Pretoria, 2016). 118 S&P, ‘South Africa long-term foreign currency rating cut to ‘BB+”, 3 April 2017 <https://www.standardandpoors.com/en_US/web/guest/article/-/view/type/HTML/id/1825424> (4 January 2020). 119 Futuregrowth Asset Management, ‘SOE governance unmasked’ (Futuregrowth, Cape Town, 2018). 120 Eskom annual financial statements, various years. 121 GEPF, ‘Annual Report 2016/17’ (GEPF, Pretoria, 2017). 122 PIC, ‘Annual Report 2016/17’ (PIC, Pretoria, 2017). 123 E.g. Eskom, ‘Annual Report 2011’ (Eskom, Johannesburg, 2011). 124 DTI, ‘Industrial policy action plan 2017/18’; Parliament, ‘Report of Standing Committee’; Zavareh Rustomjee, Lauralyn Kaziboni and Ian Steuart, ‘Structural transformation along metals, machinery and equipment value chain’ (Industrial Development Think Tank, Johannesburg, 2018); Lionel October, Director General, DTI, comments at Trade and Industrial Policy Strategies Annual Forum 2017, Midrand, 13 June 2017; Interview, former Department of Public Enterprises official, Johannesburg, 23 July 2017. 125 DTI, ‘Industrial policy action plan 2017/18’, p. 101. 126 Eskom Annual Reports, various years. 127 See Brunette, ‘Reform in the contract state’; Chipkin et al., Shadow State. 128 E.g. Nimrod Zalk, ‘Selling off the silver’ New Agenda 63 (2016), pp. 10–15. See also PRCSOE, ‘Key themes arising from the state of supply chain management within the state owned enterprises’ (PRCSOE, Pretoria, unknown year); Interview, former Department of Public Enterprises official, Johannesburg, 23 July 2017. 129 Author’s calculations, company annual reports and Department of Mineral Resources data. 130 South African Coal Roadmap, ‘South African Coal Roadmap’ (Fossil Fuel Foundation/South African National Energy Development Institute, Johannesburg, 2013). 131 Mudzi Marageni, ‘Eskom’s coal supply and the role of emerging miners’ (Eskom, Johannesburg, 2015). 132 Interview, coal procurement advisor, Johannesburg, 2 August 2017; Kannan Lakmeeharan, ‘Emerging miner strategy workshop’ (Eskom, Johannesburg, 2014); Eskom, ‘Eskom will continue to insist on procuring coal from 50%-plus black-owned coal miners’ 12 May 2015 <http://www.eskom.co.za/news/Pages/Dec5B.aspx> (10 July 2018). 133 Ibid. 134 Interview, former Eskom executive 1, conducted remotely, 5 September 2016; Interview, former Eskom executive 2, Johannesburg, 26 July 2016; Interview, Eskom coal procurement staff, 22 April 2015. 135 Portfolio Committee on Public Enterprises (PCPE), ‘Inquiry into governance, procurement and the financial sustainability of Eskom’ (PCPE, Cape Town, 2018); National Treasury, ‘Forensic investigation into various allegations at Transnet and Eskom’ (Funduzi/National Treasury, Pretoria, 2018); Dentons, ‘Report’. 136 Hansard (National Assembly), 3 December 2019, p. 161. 137 PCPE, ‘Inquiry into Governance’; Dentons, ‘Report’. 138 The political economy of REIPPPP, and the resilience of Eskom and coal interests to associated reform efforts, is analysed extensively by Lucy Baker. Among other works, see Lucy Baker ‘Commercial-scale renewable energy in South Africa and its progress to date’, IDS Bulletin 48, 5–6 (2017), pp. 101–118. 139 Ibid. 140 For a deeper political analysis of the mutually-dependent relationship between Zuma and his political allies and the Guptas, see Chipkin et al., Shadow state. 141 Public Protector, ‘State of capture: Report no: 6 of 2016/17’ (Public Protector, Pretoria, 2016). 142 Ibid, pp. 23–24 and 338. 143 Ibid, pp. 315–348. 144 See Anton Eberhard and Catrina Godinho, ‘Eskom inquiry reference book’ (UCT, State Capacity Research Project, Cape Town, 2017). 145 Public Protector, ‘State of Capture’; Bekezela Phakathi, ‘DA calls for review of BEE laws after explosive findings on Gupta-linked firms’ Business Day 25 September 2019 <https://www.businesslive.co.za/bd/national/2019-09-25-da-calls-for-review-of-bee-laws-after-explosive-findings-on-gupta-linked-firms/> (15 December 2019). 146 Haroon Bhorat, Mbongiseni Buthelezi, Ivor Chipkin, Sikhulekile Duma, Lumkile Mondi, Camaren Peter, Mzukisi Qobo, Mark Swilling, Hannah Friedenstein, Betrayal of the promise: How South Africa is being stolen (Public Affairs Research Institute, Johannesburg, 2017) 147 PCPE, ‘Inquiry into governance’; Treasury, ‘Forensic investigation’. 148 Dentons, ‘Report’, pp. 76–110. 149 Eskom, ‘System status briefing, 16 November 2018’ (Eskom, Johannesburg, 2018). 150 Interview, former Eskom executive 2, Johannesburg, 26 July 2016. 151 Ibid. 152 Interview, former Eskom executive 1, conducted remotely, 5 September 2016. 153 PCPE, ‘Inquiry into governance’. 154 Public Protector, ‘State of capture’; Eberhard and Godinho, ‘Eskom inquiry’. 155 PCPE, ‘Inquiry into governance’; Public Protector, ‘State of capture’, p. 90. 156 Futuregrowth, ‘SOE governance unmasked’. 157 von Holdt, ‘The political economy of corruption’. 158 Ibid; Chipkin et al., ‘Shadow state’; Brunette, ‘Reform in the contract state’. 159 Baker, ‘Commercial scale renewable energy’. 160 von Holdt, ‘The political economy of corruption’. See also ANC, ‘Legislature and governance discussion document’ (Discussion document, ANC, 5th National Policy Conference, Midrand, 30 June—5 July 2017). 161 von Holdt, ‘The political economy of corruption’ 162 Gray and Whitfield, ‘Reframing African industrial policy’. 163 Whitfield et al., The politics of African industrial policy; Hickey, ‘The politics of state capacity’. 164 Behuria et al., ‘Studying political settlements’. 165 See also Beresford, ‘Power, patronage, and gatekeeper politics’; Zalk, ‘Selling off the silver’. 166 von Holdt, ‘The political economy of corruption’. A similar argument, though with some distinctions, is made in Chipkin et al., Shadow. 167 Ibid. 168 Beresford, ‘Power, patronage, and gatekeeper politics’. 169 Nimrod Zalk, Things we lost in the fire: the political economy of post-apartheid restructuring of the South African steel and engineering sectors (University of London, unpublished PhD dissertation, 2017). 170 Ibid, pp. 39–40. See also von Holdt, ‘Nationalism, bureaucracy and the developmental state’. 171 Atul Kohli, State-directed development: Political power and industrialization in the global periphery (Cambridge University Press, Cambridge, 2004), p. 9. 172 Ibid. 173 Whitfield et al., The politics of African industrial policy, Behuria et al., ‘Studying political settlements’. 174 See also von Holdt, ‘The political economy of corruption’. 175 DPE, ‘Roadmap for Eskom in a reformed electricity supply industry’ (DPE, Pretoria, 2019) 176 Antoinette Slabbert, “SA bent over backwards for Eskom’: Nersa heads to court’ City Press, 5 January 2020, <https://city-press.news24.com/Business/sa-bent-over-backwards-for-eskom-nersa-heads-to-court-20200105> (6 January 2020). 177 Eskom, ‘Annual financial statements’ (Eskom, Johannesburg, 2019). 178 DPE, ‘Roadmap’. 179 Ibid, p. 4. © The Author(s) 2020. Published by Oxford University Press on behalf of Royal African Society. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Parastatals and economic transformation in South Africa: The political economy of the Eskom crisis
JF - African Affairs
DO - 10.1093/afraf/adaa013
DA - 2019-09-11
UR - https://www.deepdyve.com/lp/oxford-university-press/parastatals-and-economic-transformation-in-south-africa-the-political-e6ECcIFkQx
DP - DeepDyve
ER -