South Africa was one of the 1990s iconic cases of democratization. Yet starting in the mid-2000s, the country began to experience a disruptive collision between its strong political institutions and massive economic inequality. The collision intensified across the 2010s, resulting in economic stagnation and increasing threats to institutional integrity. Understanding why this collision occurred and worsened over time is relevant not just for other middle-income countries but also many higher-income democracies wrestling with similar tensions between a declining tolerance for high or rising inequality and institutions that seemed strong in the past but find their legitimacy increasingly being questioned.

Ideally, ideas, institutions, and growth all reinforce one another in a virtuous developmental spiral. Ideas offer hope by encouraging cooperation and the pursuit of opportunities for win-win gains. Institutions assure that the bargains underpinning cooperation will be monitored and enforced. Together, ideas and institutions provide credible commitment, which fuels economic growth. However, such a benign scenario does not reckon with the ways in which persistent high inequality, accompanied by unresolved tensions between the distribution of economic and political power, can both put pressure on institutions and quickly change hope into anger. The result can be a cascading set of pressures and an accelerating downward spiral.

For the first fifteen years of democracy, South Africa enjoyed the advantages of both effective institutions and a shared willingness of stakeholders believed in the power of cooperation. This enabled the country to move beyond counterproductive conflict and pursue win-win outcomes. Growth began to accelerate, which created new opportunities for expanding the middle class. Increased fiscal space made it possible to broaden access to public services and to social grants, which reduced absolute poverty.

There were, however, some stark limitations in what was achieved. Gains for the poorest did little to alter their difficult economic and social realities. Less than a quarter of the total population, including essentially all white South Africans, enjoyed a standard of living that was middle class or better. There was ample reason for the majority of South Africans to feel that, notwithstanding the promises of mutual benefit, the deck remained stacked against them. This increased the vulnerability of South Africa’s political settlement.

South Africa’s political settlement was built around four distinct subbargains:

  1. A deal between the overwhelmingly white economic elite and the country’s new political leadership. This included commitments to the rule of law (including protection of private property), and to economic transformation, including through the Black Economic Empowerment (BEE) program.
  2. A deal among the new political elites within the majority political party, the African National Congress (ANC). The ANC is a broad tent encompassing many ideological proclivities; degrees of public-spiritedness; and regional, ethnic, and economic interests. Its implicit promise was that its formal structures, plus the structures of government, would channel this diversity toward a shared national purpose.
  3. A promise of upward mobility. One aspect was a commitment to protect the interests of new (predominantly black) middle-class insiders. Another aspect was a promise that a combination of education, job creation, and an end to racial discrimination would open up readily accessible opportunities for those on the cusp of middle-class status.
  4. A promise to reduce extreme poverty through a post-minority-rule redirection of public resources and services.

All of these subbargains except for the last one, which was pursued at least into the 2010s, were built on shaky foundations. Pressures on institutions were building. Many BEE transactions straddled the boundary between rules-based and more personalized deal-making; who should participate in BEE initiatives became part of the ANC’s inter-elite conflict. Adapting to a transformed political order created new pressures for the public sector. Had South Africa been able to enjoy a combination of visionary leadership and East Asian rates of rapid economic growth for a sustained period, the expansion of opportunity throughout society might have trumped the limitations of the aspirational commitments. In reality, the country only briefly reached an annual rate of 5 percent from 2005 to 2008. In 2009 Jacob Zuma became president, having won a bitterly contested struggle for ANC leadership. Over the course of the subsequent eight years, an accelerating downward spiral of decline seemed to take hold.

Upper echelons of the ANC competed for control over resources and positions of influence, further ratcheting up pressure on institutions. After 2009, state-owned enterprises emerged as an especially lucrative area for powerful and well-connected individuals and factions to engage in systematic looting.

In Zuma’s second term, per-capita income growth entered negative territory. Zero-sum contestation over positions and resources at the national, provincial, and local levels became acute. A rising tide of disillusion across a broad swathe of society further undermined the political settlement. The losers in elite conflicts within the ANC felt aggrieved. Those on the cusp of the formal economy found themselves unable to consolidate their middle-class status; job creation in the formal sector was anemic. Unemployment steadily increased.

Disillusion led to an ideational turn toward anger, fueled by both genuine grievance and political opportunism. In the face of thwarted opportunity, an increasing number of South Africans came to see the privilege enjoyed by the mostly white economic elite—and the tide of apparent corruption that seemed to be the only way that new elites could share in that privilege—as a provocation. In turn, opportunistic ethnopopulist political entrepreneurs sought to use the disillusion to strengthen their position within inter-elite political struggles.

All the elements seemed to be in place for a rapidly accelerating cumulative slide, with weakened economic performance, institutional decay, and anger and ethnopopulism feeding on one another. While the December 2017 election of Cyril Ramaphosa as leader of the ANC and his subsequent accession to the country’s presidency signaled a pause to this slide, three years later the country is not out of the woods. Ramaphosa has not been able to move decisively beyond a promise to stop the rot, and to also offer a renewed positive vision. The country’s deep-seated, underlying challenges remain, exacerbated by the economic consequences of the coronavirus pandemic.

South Africa’s experience suggests four potentially useful lessons for the many countries struggling to maintain a positive social, political, and economic trajectory in the face of high or rising inequality.

  1. The trajectory of change is a knife-edge. There is potential to create virtuous circles of positive interactions among ideas, institutions, and economic growth. But there also is substantial risk that unaddressed distributional imbalances could cause a cumulative downward spiral of decline.
  2. Ideas matter—a hopeful vision of change, when combined with a “good enough” responsiveness to distributional concerns, can be sufficient to launch a positive trajectory.
  3. Both ideas and institutions can be shields against adversity—but only up to a point. Hopeful ideas can evoke positive agency and help mobilize people for collective action. Institutions can function as shock absorbers. However, both ideas and institutions need reinforcement, including by addressing festering imbalances in economic and political power.
  4. Initiating a new cycle of renewal requires a set of ideas and actions that address in a “good enough” way the imbalances resulting from derailment.

What is most important is a credible promise of upward mobility, which would both offer rapid gains for those on the cusp of the middle class and renew a vision of hope and possibility across society as a whole. South Africa’s leaders need to mobilize new coalitions capable of overcoming the vested interests that stymie inclusive change. Can South Africa’s leadership—and the leadership in other countries where similar disillusion has also taken hold—summon the necessary boldness to rise to this challenge?


For champions of democratic governance, the 1990s was a time of exuberant optimism. A third wave of democracy was cresting; between 1988 and 1996, the number of electoral democracies worldwide rose from 67 to 118 countries.1 Three decades later, however, the world has gone into a democratic recession. Between 2006 and 2017, the proportion of countries classified as democracies declined from 58 percent to 51 percent.2 Even within long-established democracies, the tenor of political discourse has become increasingly polarized. Many democracies are finding it difficult to self-correct—that is, to use their experiences to reassert democratic norms and practices in the face of challenges to their systems of governance.

A crucial part of the optimistic outlook of the 1990s was the assumption that democratization and development naturally would go hand in hand in a mutually reinforcing virtuous circle. Yet the experience of the past two decades has called this assumption into question: Many developing-country democracies have confronted both economic difficulties and increasing challenges to their political health.

This report uses the case of South Africa to explore the proposition that a critical weakness for many democracies is the complexity of addressing challenges of inequality and exclusion. These challenges have become pervasive in democracies the world over. Many high- and middle-income democracies are wrestling with a volatile combination of a declining tolerance for high or rising inequality and institutions that may have seemed strong in the past but now find their legitimacy increasingly being questioned. It is no longer feasible to sweep economic and social imbalances under the rug; nevertheless, countries find it difficult to address these concerns.

Brian Levy
Brian Levy teaches at the School of Advanced International Studies, Johns Hopkins University. He was the founding academic director of the Nelson Mandela School of Public Governance at the University of Cape Town.

Over the past quarter-century, the collision between institutions and inequality has been particularly forceful in South Africa. In the 1990s, South Africa was one of the world’s iconic cases of democratization. For more than three centuries, the country had been riven by conquest, war, the ravaging of indigenous social and economic structures by colonial powers and settlers from a distant continent, and oppressive minority rule. Yet out of this disastrous history, almost miraculously, a “rainbow nation” of equal citizenship for South Africans of all ethnicities and colors—not merely the white minority that had long held the reins of power—was born.3 Its date of birth was April 27, 1994, the day South Africa’s voters went to the polls in the country’s first democratic election.

The 1994 election was a triumph for the African National Congress (ANC), a resolute opponent of apartheid, whose leadership had come home after three decades in exile. Two weeks later, Nelson Mandela—seemingly risen from the dead, having been freed in 1990 after twenty-seven years in jail for his antiapartheid activity with the ANC—was sworn into office as South Africa’s first democratically elected president. A global icon and recipient of the 1993 Nobel Peace Prize, Mandela was committed to building a foundation for peace and stability. Racial reconciliation was a centerpiece of his term in office. But as in many countries, a fairy-tale climax has not been the end of the story.

South Africa is a middle-income country (MIC), with a per capita income similar to that of Brazil and Thailand and somewhat below that of Mexico and Turkey (for comparative data, see table 3 later in this paper). Among MICs, it is an outlier in at least three ways that deserve closer examination:

  • Its democracy is underpinned by a well-designed constitution and an unusually strong set of formal checks and balances on its institutions. At the dawn of its democracy, both government effectiveness and the quality of its judicial institutions rated far higher than those of its MIC comparators.
  • Of those countries with good-quality data, South Africa is perhaps the most unequal country in the world—both in the overall distribution of wealth and income and in the ethnic composition of that distribution.
  • Its transition to democracy brought into power a political party (the ANC) with a radically different and vastly more inclusive political base than its white minority predecessor—and with a long-standing commitment to building a more equitable society.

These characteristics make the South African case relevant not only for MICs but also for higher-income countries where strong or at least relatively strong institutions are being threatened by inequality, civic dissatisfaction, and political polarization.

The country’s mid-1990s political settlement seemingly transformed both relationships among elites and the modes of incorporation of nonelites. The settlement included implicit promises to

  • rebalance political power and economic benefits between established and emerging elites;
  • navigate potential rivalries among emerging elites (especially those within the ANC’s broad tent) in a way that sustained a platform of economic and political stability;
  • create opportunities for upward mobility for nonelites, beyond a favored few; and
  • address the country’s massive legacy of extreme poverty with the promise (as per the ANC’s 1994 election slogan) of “a better life for all.”

This settlement was underpinned by a combination of ideas and institutions. The central idea was a shared willingness among stakeholders to believe in the power of cooperation, which would enable South Africa to move beyond counterproductive conflict and pursue win-win outcomes. Effective institutions, both formal and informal, provided commitment mechanisms that helped make promises of cooperation credible. As it happened, however, the daunting challenge of transforming and redressing the country’s difficult political and economic legacy prevented South Africa’s leaders from sustaining a virtuous spiral of sustained hope, a rise in confidence, accelerated growth, and continual gains in inclusion. Instead, failure to deliver on the promise of shared gains contributed to a cascading downward spiral of pressure on institutions, worsening economic performance, and of an ideational shift away from cooperation.

Yet for all of South Africa’s difficulties, there also may be positive lessons to learn from its example. In the early 1990s, South Africans demonstrated their ability to overcome formidable odds and reverse what seemed to be an inexorable downward spiral. The election of Cyril Ramaphosa, first as ANC president in December 2017 and then as president of South Africa in February 2018 following Zuma’s resignation, signaled that a large segment of society (including within the ANC) continues to be committed to building a thriving, inclusive constitutional democracy. Though the first three years of Ramaphosa’s presidency have been difficult, perhaps South Africa can surprise the world again—and others can learn from the experience.

To better understand how South Africa’s experience with inequality and exclusion has affected its democratic transition and development, this paper will establish an analytic framework and link it to the South African experience. It will then examine the evolution of governance and institutions in the post-apartheid era in three parts. The first part reviews the period from approximately 1994 to 2010, when the cooperative vision was largely intact. The second part involves the acceleration of a downward spiral under the administration of President Jacob Zuma. The final part looks at how South Africa was able to pause the downward spiral with Ramaphosa’s December 2017 victory in the contest to succeed Jacob Zuma as leader of the ANC, and what it might take to set in motion a new virtuous circle of hope, inclusion, and opportunity.

Setting the Stage

This section lays out the study’s analytic framework, provides some requisite background detail for South Africa, and highlights the connections between the framework and the South African case.

The Hirschman Cycle

Albert Hirschman’s reflections on interactions between development and inequality in Latin America provide a useful analytical point of departure. As figure 1 illustrates, Hirschman frames the relationship as a cycle with three phases: growth, anger, and reform. (As per the bottom row of table 1, the evolution of South Africa’s democracy tracks closely with these phases.)

Reflecting in the mid-1970s on his experience as a scholar-practitioner in Latin America, Hirschman summarized the previous quarter-century as characterized by a “changing tolerance for inequality.” The earlier part of the period was characterized by optimism; Hirschman’s classic 1963 book was titled Journeys Toward Progress, and his 1971 collection of essays was titled A Bias for Hope. But over time, rising polarization, civic conflict, and military rule spread across the continent. Growth, underpinned by hope, had transmuted into anger. What accounted for the shift?

Hirschman characterized the growth phase—a period when “society’s tolerance for disparities [is] substantial”—by comparing it to being stuck in intermittently moving traffic:

Suppose I run into a serious traffic jam in a two-lane tunnel. After a while the cars in the other lane begin to move. Naturally, my spirits lift considerably. . . . Even though I still sit still, I feel much better off than before because of the expectation that I shall soon be on the move. . . .

As long as the tunnel effect lasts, everybody feels better off, both those who have become richer and those who have not. . . . The tunnel effect operates because advances of others supply information about a more benign external environment; receipt of this information produces gratification; and this gratification overcomes, or at least suspends, envy.4

Acceptance of imbalances cannot, however, persist indefinitely. In Hirschman’s framing, development invariably is accompanied by rising imbalances. In South Africa, the apartheid system meant that inequality already was extraordinarily high at democracy’s dawn; the source of hope was the promise of a more inclusive economy and society. Either way, as Hirschman put it, tolerance for inequality

is like a credit that falls due at a certain date. It is extended in the expectation that eventually the disparities will narrow again. But if the expectation . . . does not occur, there is bound to be trouble and, perhaps, disaster. . . . Nonrealization of the expectation that my turn will soon come will at some point result in my “becoming furious”; that is, in my turning into an enemy of the established order. . . . No particular outward event sets off this dramatic turnaround.5

Potentially, though, the cycle might in time move from anger to a new phase. As seen in figure 1, the angry phase of the tunnel effect may be superseded by a reform phase involving “efforts at catching up on the part of lagging sectors and regions at social reforms to improve the welfare and position of groups that have been neglected and squeezed, and at redistribution of wealth and income in general.”6

Table 1 unbundles Hirschman’s depiction of the overall pattern with respect to three variables (ideas, institutions, and inequality), along with differences across the cycle in the growth orientation of economic policy.


In bringing ideas—specifically, the role of expectations about the future—to center stage, Hirschman was following the earlier lead of John Maynard Keynes:

We have as a rule only the vaguest idea of any but the most direct consequences of our acts. . . . By “uncertain” knowledge, I do not mean merely to distinguish what is known for certain from what is only probable. The sense in which I am using the term is that . . . we simply do not know. Nevertheless, the necessity of action and for decision compels us as practical men to do our best to overlook this awkward fact.7

As per Keynes, uncertainty is thus the source of the Hirschman cycle’s volatility:

Our theory of the future, being based on so flimsy a foundation, is subject to sudden and violent changes. The practice of calmness and immobility, of certainty and security, suddenly breaks down. New fears and hopes will, without warning, take charge of human conduct.8

Recent analyses of political economy have explored the role of ideas in fostering change—ideas that protagonists have about themselves, about others, and about how the world works, as well as ideas and expectations of the future.9 As Dani Rodrik put it:

Any model of political economy in which organized interests do not figure prominently is likely to remain vacuous and incomplete. . . . But a mapping from interests to outcomes depends on many unstated assumptions about the ideas of political agents.10

Democratic South Africa offers a compelling example of the existence—beneath a political discourse preoccupied with interests and policies—of an unstated assumption about a central underlying idea: that a thriving future could be built around cooperation and thus create the possibility of win-win outcomes with shared benefits. The ANCs 1994 electoral slogan—“a better life for all”—and its detailed economic electoral manifesto, the Reconstruction and Development Programme (RDP), were evidence of this thinking. The RDP took as its point of departure a vision that reconstruction and development would be achieved through “the leading and enabling role of the state, a thriving private sector, and active involvement by all sectors of civil society which in combination will lead to sustainable growth.”11

Alan Hirsch
Alan Hirsch is a professor at the Nelson Mandela School of Public Governance at the University of Cape Town (UCT) and was the school’s founding director from 2011 to 2019.

The idea of gains through cooperation is radically different from an understanding of the world framed in terms of dominance and oppression. What made South Africa’s “rainbow” miracle so remarkable was the speed with which one set of national ideas appeared to give way to their complete opposites. From an essentialist, race-based society marked by internalized patterns of dominance and oppression—which framed its central conflict in terms of a struggle between competing nationalities—an inclusive, cooperative social order emerged, seeking to embrace the principles of equal dignity and shared citizenship. It was a turn from a zero-sum to a positive-sum way of understanding the world.

An ambitious program of cooperation confronts the challenge of time inconsistency. Success requires ongoing cooperation, but not all participants share the costs or reap the benefits of that cooperation at the same time. Successful implementation of the RDP called for

  • early upfront fiscal revenues in order to expand the provision of public goods,
  • the hitherto oppressed majority to practice political forbearance in order to provide an attractive business environment,
  • a willingness of private companies to commit to the future and invest, and
  • both political forbearance and private investment to continue over the long run.

Ideally, all of these conditions would lead to mutual gains for all South Africans. However, at any point, one or more stakeholders might choose to withdraw their cooperation. They might have received an initial set of benefits but now confront a period when their burdens might be at the forefront. Alternately, they might have lost patience with the delay in receiving the benefits they felt they were due to receive. Regardless of the situation, those who backed out of the social contract might easily undermine the entire process.


Institutions, as per Douglass North’s classic definition, comprise the “rules of the game . . . humanly devised constraints for governing human interactions.”12 Institutions can be formal or informal; formalization generally is necessary for governing complex interactions. A paradoxical source of strength of South Africa’s constitutional democracy was the strong set of formal checks and balances institutions that the newly democratic country inherited from the apartheid era. For the country’s white minority, these institutions had long provided a robust economic, social, and political governance framework; the transition to democracy extended their umbrella of protections to the population as a whole.

The country’s new constitutional order was built around the protection of political, civil, and social and economic rights. The political order was based on the principle of equal citizenship for all. The civic order provided constitutional protection for the rights of all citizens, as individuals—with the latter consistent with the nonracial ethos of the ANC. The rights established in the new social order included the rights to education, healthcare, environmental protection, and sexual preference. Similarly, the economic order protected property rights: South Africa would continue to embrace market principles.

Table 2 uses four sets of measures taken from the V-Dem and Worldwide Governance Indicators (WGI) datasets to provide a comparative perspective on South Africa’s governance quality as of 1996 (two years after the political transition), relative to other MICs. For three of the measures, the country emerges as a strongly positive outlier, though its score on a fourth is more mixed—albeit for an understandable reason.

  • Government effectiveness: The WGI’s composite measure of government effectiveness captures “perceptions of the quality of public services, the quality of the civil service and the degree of its independence from political pressures, the quality of policy formulation and implementation, and the credibility of the government’s commitment to such policies.” South Africa ranked in the 83rd percentile, well above the MIC comparators in the table and on a par with countries such as Italy and France.13
  • Rule of law: V-Dem’s composite measure of the rule of law measures “the extent to which laws are transparently, independently, predictably, impartially, and equally enforced, and the extent to which the actions of government officials comply with the law.” Again, South Africa scores well above its MIC comparators.
  • Control of corruption: The WGI’s control of corruption score is intended to capture “perceptions of the extent to which public power is exercised for private gain, including both petty and grand forms of corruption, as well as ‘capture’ of the state by elites and private interests.”14 Corruption is an outcome of how institutions function; the extent to which corruption is controlled is a useful proxy for the overall quality of a country’s institutional system. South Africa’s 1996 percentile ranking was similar to that of southern European countries and was well above its MIC comparators.
  • Political party institutionalization: As per V-Dem, the indicator captures “the level and depth of political party organization, links to civil society, cadres of party activists, party supporters within the electorate, coherence of party platforms and ideologies, party-line voting among representatives within the legislature.”15 Here, South Africa’s score as of 1996 was higher than Thailand and moderately above Brazil but well below both Mexico and Turkey. Given both that South Africa’s democracy was new and that from 1960 to 1990 the ANC had been forced to operate as an exile/underground movement, the fact that the ANC’s domestic organizational networks were only moderately well developed as of the mid-1990s is unsurprising.

Though imperfect, South Africa’s strong formal institutions thus seemed to provide an unusually strong foundation for addressing the country’s daunting political, economic, and social challenges. In a benign world, positive interactions between the idea of shared gains through cooperation, strong institutions, reasonable policies, and a broadly supportive global environment might be sufficient to enable a virtuous spiral to take hold, for mutual trust to grow, and for the country to develop into a thriving, inclusive affluent society. However, this benign scenario does not reckon with inequality and its consequences.


A useful point of departure for understanding how inequality affects cooperation is the notion of a “political settlement,” a growing area of focus for development scholarship. Crystallizing a decade’s worth of contributions, Tim Kelsall and colleagues define a political settlement as “an ongoing agreement (or acquiescence) among a society’s most powerful groups over a set of political and economic institutions accepted to generate for them a minimally acceptable level of benefits, and which thereby ends or prevents generalized civil war and/or political and economic disorder.”16 A paradoxical implication of this definition is that high inequality need not undermine a political settlement’s stability. What matters for stability is not inequality itself but rather the alignment between the distribution of power in society and the distribution of the economic benefits that result from the settlement. A settlement can be stable if both political power and economic benefits are concentrated and reproduced within a narrow, dominant elite. This is a reasonable approximation of South Africa during its years of peak apartheid.

South Africa’s 1994 democratic political settlement had to wrestle with the challenge of realigning economic and political power. The challenge had two distinct aspects: new bargains among elites and new modes of incorporating nonelites. To address the former, the settlement provided a platform for cooperation between established and emerging elites. To address the latter, the settlement seemingly committed to massively expanding inclusion, thereby enhancing nonelites’ access to resources and opportunities. Indeed, one of the attractions for apartheid-era elites of the elite bargain was that the ANC’s certain accession to power via a democratic election would confer legitimacy across society for a negotiated constitutional order that promised equal rights to all. Going forward, of course, the challenge was to match the aspirations of cooperation and inclusion with actions to address the inherited distributional imbalances in a coherent and mutually beneficial (non-zero-sum) way. The challenge was significant.

Table 3 contrasts South Africa’s inequality in the early 2000s with that of four other MICs. As the table shows, inequality was far greater in South Africa than in the comparators (including Brazil, which in the early 2000s was the perceived exemplar of inequality among MICs). A higher proportion of purchasing power was concentrated within the top 10 to 15 percent of the distribution, with a correspondingly smaller share for everyone else and with especially stark economic exclusion among the bottom 40 percent.

The data in Table 3 do not capture the full measure of the challenge at the top of the distribution. For one thing, wealth generally is more unequally distributed than income, especially in South Africa. As of 2015, the richest 10 percent of South Africans held 92 percent of all of the country’s wealth.17 At least as challenging for the stability of the 1994 settlement was the ethnic composition of the economic elite: South Africa’s apartheid legacy resulted in the commanding heights of economic power being wholly dominated by the white minority. As of 1994, only one-half of 1 percent of the shares on the Johannesburg Stock Exchange (JSE) were owned by black investors.18

As Table 4 details, between 1995 and 2010 there have been some changes in the ethnic composition of the most affluent 10 percent of South Africa’s population. Even so, as of 2010, white South Africans (who comprised only 9 percent of the population, down from 13 percent in 1995), accounted for 56 percent of the top 10 percent. Six out of ten white South Africans were in the top 10 percent of earners, and all but a handful were in the top quartile. Given South Africa’s history, these patterns are ethically intolerable, politically unsustainable, and (as will be explored further below) corrosive to institutions.

Now consider the middle of the distribution. As the data in Table 3 suggest, South Africa confronted a stark cliff—a missing ladder of opportunity in the middle of the distribution. South Africans were (and continue to be) either affluent or poor, with few in between. In Brazil, Mexico, Thailand, and Turkey, income declines relatively slowly within the top part of the distribution—in the early 2000s, those in the 35th percentile earned about half of those in the 15th percentile. By contrast, in South Africa the income of those in the 35th percentile of the distribution was less than 30 percent of that of the 15th percentile.19

A comprehensive assessment of the reasons for the origins and persistence of this income cliff is beyond the scope of this work.20 However, South Africa’s problems have much to do with the structural legacies of the apartheid era. The country inherited the rigidities of a political system that supported an economy geared to generate high incomes for the white minority by using legally oppressed unskilled labor in mining and agriculture, while limiting upward mobility for black South Africans. The business insiders whose investments and skills were concentrated in capital-intensive sectors of the economy—and formal employees (by the mid-1990s, predominantly black) who enjoyed relatively high wages as a “labour aristocracy” of insiders—reproduced the inequalities of the past, with little economic incentive to change. Moreover, the inner workings of the ANC alliance resulted in a pattern of decisionmaking that disproportionately empowered unionized blue- and white-collar workers at the expense of those of its supporters who were outside the formal economy.

Vinothan Naidoo
Vinothan Naidoo teaches public policy and public administration at the University of Cape Town.

Whatever the origins of the cliff, its continued persistence made it difficult for the ANC government to deliver on its promise of upward mobility. It faced an especially difficult tension between creating new opportunities for those outside the formal economy and satisfying the concerns of vulnerable black middle-class insiders, who were an influential block within the ANC. One notable institutional consequence of the pressure to expand the black South African middle class was that in the absence of a thriving, market-driven ladder of opportunity, public sector employment became an important pathway to upward mobility. Insofar as this shift has been combined with a rising propensity to patronage and clientelism, the public sector has had to contend with growing pressures on the fiscus and on service quality (see “Transformation of the Public Sector” later in the paper­).

Virtuous and Vicious Spirals

Table 5 lists some of the ways in which continuing imbalances between economic and political power might undermine the stability of the broader political settlement. One mechanism, highlighted by Hirschman’s “tunnel effect,” is ideational—a turn from hope to anger, evoked by a combination of disillusion and ethnopopulist political entrepreneurship.21 A second set of mechanisms works by corroding institutions. Pressure on institutions can come from a lack of robustness in the elite bargain and thus ongoing contestation among elites; it also can come from ongoing challenges in incorporating nonelites.

Figure 2 illustrates how interactions among ideas, institutions, inequality, and growth can feed on one another for good or ill. In the benign scenario on the left-hand side of the figure, ideas, institutions, and growth all reinforce a virtuous spiral. Ideas offer hope, encouraging cooperation. Institutions provide credibility that the bargains underpinning cooperation will be monitored and enforced. Together, ideas and institutions provide credible commitment, fueling economic growth. However, as the right-hand panel highlights, this benign scenario does not reckon with the ways in which persistent unresolved tensions between the distribution of economic and political power can threaten both elite bargains and the credibility of promises to incorporate nonelites. As the solid arrows in the figure suggest, persistent inequality can both put pressure on institutions and, as per Hirschman’s tunnel effect, catalyze an ideational turn from hope to anger. The cumulative consequence, as per the dotted lines in the figure, can be a cascading set of knock-on pressures, a slowdown in growth, and an accelerating downward spiral.

1994–2007: Season of Hope—A Better Life for All

The first dozen or so years of South Africa’s democracy was a season of hope (as per the title of Alan Hirsch’s book on economic reform under the country’s first two presidents).22 In this hopeful phase of the Hirschman cycle, the country embraced the ANC’s electoral slogan, “a better life for all,” and its vision of realizing mutual gains through cooperation. Realizing this vision was no easy task. As Hirsch (then South Africa’s chief director of economic policy within the office of the presidency) summarized in his book, the crucial challenge was to combine a commitment to growth with a commitment to sharing the benefits of that growth:

Was there any point in redistributing a shrinking patrimony, with everyone getting poorer? And, yet, would not fueling economic growth simply put more wealth in the hands of those who already had it? The big economic question faced by the ANC was: What would be the ideal relationship between growth and redistribution in South Africa? Or, more precisely how could it set South Africa on the path of economic growth and at the same time ensure fair, just and politically necessary redistribution outcomes? Put yet another way: was there a way in which growth and redistribution in South Africa could complement each other?23

The effort to combine growth and redistribution required a collective, society-wide commitment—one that cut profoundly against the grain of South Africa’s historical legacy. Centuries of exploitation and oppression had left large residues of suspicion and distrust, along with deeply rooted dysfunctional habits of unequal communication across racial lines. Inevitably, the process of effecting change was difficult, with many compromises and not a little conflict. Gradually, gains began to be evident; economic growth accelerated and extreme poverty declined. However, achieving gains in growth and poverty reduction turned out to be the relatively easy parts of the agenda of building a thriving inclusive society. Addressing other dimensions of exclusion and inequality turned out to be more difficult—and the lack of resolution in these other dimensions placed new pressure on institutions.

The pressures on institutions were particularly substantial in two aspects of the ANC’s agenda: initiatives to transform the elite through proactive efforts such as the Black Economic Empowerment (BEE) program, and initiatives to use the public sector as part of a broader agenda of transformation. As will become evident, even though some gains were achieved, these initiatives came with internal complications that would set the stage for accelerating difficulties in the 2010s. Before examining the complications, it is worth reviewing South Africa’s achievements in reducing poverty in the post-apartheid era.

Poverty-Reducing Growth—Some Achievements

Table 6 summarizes South Africa’s real economic growth rates since the dawn of democracy. By the end of the apartheid era, the economy had stopped growing. In the first dozen years of democracy, as confidence in the new system took hold, growth began to accelerate. Indeed, signaling the private sector’s rising confidence, private investment gradually rose as a share of GDP from an average of about 14 percent in the 1990s to 17 percent in 2007. Growth provided the fiscal means for addressing absolute poverty; the growing economy also offered new opportunities for expanding the middle class and for upward mobility via the private sector. In the absence of growth, however, everything else would become more difficult—as South Africa would discover in time.

Table 7 summarizes the gains in reducing absolute poverty in South Africa. In the early years of democracy, the extension of services was a major focus. Between 1994 and 2003, as a central part of the RDP implementation, more than 1.5 million houses were built for the poor. Access to potable water grew from 60 percent to 80 percent of households between 1998 and 2003, and access to electricity grew from 50 percent to 70 percent of households between 1994 and 2000. Immunization coverage also grew.

Beginning in the latter 1990s, as government finances strengthened, South Africa greatly expanded its social grants. Pension payments (which during apartheid years had been generous for white South Africans) were universalized and equalized. A Child Support Grant, the main form of support for indigent families, was introduced in 1998 and had reached 3 million children by 2003. By 2010, after the maximum eligibility age was raised several times, more than 10 million of the poorest children in a country with a total population of less than 50 million at that time were receiving child support grants. Social security transfers rose to 3.5 percent of GDP by 2009, representing a concerted attempt to ameliorate severe poverty.

For all of the above gains, there were some stark limitations on the progress experienced by those at the bottom of South Africa’s distributional pyramid. The poorest four deciles are largely unemployed or underemployed and are found mostly in rural areas (designated as “reserves” or “homelands” in the apartheid era) and informal settlements around towns and cities. Many of these informal settlements are poorly integrated into the social life of the towns and cities. Social grants are the principal source of livelihood. Gains for the poorest 40 percent were achieved without much alteration of their difficult underlying realities.

Elite Transformation via Black Economic Empowerment

South Africa’s transition to democracy brought a far-reaching shift of political power from a white minority to the majority of its citizens. However, the economy continued to be structured along lines built by apartheid: white economic elites dominated ownership and control of the private sector. Newly empowered political elites were unlikely to be satisfied with a shift in formal political authority if there was next to no change in economic power. On both moral and pragmatic grounds, the status quo of South Africa’s economic power was untenable.

BEE efforts to increase the proportion of black South Africans in the country’s economic elite have been part of South Africa’s policy agenda since the end of apartheid. Initial efforts were cautious and wary, but momentum soon accelerated. A strong platform of formal institutions capable of monitoring and enforcing carefully negotiated (and codified) agreements offered a seemingly straightforward path: codify a carefully constructed set of rules to govern the process of change. At the outset, the ANC indeed mostly proceeded in this rule-governed way. But this approach, in turn, posed a further dilemma: the “rules-deals” conundrum.

Musa Nxele
Musa Nxele teaches political economy of development at the Nelson Mandela School of Public Governance at the University of Cape Town.

As mentioned earlier, the combination of a well-constructed set of formal rules and credible enforcement mechanisms were a key underpinning for cooperation in South Africa. However, formalism alone could not be a sufficient condition for success: rules cannot be written to cover every contingency.24 Both private and public spheres require tacit understandings, or norms, that facilitate cooperation.25 A 10-year retrospective, issued by the South African presidency in 2003, pointed out the issue:

One of South Africa’s characteristics . . . is the fact that the political leadership in government is quite distinct in terms of race, culture, background and lifestyle from the economic leadership. This has had some benefit in that the political leadership is not beholden to an established “Old Boys Club.” The disadvantage is that the kind of trust needed between the economic and political “elites” for high levels of investment was missing [in] post-1994 South Africa.26

As Alan Hirsch and Brian Levy examined in depth, South Africa’s so-called bargain between established economic elites and new political elites lacked these tacit understandings.27 Beyond agreement on the formal rules, any cooperation had high transaction costs.

Operating within a tightly circumscribed set of formal constraints is an especially unnatural act within the political realm, where deal-making and alliance-building are at the heart of any activity. In a rapidly changing polity such as South Africa’s, in which elite transformation (including internal contestation among aspirant elites as to who would be the new “insiders”) was central to the agenda, the strains became apparent before long. As a recent wave of scholarship has underscored, deals and decisionmaking within personalized networks are not necessarily antithetical to development. Informal commitment mechanisms can provide robust credibility. Loyalty is key to building political machines of all types.28 There is, however, a gray area between the developmental and the predatory. Likewise, there is an associated risk that deal-making processes can corrode a country’s sense of public purpose and undermine the norms and formal institutions that support development structures. Initially, South Africa tried to avoid this slippery slope, but as the process unfolded the hazards loomed larger. Multiple distinct waves of BEE were evident.

A first wave was shaped by a combination of proactivity by established economic elites and caution from newly empowered politicians and policymakers. In a common scenario, established elites would transfer (usually minority) shareholdings in large companies to a consortium of black investors, paid for with debt that would be wound down as dividends flowed in. (The voting power of these shares often was compromised.) An early example of such proactive initiatives was the privatization of the government-owned sorghum beer industry in 1990 and its transfer to the hands of black entrepreneurs, forming the National Sorghum Breweries Ltd.29 Soon after, in 1993, the Afrikaner-dominated conglomerate SANLAM sold 16 percent of the Metropolitan Life insurance company to a black empowerment consortium, New African Investments Limited (NAIL). NAIL’s founder, Nthato Motlana, was a Soweto-based medical practitioner who built a hospital from his practice and had been a brave political leader in the 1970s. NAIL subsequently purchased, at a discount, the industrial conglomerate Johnnic from the Anglo American Corporation. In 1998, before things subsequently turned sour, NAIL had a market capitalization on the Johannesburg Stock Exchange in excess of $1 billion.30 Some initiatives involved creating entirely new businesses, as in the case of the mobile telecommunications company MTN, founded in 1993 by private Afrikaner and Anglophone business interests, the British company C&W, the outgoing Nationalist government, and the incoming ANC. By the early 2000s, MTN’s ownership structure had stabilized; it had emerged as the highest-profile BEE-controlled firm in South Africa. Cyril Ramaphosa would be chair of the MTN board.31 Box 1 presents a detailed case study: the emergence of Patrice Motsepe, the brother-in-law of Cyril Ramaphosa, as a mining magnate.

Box 1: The Rise of Patrice Motsepe

Patrice Motsepe, whose current net worth is approximately $2.1 billion, is a successful BEE entrepreneur who, contrary to the politician-turned-businessman stereotype, built his track record within the private sector. After earning a law degree from the University of the Witwatersrand in 1988, he commenced his career as a lawyer at Bowman Gilfillan, becoming the first black South African partner in 1994. Having specialized in mining and business law, he took advantage of the changing political and business climate in 1994 by founding Future Mining, which provided services to mining companies.32 As gold prices plummeted in 1997, Motsepe secured his first opportunity in gold mining by buying marginal, low-performing mines being shed by AngloGold.33 He was unable to secure a bank loan of approximately $8 million to back his purchases, and so he entered into an agreement with AngloGold to repay the price through the company’s profits—which he did within three years through restructuring and skillful mine management.34

In 2001, Motsepe ventured into platinum mining by winning a bid to be a 50-50 partner with Anglo Platinum to develop the Modikwa platinum project, a deal worth $164 million.35 These early successes placed Motsepe on a firm footing to go public, listing his consolidated assets as ARMGold on the JSE in 2002, with market capitalization of $492 million.36 During the same year, the Menell family’s Avmin mining company, which had platinum, nickel, and gold assets, was in financial difficulty. Through a complex BEE transaction, Motsepe sold ARMGold and absorbed Avmin into a new company, African Rainbow Minerals (ARM), launched in 2004.37 ARM became Motsepe’s vehicle to expand his assets in mining through a number of BEE deals, reaching a market value of $8 billion in 2008.38 By 2011, almost 80 percent of ARM’s revenue was drawn from iron ore, manganese, and platinum.39 In 2014, Motsepe used the proceeds from another long-term BEE deal to found African Rainbow Capital (ARC), which listed on the JSE in 2017 with a diversified portfolio valued at $350 million. Through ARC, Motsepe intends to build black capital along the lines of SANLAM, historically the capital vehicle of the Afrikaner business class.40

For newly empowered politicians and policymakers, the challenges raised by the need for elite transformation were formidable. They had to contend with an incumbent (white) business class, which recognized the necessity of diversifying the ethnic makeup of ownership and control in the South African economy but sought to do so on its own terms. Equally challenging was the task of navigating among disparate ideological factions within the ANC. As so often happens with transfers of power, South Africa’s rising waves of political leaders had strong incentives (paralleling those of an earlier generation of white Afrikaner politicians) to use their newfound political power and control over state resources to open up business opportunities for their allies.

The initial response of policymakers and political leaders was to try and contain the struggle over ownership within a broader discourse around economic transformation. They sought out ways to address the seemingly contradictory interests of the various protagonists without undercutting the momentum of efforts to reincorporate South Africa into the global economy, as well as more broadly to foster inclusive economic growth and development. Efforts generally focused on new firm development, affirmative action in the workplace, and measures to support the enhanced education of black South Africans who had the potential to be professionals, managers, and business leaders.41 In ANC economic policy documents before the 50th National Conference in Mafikeng in December 1997, the ANC called for the following:

National Empowerment Policy: The ANC should clearly articulate a National Empowerment Policy that will focus on those who have been historically disadvantaged and particularly black people, women, youth and the disabled and rural communities. The empowerment process must constitute part of a more radical and profound change in social relations. Changing ownership and workplace relations are part of this wider process of empowerment.

Within the National Empowerment Framework government should establish a National Empowerment Fund which must lead to the stimulation of saving, shift people from the informal to the formal sector, and from predominantly retail to more manufacturing SMMEs [small, medium, and micro enterprises].

The ANC government should ensure the implementation of a vigorous affirmative procurement policy which will ensure that government and parastatals facilitate awarding of tenders to our people through approved mechanisms.42

Subsequently, though, it increasingly became apparent that lower-key approaches to transformation were doing little to transform the ethnic hue of ownership. Thus, subsequent to 1997 the BEE process accelerated—spurred in part by a 1998 economic contraction and stock market decline that left many black South African beneficiaries of initial rounds of deals without the cash flow needed to meet their BEE-related financial obligations. During this period, black South African ownership on the JSE declined from 8 percent to less than 4 percent.43 BEE became increasingly central to the objectives of the ANC, and its protagonists and beneficiaries became increasingly influential within the party.

A central pillar of this accelerated response was an effort to establish a system of rules within which to manage BEE expectations and pressures.44 This effort had a number of notable milestones. In 1998, at the urging of the Black Management Forum, the government established a BEE Commission with the task (as then president Thabo Mbeki put it) of promoting “the formation of a black bourgeoisie which will itself be committed and contribute to black economic empowerment.”45 The commission produced a final report in 2003 and released an official strategy document, which led to the promulgation in 2004 of the Broad-based Black Economic Empowerment (BB-BEE) Act. Also, in 2002, the government passed the Mineral and Petroleum Resources Development Act (MPRDA), which required owners of “old-order” mineral rights to convert to “new-order” rights subject to conditions of meeting BEE imperatives (see box 2).

Box 2: The Struggle Over Mining Rights

In the mining sector, the efforts to introduce BEE ratcheted up uncertainty. Between 2001 and 2008, the longest sustained commodity boom in recent history, South Africa’s mining industry contracted at a rate of 1 percent per year, even though its closest competitors grew an average of 5 percent per year—notwithstanding a 2010 valuation by Citgroup of South Africa’s proven mineral resource wealth at $2.5 trillion, the largest in the world.46 The MPRDA-mandated process of converting old-order to new-order mineral rights proved to be rife with ambiguity and scope for ministerial discretion as to what entailed a BEE-compliant conversion, resulting in numerous instances of conflict (resolved juridically) involving incumbent and BEE companies, and discretionary government action. An attempted revision of the legislative framework via a 2013 Mineral and Petroleum Resources Amendment bill added to the ambiguity.

An especially thorny sticking point concerned a specific question: Would a new round of BEE be required if a mining company had reached a required minimum threshold of 26 percent black-owned equity but some of the black South African shareholders subsequently chose to sell their shares? Ongoing disputation and revisions continued. In 2018, a compromise seemingly was reached to ramp up BEE to 30 percent ownership within five years of the charter‚ including a contentious 10 percent free-carry for communities and employees as part of ownership structures.47 In December 2019, draft amendments to the regulations were published. The mining establishment saw the draft as an improvement but expressed reservations regarding whether some obligations were unduly burdensome; as of the time of writing, the regulations had not been finalized.48 Unless the question of mineral rent rights is resolved decisively, South Africa may miss the next minerals boom as well.

As per its name, the 2004 BB-BEE Act went beyond a narrow preoccupation with ownership. Its implementing framework laid out a balanced scorecard that set targets for ownership, management control, skills development and support for subcontractors and other small and medium enterprises, and socioeconomic development more broadly. The expectation was that each economic sector would detail a quasivoluntary sectoral charter aligned with the framework laid out in the legislation. The only enforcement mechanism was that to have access to public procurement opportunities, companies would need to be BEE compliant.

The construction sector illustrates the results of these efforts. Immediately following the passage of the BB-BEE Act in 2004, the sector began negotiating its own sector-specific charter, which was finalized a few years later.49 Four of the largest firms have complied with some of the provisions of the charter in exemplary fashion.50 As of 2014, the black South African ownership percentage in both Murray & Roberts and Wilson Bayly Holmes Ovcon was above 45 percent. (The other two, Group Five and Basil Read, also were above the target of 30 percent.) All four firms had met the charter’s BB-BEE targets for providing capacity-building support to their subcontractors and for enterprise development.

However, the construction sector’s BB-BEE program was not an unambiguous success. For one thing, as of 2014, all four firms had fallen well short of targets (which ranged from 30 to 60 percent) for black South Africans’ inclusion in senior, middle, and junior management positions. For another, the major construction companies repositioned themselves as global companies, with only a minority of their business in South Africa. There also was domestic scandal. South Africa’s 2010 hosting of the World Cup soccer championship was a bonanza for the construction companies who built the stadiums and related infrastructure. Subsequently, following an in-depth investigation by the independent Competition Commission, all of the companies confessed to extensive World Cup construction-related price-fixing.51 Finally, in part as a consequence of the World Cup scandals, and in part as a consequence of broader economic stagnation, the industry went into the doldrums, with a major decline in share prices.

For all of its limitations, the combination of voluntarism and rule-focused BEE was consistent with the continuation of strong formal institutions. But there was also a shadow side to the new government’s use of economic power. Top political leadership wanted trusted allies to be the beneficiaries of ongoing efforts to change the racial hue of business. The ANC also needed support to finance its activities. Inevitably, some influential insiders went further, seeing an opportunity to use their political influence as a platform for private economic enrichment and empire building.

The problems began almost as soon as the new government assumed office. When the ANC first came to power, it found that the South African armed forces had set up an arms-purchasing deal, with suppliers identified. It quickly torpedoed this initiative. After a defense review was completed, requests for proposals were issued for a range of conventional naval and air force equipment. The exercise was muddied by several kickbacks and side-deals under the cover of helping the ANC and supporting empowerment, including one that implicated then deputy president Jacob Zuma.52 This was the first major set of transactions tainted by corruption.

South Africa’s state-owned enterprises (SOEs) also provided opportunities for murky deals. In the early 2000s, the newly formed SOE PetroSA and a little-known BEE consultancy company were involved in a questionable deal related to the United Nations Oil-for-Food Programme in Iraq. This deal also appeared to be connected to the ANC’s efforts to raise finances for the 2004 election campaign.53 Similarly, in 2004, a multi-billion-dollar procurement contract for the boilers for the massive coal-fired Medupi electricity generation plant was awarded through an obscure process by the state-owned electricity company, Eskom. The contract went to a joint venture between Hitachi Power Systems and Chancellor House—the investment arm of the ANC.54

The telecommunications industry also saw some questionable transactions. In 2004, a BEE “elephant consortium” group headed by the former director general of the Ministry of Telecommunications and dominated by close allies of Mbeki acquired a controlling share of South Africa’s leading fixed line telecommunications operator, Telkom, with subsidized support from the state-owned Public Investment Corporation. The equity was purchased from an international consortium, which had entered South Africa as part of a 1997 partial privatization deal but exited just seven years later.55 Between 2007 and 2009, a series of deals by Vodacom—one of South Africa’s two largest mobile telecommunications operators (at the time a joint venture between Telkom and UK-based Vodaphone, controlled by the latter)—resulted in the following:

  • A massive windfall to the elephant consortium (as part owner of Telkom, which sold some of its equity);56
  • Establishment of a new BEE consortium to partner with Vodacom—to be headed by Bulelani Ngcuka, a close ally of Mbeki and, as the director of public prosecutions, a lead actor in efforts to prosecute Zuma for corruption;57 and
  • After Thabo Mbeki lost an ANC leadership struggle to Zuma, the setting aside of the Ngcuka-headed consortium in favor of a more politically neutral BEE arrangement with Royal Bafokeng Holdings.58

In sum, the inclusion of black South Africans into the capitalist class began through an ad hoc set of private initiatives in the early 1990s and transformed into a rule-based set of programs during the Mbeki presidency in the early 2000s. Notwithstanding the ambiguous character of BEE and some signs of its corrosive effect on formal institutions, the balance between rules and deals was good enough to provide a reasonably robust platform for the economy. Though it might be argued that the seeds for a more anarchic series of deals and maneuvers to enrich black South Africans (and others) were planted during the Mandela and Mbeki presidencies, prior to the assumption of power by Zuma, legitimate order remained dominant. The prospects seemed positive for accelerating growth and continuing declines in extreme poverty.

Transformation of the Public Sector

A long-standing idealized Weberian vision of public sector governance draws a sharp boundary between, on the one hand, the acknowledged necessity for political control over the apparatus of the public sector and, on the other, a professional bureaucracy that has the technocratic skills to solve public problems as prioritized by political leadership. Reality almost everywhere falls short of this idealized vision. In the case of South Africa, elements of Weberianism had been visible in public sector governance since early in the twentieth century. However, the objectives that this bureaucracy was charged with pursuing were set by the globally condemned apartheid political leadership. Since the mid-1920s, employment in the public sector had been used as a political tool of targeted employment creation, acting as something of an employer of last resort to address the “poor white” problem.59 The policing arm of the public sector was the mechanism for ruthless enforcement of apartheid controls. Puppet quasi-independent bantustans—complete with explicitly patronage-oriented bureaucracies—became a key mechanism to deny black South Africans their citizenship. The political-bureaucratic interface was in practice governed through the influence of a secretive parallel organization, the Broederbond—good standing in which was necessary for both political careers and careers within the public bureaucracy.

The coming of democracy presented public governance with both opportunities and challenges. The challenges included asserting political control over a public service that had been instrumental to implementing apartheid laws; deracializing the bureaucracy; reincorporating the quasi-independent bantustans into a new, nonracial semifederal system, with responsibilities aligned across national, provincial, and local levels; and providing equitable public services, which to date had disproportionately benefited the white minority population. A representative bureaucracy was a key element in transforming the post-apartheid public service. It was rooted in the ideal that a bureaucracy that represented the demographic diversity of the country could help rebuild citizens’ trust in government. In practical terms, this meant increasing the employment of black South Africans in the hierarchy of government departments. This contributed to the broader goal of accelerating the emergence of a black South African middle class.

With these historical realities (rather than a benchmark of perfection) as a backdrop, this daunting transition was, for the most part, quite well managed. Contrary to a familiar domestic catastrophizing narrative of bureaucratic collapse, post-apartheid South Africa maintained relatively tight control over the economically sensitive aggregates, with continuing (though certainly imperfect) focus on the expansion of public service provision to the population as a whole.

To be sure, the transition of political control brought new opportunities for politicizing public sector employment. Over time, the impact this situation had on bureaucratic performance was one of decay, accelerating during the Zuma era. Even so, as of 2020, South Africa’s bureaucracy remains largely functional—if strikingly inefficient in terms of its delivery potential and outcomes—and is on the better-performing end of the spectrum of bureaucratic performance among MICs.

Employment and economic aggregates. A useful place to begin to understand what has happened to public sector governance in the quarter-century of democracy in South Africa is with the big picture: the role of the public sector in the expansion of an emerging black South African middle class. Tables 8 and 9 provide estimates of the magnitudes. Table 8 uses 1995 mean per capita income as a threshold minimum for being characterized as middle class.60 (Per capita income as of 1995 was roughly $5,500; 23 percent of the population earned above that mean.61) As the table shows, in 1995, approximately 3.9 million black South Africans earned above that threshold; by 2010, the number had risen to 7.5 million—an expansion of 3.6 million.62

Against that backdrop, consider table 9. The table consolidates data on trends in public sector employment, disaggregating by level of government, race, and job classification; the job types highlighted in the table comprise those that might be characterized as middle class (and elite) jobs. Between 1995 and 2005 (and continuing between 2005 and 2018), there was a large shift in the racial composition of management and other skilled positions within the bureaucracy. Already in 1995, and factoring in the integration of public servants from former bantustan bureaucracies, nearly 360,000 black South Africans were employed within the public sector at levels that supported a middle-class (or better) lifestyle. Between 1995 and 2005, middle-class employment of black South Africans in national and provincial government increased by nearly 115,000 positions. This increase enabled approximately 344,000 additional black South Africans (assuming an average household size, within the top third of the distribution, of three people) to enjoy a middle-class status. These numbers are not small. Even so, increases in public employment accounted for only a modest proportion of the total expansion of South Africa’s black middle class.

Consistent with the data in table 9 and the pattern evident in figure 3 below, control of the economic aggregates associated with public employment—numbers employed and wage levels—largely has been robust. Strikingly, between 1995 and 2005, there was a decline in aggregate (national and provincial) public employment, resulting from a transitional period of public sector rationalization. Subsequently, there was some increase, taking employment back to and beyond 1995 levels. However, as per table 9, throughout the 1995–2018 period, employment in the single largest middle-class employment category—highly skilled production, where jobs are predominantly at the provincial level and mostly in the education and health sectors—actually declined. There is no sign of any effort to cater to middle-class aspirations by expanding the number of posts in job-rich, highly skilled areas of frontline public service provision.

The number of middle and senior managers has experienced significant growth.63 The National Treasury put this into perspective by reporting that in 2006/2007, 31 percent of public servants were employed in salary levels 1–4, with 10 percent in levels 9–16 (middle and senior management).64 By 2017, the share of public servants employed in levels 1–4 had dropped to 19 percent, whereas the share of officials employed in levels 9–16 had increased to 21 percent.65 The most striking outlier in the expansion at management levels has been in the number of “highly skilled supervision” (middle managers) positions at the provincial level. The bulk of the increase occurred over a three-year period between 2006 and 2009, crowding out growth in the number of skilled production staff. (Over this three-year period, the number of positions in this category nearly doubled, reaching just below 131,000.66) This increase, which has constrained service delivery performance at subnational levels, may have been exacerbated by political interference in senior staffing processes.

Local government appears to have experienced a more rapid increase in employment than other levels of government. Total employment increased by about 30 percent between 2005 and 2018, with managers accounting for the largest share of growth.

There have been increases in the aggregate amount the public sector spent on wages. Occupation-specific wage settlements have enabled employees in the provincially concentrated education and health sectors to enjoy increases in real earning that to some extent have outpaced both inflation and expansion in public revenues, with relative increases greatest between 2010 and 2017. The result was that compensation spending as a share of total provincial budgets had risen to nearly 70 percent in 2017/2018. However, as of 2019, labor costs had not spiraled out of control. Between 1995 and 2019, spending on personnel ranged between 32 and 40 percent of total government spending and remained between 32 and 36 percent from 2002/2003 onward, creeping up to 38 percent in 2019.67 Total government consumption, including the wage bill, amounted to approximately the same percentage of GDP (20 percent) in 2006 as prevailed in 1994. In 2010, in the wake of the global financial crisis, it rose modestly to 21.5 percent of GDP and subsequently restabilized at marginally below that level.68

In its 2020 budget, the South African government expressed an intention to aggressively address the public sector wage bill by moderating cost-of-living increases, pay progression, and other benefits in order to alleviate the country’s fiscal burden and crowding-out effect on other government spending.69 The upward trend in public sector earnings had become an acute problem politically, given that it had not been driven by a dramatic expansion in aggregate personnel numbers (which, as figure 3 showed, peaked in 2012 and have leveled off).70 Rather, the increases seem to have benefited public servants in the more highly paid supervisory/managerial ranks, whose numbers have grown disproportionately compared to highly skilled frontline public sector workers.

Trends in the quality of public management. Two pathways embraced by the incoming ANC government had far-reaching consequences for how the public sector was managed. One of these was the government’s experimentation with New Public Management, a concept popular worldwide in the 1980s and 1990s. This is not the place for a comprehensive review of the impact of New Public Management on public governance in South Africa.71 What is relevant for present purposes is that its introduction weakened centralized control over the administrative processes of government as a result of the extensive contractualization of service delivery through public procurement and supply chain arrangements. The use of limited-term contracts was extended to the employment of top public servants, in order to engineer the entry of politically pliable officials as well as to discourage competent career-oriented officials from taking a principled stand against various forms of abuse of authority for fear of falling out with their political principals.72

A second and related pathway embraced by the ANC government in 1997 was cadre deployment. The image of the party cadre has a long history in the ANC, initially conceived as part of a strategy to develop the political, ideological, and operational discipline of party members (then in exile) in preparation for political leadership. The deployment of party cadres as an ANC policy imperative was resurrected post-1994 as a way of actualizing that goal and was articulated as part of the wider public service transformation agenda. It was closely aligned with the goal of representation and was justified on the basis of the need to reassert political control over a potentially recalcitrant bureaucracy, which, only a few years earlier, had constituted a repressive state apparatus.

Over the past twenty-five years, the ANC has struggled to reconcile the contradictory pressures associated with its efforts to internally manage the deployment process. This has seen the practice degenerate into factionalized contestation, rent-seeking, and patronage-driven appointments across a wide range of institutional sites in the public service and SOEs. In this regard, cadre deployment was perhaps less of a shift than it might superficially appear. South Africa has no history of using an examination system as a nonpolitical means for merit-based recruitment; good standing within the Broederbond was the Afrikaner Nationalist government’s own form of cadre deployment. After 1994, senior positions generally have been subject to more rigorous competency criteria and tertiary qualifications. Nonetheless, the explicit embrace of cadre deployment has added to the difficulty of instilling professional norms and capacity in the public service. South Africa’s National Development Plan, produced in 2012 during the early period of Zuma’s presidency, recommended reforms to the appointment of senior bureaucrats by instituting more independent checks in the selection process to reduce the scope for political manipulation. These policy reforms failed to materialize in the ensuing years of the Zuma presidency, or subsequently.

What impact did the combination of loosened process controls and cadre deployment have on the quality of public services? Table 10 reports the trends in South Africa’s performance for two widely used indicators: the WGI measures of governance effectiveness and control of corruption between 1996 and 2017. As the table shows, over the course of two decades both measures declined steadily but modestly. Surprisingly, there is no evidence of an accelerating decline from the Mbeki to the Zuma eras. For both measures, South Africa’s score as of 2017 was at least as high as the scores of the MIC comparator countries featured in table 2.73

To better understand what lies behind the WGI’s patterns for South Africa, including why they seem so out of sync with South African perceptions, the following paragraphs (and Appendix A) report on changes over time in indicators on the quality of public management processes and public service provision. Box 3 highlights a set of five inflection points in the character of public sector management between the Mbeki and Zuma eras.

Box 3: From Mbeki to Zuma: The Changing Character of Public Management, 2008–2018

In South Africa’s discourse, it is common to catastrophize public sector performance and to ascribe blame for the declining quality of the public sector to Zuma’s mode of governance. The Zuma presidency was indeed associated with predatory actions, especially with regard to the governance of SOEs. However, an analysis of trends in service quality over the Zuma and Mbeki eras does not reveal a dramatically accelerating decline. Rather, what appears to have happened is a gradual erosion of core public management practices; five of these are highlighted here.

  1. Acceleration of growth in the managerial class (middle and senior) of public servants to the detriment of highly skilled frontline staff, especially in provincial and local governments. This shift constrained subnational service delivery, increased pressure on the wage bill, and failed to instill desired levels of professionalism in the public service at the cost of political interference. In a briefing to parliament, the Public Service Commission recounted various problems associated with management-level officials that continued to frustrate efforts to professionalize the public service. These included excessive turnover of officials (also linked to interference by political heads below the senior management level) without demonstrating requisite job proficiency, a lack of preparedness among officials entering middle management, and insufficient checks and incentives to prevent manipulation of the performance management system.74 Advocacy groups have highlighted how politicization can affect public sector managers: “Regardless of the system of delegations that pertains . . . political appointment of a politically-allied head of department . . . enables political preferences to extend to their subordinate, and so on as politicization cascades down the hierarchy.”75
  2. Deteriorating trends in public finance management, marked by an almost endemic continuity in contracting/supply chain violations and increasing irregular expenditure, as expressed in this statement by South Africa’s Auditor-General’s office: “[O]ver the past four years since 2014–15, there has been a similar pattern with more audit outcomes regressing than improving.” Auditor General Thembekile Kimi Makwetu said that for years his office had reported weaknesses in internal controls and the risks that needed attention at national and provincial governments: “As our latest report reveals, various role players have been slow in implementing and, in certain instances, even blatantly disregarded our recommendations.”76
  3. A more pernicious strand of political interference in senior municipal staffing, often tied to contestation over service delivery contracts, resulting in a deterioration of financial governance and severe instability in municipal management: “[A]lthough in the municipal sphere, too, political criteria are formally excluded from appointment decisions, in fact they figure prominently. A vacancy in the post of municipal manager is often a pretext for some of the most vicious and debilitating factional conflicts in councils. Political appointment and control of a municipal manager enables politicisation of personnel practices right down to the lowest grade.”77
  4. An increasingly inflated and fragmented government machinery, especially at the national level.78 A stable peak of twenty-nine ministerial departments in the Mbeki period rose to a high of thirty-nine under Zuma, frustrating coordination, fueling policy uncertainty, and facilitating the rewarding of political allies.
  5. Significant turnover of cabinet ministers and senior bureaucrats.79 This trend has exacerbated leadership instability, strained political-administrative relations, and enabled appointments suspected of facilitating corruption: “The essential mechanism of ‘state capture’, where administrative decisions regarding procurement and other matters are effectively externalized into undemocratically-constituted and opaque fora, thus comes into view.”80

Figure 4 reports on trends in a process indicator: the magnitudes of irregular expenditure at the national/provincial level and at the local government level. These measures are collected as part of a mandatory audit process, required by South Africa’s Public Finance Management Act, for all units across all levels of government. Both sets of measures increased between 2009 and 2018, illustrating a growing propensity across government to flout procurement processes. However, for both sets, even at the end of the period (that is, at the peak of the increases shown in figure 4), the total amount of irregular expenditures was only about 10 percent of total nonwage, noninterest spending.81

From the perspective of citizens, the quality of service provision matters most in assessments of public sector performance. In the current, catastrophizing narrative, the perception is one of widespread, rapidly decaying quality. However, the data on service quality exhibit a more complex story of longer durée, revealing demonstrable long-term efforts to broaden access to public services. However, these initiatives have been constrained by spatial/regional and sectoral unevenness and underperformance. Urban/rural disparities, as well as divisions within urban areas between more affluent historically white community areas (with much wider access to private services) and formerly black community areas that are predominantly dependent on public service providers, are additional factors. Appendix A presents detailed sector-specific information. Here are some highlights:

  • For education, where responsibility for provision is located at the provincial level, there is anecdotal evidence that in the mid-2010s teacher recruitment and promotion began to be subject to purchasing-of-posts corruption. It was a symptom of long-standing abuses in the management of teacher allocation stretching back to the amalgamation of new provincial governments following the end of apartheid. The response to these latest reports was a high-profile ministerial investigation that, although questioning the perceived scale of the problem, found more widespread nonmonetary irregularities and undue influence in appointment practices—all of which threatened to undermine professional standards in teaching.82 Despite these challenges, the aggregate pattern is not one of accelerating decline in sectoral performance. The best available systematic measures of trends in performance over time come from global standardized tests. As of 2015, South Africa was the poorest performer among all participating countries in the Trends in Mathematics and Science Survey (TIMSS). However, between 2003 and 2015, South Africa achieved the most rapid gains in test scores of any country, with substantial gains achieved across all provinces. The big picture summary is that overall performance improved from catastrophic to only comparatively disastrous.
  • For health (also a provincial responsibility), the public sector wage bill has had to absorb an increasing number of higher paid managerial officials. This has led to a growing shortfall in the availability of nursing staff within an already overloaded public health service, which caters to the vast majority of citizens who cannot afford private health insurance and care. However, South Africa has also succeeded brilliantly in meeting what was (before the coronavirus pandemic) the largest challenge of the public health system over the past decade: the adequate stocking and provision, countrywide, of antiretroviral medications for the roughly one-quarter of the country adult population who are HIV-positive.
  • For road maintenance, a service provided by local government and a key enabler of social and economic activity, in both 2004 and 2015 the majority of respondents to the Afrobarometer survey expressed dissatisfaction with quality (with marginally less dissatisfaction in the later survey). The trends vary across provinces: two provinces show significant improvement, two show significant decline, and in five there is no trend.
  • For water and sanitation provision (also a local government responsibility), despite consistent progress in expanding service access to low-income households following the first democratic local government elections in 2000, municipalities have struggled to keep pace with the complex scale of urban demand for these services. These issues are made more complicated by a crisis of formal housing in formerly black metropolitan areas; and only marginal progress in improving rural access. Overall expenditure on repairs and maintenance is inconsistent, a problem caused in part by a lack of internal technical capacity.

In sum, public sector quality at the end of the Mbeki era remained good enough to sustain a season of hope—a continuing virtuous circle of gains along the lines illustrated in figure 3. The data summarized above and detailed further in Appendix A show some continuing decline over the subsequent decade in the quality of public sector governance and performance. However, the overall picture suggests that a “good enough” characterization continued to apply even at the end of the Zuma era. Note, though, that a focus on quantitative indicators of performance likely underestimates the damage done to the integrity and performance potential of the public sector during the Zuma period.

2011–2017: “Becoming Furious”

In May 2009, Zuma was sworn in as South Africa’s fourth president. He inherited an economy that, though buffeted by the 2007/2008 financial crisis, seemingly was fundamentally sound. Over the course of the Mandela/Mbeki era, economic growth had been accelerating and absolute poverty falling. Indeed, in the initial years of Zuma’s presidency—which included the wildly successful, celebratory atmosphere of South Africa’s June 2010 hosting of the soccer World Cup—it seemed likely that the country would continue on a positive trajectory and might even begin a new phase of renewal.

However, a hopeful scenario was overtaken by a combination of events and of deep-seated ongoing challenges caused by South Africa’s continuing extreme inequality. The events comprised Zuma’s accession to leadership and South Africa’s weak, undisciplined, and uncoordinated response to the global financial crisis, which short-circuited a virtuous circle of an economy and society on the mend. As a result, despite South Africa’s resilience immediately after the global crisis, it failed to build momentum and (contrary to other MICs) stagnated, signaling that the global shock is not sufficient to account for the subsequent reversal. This divergence is shown in figure 5.

The deep-seated ongoing challenge was the country’s extreme inequality. As per table 11, two decades after the dawn of democracy, half of South Africa’s population remained mired in chronic poverty, with limited upward mobility into the middle class. Less than a quarter of the total population, but essentially all white South Africans, enjoyed a standard of living that was middle class or better. Moreover, even though white South Africans comprised only 9 percent of the total population, they were two-thirds of the country’s high earners and accounted for 37 percent of all spending. There was ample reason for the majority of South Africans to feel that, notwithstanding the promises of mutual benefit at the dawn of democracy, the deck remained stacked against them. The stage was set for a turn of the Hirschman cycle—for “nonrealization of the expectation that my turn will soon come [to] at some point result in my ‘becoming furious’ that is, in my turning into an enemy of the established order.”83

By the mid-2010s, South Africa’s nascent virtuous spiral had gone into reverse. As per the conceptual framework laid out earlier, the downward spiral was characterized by growing recourse to personalized deal-making, rising pressure on institutions, a turn from cooperation to conflict, a rise in ethnopopulist politics, and (as per table 6 and figure 5) a stagnant economy. To better understand how and why each of these trends took hold, and how they reinforced one another, this section will probe three pivotal aspects of the process: the conflict over leadership succession that brought Zuma into power; the repurposing of SOEs; and the drivers of the ideational turn from hope to anger.

What Happened at Polokwane

South Africa’s embrace of constitutional democracy resolved two key aspects of the question of who would lead politically. Leadership would be selected by the country’s (black) majority via competitive elections. Given the ANC’s political majority and role as the party of liberation, it would emerge as electorally dominant, with sustained loyal support from the majority of the electorate. However, this left open the question of who would lead the ANC. In South Africa’s constitutional system, voting is by proportional representation; the leader of the governing party becomes the country’s president. Thus, contestation for ANC leadership has a huge impact on the polity more broadly. The ANC’s electoral conferences can be pivotal moments in shaping the country’s future. The 2007 Polokwane conference was such a moment.

The conference, at the height of South Africa’s growth boom, marked the defeat of the ANC’s post-1994 governing establishment and its guiding vision. Prior to Polokwane, South Africa’s center largely had held, albeit messily. The expectation had been that a vision of mutual gains through cooperation would be sustained and that programmatic, developmental purpose generally would be decisive in government decisionmaking. (Specifics would, of course, be contested on both technocratic and ideological grounds.) Though he was not eligible to serve a third term as the country’s president, Mbeki proposed that a close ally would succeed him as the country’s president and put himself forward as a candidate for a third term in office as president of the ANC. In this he was decisively beaten: 1,505 delegates voted for him, and 2,329 for Zuma.

There were many reasons why Mbeki lost and Zuma won. They included hubris, unforced policy errors, ideological disputes, and the emergence (fueled in part by the BEE agenda) of a political culture in which developmentally driven decisionmaking began ceding space for more narrowly personal goals. Disentangling their relative influence is beyond the scope of this discussion. Instead, Hein Marais’s vivid depiction of what happened en route to Polokwane will be sufficient introduction:

Polokwane was to be the stage for a mutiny Mbeki could no longer avoid—or survive. . . . He and his circle stood accused of abusing their power and authority, sidelining the ANC and inflaming tensions inside the organization, humiliating key allies, mismanaging major issues and using state structures to settle scores. The AIDS fiasco had squandered goodwill inside and beyond South Africa, as had Pretoria’s indifferent response to Zimbabwe’s descent into despotism and misery. . . .[Internally] he was deemed guilty of cardinal sins: ramming the 1996 Growth Employment and Redistribution (GEAR) plan into policy, steering South Africa onto a neoliberal course and supervising the emergence of a bumptious black elite while jobs vaporized and inequality deepened. . . .

An underlying frailty in the ANC’s dominance was becoming evident. . . . Perhaps most galling has been the sense of rampant unfairness—that the fruits of liberation are not being distributed fairly. Not only does daily life reiterate the scandal of inequality, but rightful access to elementary sources of wellbeing too often seem to require connections, payoffs or favors.

In the run-up to the conference, supporters were abandoning Mbeki in droves. Overnight, many were converted into ardent champions of Zuma. . . . In key respects, the succession battle was an undignified jostle for positions, leverage and influence. Individuals who had been slighted, wounded or marginalized during the Mbeki years now found common purpose with “comrades” nursing old grudges, those whose entrepreneurial or political ambitions had been thwarted, whose ineptitude or greed had upended their careers or those who had been shut out for their politics or principles.84

Jacob Zuma acceded to South Africa’s presidency in 2009, following another (decisive) ANC victory at the polls.85 Over his nine years in office, he governed in an increasingly personalized way, with increasing recourse to polarizing rhetoric. This increasingly personalized mode of governance was in part a consequence of deeply rooted pressures emanating from South Africa’s unequal society. But the quality of political leadership also was crucial for what transpired.

When Zuma took office, many who backed him hoped that he would bring an inclusive, coalition-building, popular touch to leadership—a contrast to Mbeki’s remote, technocratic, and somewhat imperious style. In the event, Zuma proved to be a cunning, ruthless, and charismatic tactician. The Moscow-trained head of intelligence of the ANC in exile, he was said by many to live up to the meaning of his middle name, Gedleyihlekisa (“the one who laughs, while grinding his enemies”). Had Zuma, instead of deploying his skills to loot the economy on behalf of his friends and clients, used his credibility with the poor and the organized working class to recast the social contract there could well have been a different outcome instead of steady decline.

A Turn Toward Personalized Governance: South Africa’s State-Owned Enterprises

During the Zuma presidency, the focus of efforts at elite transformation shifted away from BEE initiatives to reshape ownership and control of the business establishment and toward the repurposing of SOEs. This change created new pressures for South Africa’s institutions of accountability.

SOEs have a long history at the center of South Africa’s political economy. State-owned railway, electricity, and steel companies date back to the early 1920s. SOEs were an integral part of the country’s import-substituting industrialization strategy in the 1940s and 1950s, and new SOEs were established in the 1960s and 1970s in response to anti-apartheid economic boycotts.86 The way in which these enterprises have been governed since 1994 encapsulates both the best of the earlier and the worst of the later democratic era.

With the issuance in 1997 (revised in 2002) of a Protocol on Corporate Governance in the Public Sector, South Africa promulgated a best practices set of institutional arrangements. These arrangements feature in a 2006 World Bank publication, Held by the Visible Hand, which was intended to provide a global template for SOE governance. As the publication details:

In South Africa, the Department of Public Enterprises has issued Protocols on Corporate Governance in the Public Sector that apply to the six SOEs it supervises, but other SOEs as well. These protocols provide detailed guidance on the board and financial oversight of SOEs as well as their obligations to stakeholders and social objectives. . . . A shareholder compact is agreed between the SOE and its shareholders.

It is a joint responsibility of the SOE board and the government (and other shareholders) to ensure that the shareholder compact is developed . . . all SOE boards are required to issue Directors Reports [which include] a review of the financial performance of the past year, information relating to internal and external factors influencing SOE performance, stressing risks and opportunities and strategies to manage them. . . . [F]inancial and other effects of directions from the Executive Authority or other political body.87

It is easy to overstate the practical impact of this protocol in the years immediately following its issuance: both the PetroSA and Eskom/Medupi examples noted earlier happened during this supposed good governance heyday. Nonetheless, the way in which SOEs were governed before the 2010s was part of the Mbeki-era “good enough” balance between rules and deals—one which enabled confidence in the future to rise, growth to accelerate, and poverty to fall.

In retrospect, the moment when SOE governance turned from largely benign to something different can be identified: the November 2010 resignation of the minister of public enterprises, Barbara Hogan, and the appointment of Malusi Gigaba. The 2016 State of Capture report issued by Thuli Madonsela, the determinedly independent head of South Africa’s Office of the Public Protector, provides details.88 According to that report, Hogan’s resignation came after repeated efforts by Zuma and ANC Secretary General Gwede Mantashe to intervene in the appointment of board members in the two largest SOEs, Eskom and Transnet. The final straw was a conflict over whether South African Airways should cede the potentially lucrative route to Mumbai to an Indian-owned airline championed by the Gupta family.

How and why the governance of South Africa’s SOEs took a different turn is spelled out in compelling detail in the 2016 Betrayal of the Promise report.89 Here is how the authors put it:

Jacob Zuma’s presidency has been aimed at repurposing state institutions to consolidate a Zuma-centered power elite. . . . These [Zuma-centered] networks are pursuing two aims. The first is to drive a transition from traditional BEE, which was premised on the possibility of reforming the white-dominated economy, to “radical economic transformation” driven by groups disguised as a black capitalist class independent of white monopoly capital. The second is to drive a transition from acceptance of the constitutional settlement and the “rules of the game” to a re-purposing of state institutions that is achieved, in part, by breaking those rules. . . .

The battleground for economic transformation was shifting away from the economy itself to the state and, specifically, to SOEs that outsourced massive industrial contracts to private-sector service providers. . . . [This required] preferential procurement from black-owned companies . . . [but] the existing constitutional and legal environment constrained this model of economic transformation by insisting that bidders for state contracts satisfy more than racial conditions. Price and experience were also considerations.90

Once constraints of price and quality are eased, the boundary between a political agenda of radical economic transformation and a corrupt private agenda of state capture becomes paper-thin. The role of the Gupta family business empire during the Zuma presidency is one illustrative example.91

The Guptas venture into South Africa began in 1993 when Atul Gupta, having failed to find an entry point for business in China, arrived in South Africa in search of opportunities. He was joined there by his brother Ajay in 1995 and other brother Rajesh in 1997. Within a short time, the brothers began associating informally with senior ANC figures. They apparently first met Zuma in 2002. In 2008, the year after Zuma was elected ANC president, his 26-year-old daughter Duduzile was invited to join the board of one of the Gupta family’s companies; his son Duduzane joined a Gupta company a little later on. In 2010, the Gupta family started a media company that quickly received major advertising buys from SOEs. The extended Gupta family prudently left South Africa in April 2016. Between June 2015 and December 2017, there were “known outflows from Gupta-linked companies and individuals to a group of Hong Kong companies amounting to R4.9 billion (somewhat under US$500 million).”92

Tracking the governance and operation of South Africa’s SOEs from 2010 to 2017 reveals the formula used to try and repurpose them to serve the overlapping agendas of radical economic transformation and private enrichment. This formula had the following steps:

  • Put in place a supportive minister of public enterprises.
  • Reshuffle SOE boards of directors.
  • Appoint a compliant/supportive chief executive officer (CEO).
  • Override established rules governing decisionmaking, especially those concerning procurement—including via direct board intervention in operations.

Overriding the rules requires not only capture of the SOEs but also compliant institutions beyond the SOEs themselves. This is why sustained and sometimes successful efforts to deploy loyal functionaries in both the justice system and the National Treasury have been part of the state capture agenda.

The examples of South Africa’s two largest SOEs, Transnet and Eskom, indicate how this process played out between 2010 and 2017.93 To give a sense of scale, Transnet’s procurement in 2010/2011 amounted to R70 billion and Eskom’s to R74 billion; between them they accounted for 70 percent of total procurement of R204 billion in that year by South Africa’s thirteen major SOEs. (Over the 2009–2012 period, the exchange rate was approximately R8 per dollar.)

Transnet. Transnet comprises rail, port, and pipeline divisions. The story of how the SOE was built and the ongoing challenges posed by its unwieldy divisional structure (cobbled together from multiple distinct entities) is outside the scope of this discussion. Rather, the focus is on some aspects of its governance and operation subsequent to 2010.94

In February 2011, three months after Gigaba had become the overseeing minister of public enterprises, Brian Molefe was appointed Transnet CEO.95 During both Molefe’s tenure as CEO and that of his successor Siyabonga Gama, there were multiple reports of questionable contracting arrangements.96 Molefe’s appointment was accompanied by significant reshuffling of the Transnet board. These included the resignation of Juergen Schrempp, who had been appointed as a nonexecutive director only a few months earlier; and the appointment of Iqbal Sharma, a former official in South Africa’s Department of Trade and Industry (from which position he had come into contact with the Gupta family).

This reshuffle would have consequences that seemed beyond coincidental. In 2012, Sharma was appointed to chair a new Board Acquisitions and Disposals Committee to supervise the planned pipeline of future infrastructure spending. A central part of the committee’s work was supervision of a massive (R51 billion) tender for over 1,000 new locomotives. China South Rail (CSR), which had apparent ties to the Guptas, had an inside track in the locomotives deal and was one of the four winning bidders.97 Even beyond the initial tender, a separate contract for one hundred locomotives was carved out and awarded to CSR—ignoring the recommendation of the technical team responsible for the program and notwithstanding the fact that CSR’s winning bid of R4.4 billion was R1.2 billion higher than a technically better-aligned and lower-cost (R3.2 billion) bid by Mitsui. Local sourcing was a key aspect of the locomotives deal; the winning bidders were required to source up to 60 percent of their components from South African contractors. Prior to the bid award, all four of the winning bidders had visited VR Laser, a local metal-body construction company. A few weeks before the announcement of the winners, “a company in which Iqbal Sharma, Rajesh Gupta and Duduzane Zuma were partners, acquired a stake in VR Laser Services.”98

Eskom. State-owned Eskom is South Africa’s dominant supplier of electricity. It accounts for the vast majority of electricity generated in the country, controls the transmission grid, and supplies power both to municipally owned distribution networks downstream and (where there is no independent municipal provider) to its own distribution network. In 2001, Eskom was named “power company of the year” at the Financial Times Global Energy Awards in New York. Subsequently, things changed.99 In June 2011, six months after becoming minister, Gigaba dismantled the long-standing Eskom board, replacing all but two of its nonexecutive directors. Soon thereafter, he imposed the requirement that, by 2018/2019, all of Eskom’s coal supply contracts would be from mining companies with more than 50 percent black ownership.

The next few years were times of continuing challenge for Eskom governance. Gigaba refused to sign off on a R4 billion tender to replace steam generators at South Africa’s only nuclear power plant, which had been competitively awarded to Westinghouse. He also intervened to mandate that Eskom work with a specific audit firm. The board was deadlocked over controversial contracts, including a contract with the Gupta media empire, and a set of (technically sound) proposals to insource management of Eskom’s information technology systems—proposals that were estimated to save R1 billion. “From 2013 onwards, Eskom began to hemorrhage executives and senior staff, many of whom were suspended on trumped up charges (without proof) and refused fair disciplinary hearings.”100

Following this upheaval, 2014 was a watershed year, and not least because of the onset in November of an unexpected new round of large-scale “load-shedding” that had plagued South Africa’s electricity supply in 2007. Back in March, CEO Brian Dames, no longer able to weather the continuing governance storm, resigned. Shortly thereafter, the nuclear steam generator contract was awarded—and not to Westinghouse—and the Guptas received a new, expanded media contract. In May, Lynne Brown was appointed minister of public enterprises, and in December she “made sweeping changes to the board, bringing in at least six Gupta-connected members.”101

In March 2015, in the midst of the load-shedding crisis, South Africans were treated to the spectacle of an Eskom governance circular firing squad. This involved, in short order, the announcement of a so-called forensic inquiry into Eskom governance, the suspension of senior Eskom executives at explicit order from top political levels, and the downgrading of Eskom bonds. In addition, the board chairman, who had resisted political pressure, was removed and replaced with Ben Ngubane, who was alleged to be close to Zuma. In April 2015, Molefe was redeployed from Transnet to become Eskom CEO. In August, Transnet’s chief financial officer, Anoj Singh, also was redeployed to Eskom to take on a similar high-level role.

Accompanying the governance issues was a series of actions on the coal supply front.102 Coal is Eskom’s largest procurement line item at 120 million tons per year for around R50 billion.103 A Gupta-owned company, Tegeta Exploration and Resources, purchased the Brakfontein coal mine in 2011. However, coal from the mine was reported to be substandard, and purchases were not approved by Eskom’s technical team; only in January 2015 did Eskom sign its first contract with the mine. Meanwhile, back in July 2013, a major supplier, Glencore-owned Optimum Coal, had invoked a hardship clause; it claimed that its per-ton supply costs had escalated to R300, too high to be covered by a R150 per ton fixed-price contract. In March 2015, Eskom’s executive procurement committee approved a cost escalation, but this increase was rejected by Molefe immediately upon becoming CEO. Glencore subsequently sold Optimum Coal to Tegeta, a sale facilitated by a direct intervention by South Africa’s new minister of natural resources, Mosebenzi Zwane, and an April 2016 prepayment by Eskom to Tegeta to enable the company to complete the purchase transaction after it had been unable to raise money for the transaction on financial markets. A variety of coal supply arrangements were restructured: in December 2015, Tegeta coal was redirected from a power station where it had been paid R150 per ton to a different power station through which it earned R467 per ton.104 More broadly, once the Optimum deal was completed, “the price of coal supplied by Tegeta rocketed from the R161 per ton paid to Glencore to R550 per ton.”105

Nuclear power. Beginning in 2011, Zuma, having met with Russia’s then prime minister Vladimir Putin, signaled his desire to conclude a nuclear power deal with Russia.106 In September 2014, a few weeks after yet another meeting between Zuma and senior Russian officials, a nuclear cooperation deal between South Africa and Russia was signed in Vienna with the Russian company Rosatom. The details as to what was binding and what was not remain obscure, as does the proposed magnitude of the deal, which was reported to involve a suite of up to eight nuclear plants, at a cumulative estimated construction cost of $40–$50 billion.107 What was not obscure was that the entire process unfolded outside of both South Africa’s procurement and energy supply planning processes. The resulting furor consumed South African politics. It featured centrally in the Public Protector’s 2016 State of Capture report.108 Once Zuma vacated the presidency, momentum for the nuclear deal seemed to have dissipated.

Although the series of deals with the Gupta family epitomizes the Zuma era, the Guptas were not the only private sector participants. The investigations that began with the public protector and continue to date have identified a range of facilitator beneficiaries, including international companies such as KPMG, Bain, and McKinsey.

Shifting Ideas: A Turn to Ethnopopulism

Along with growing predation on public assets and increasing pressure on public institutions, the 2010s also witnessed a stark turn in the tenor of South Africa’s political discourse. As a prelude to exploring the reasons for this turn from hope to anger, it is important to reiterate that the deep underlying cause was continuing economic inequity. But ideational drivers are not simply an intermediate step in the causal chain of inequality, institutional quality, and development outcomes. Ideas can have an independent impact; they are an important currency of political persuasion. As Rodrik puts it:

Policy arguments in the real world rarely rely on a naked appeal to narrow economic interests. In public discussions what we might call ‘ideational politics’ seems at least as important as interest-based politics. . . . Ideas [can] shape interests [by] altering voter preferences and/or shifting their worldview.109

Ideational politics play out across many different domains. Where inequality is the domain of interest, the ideational challenges are different for political entrepreneurs who support the status quo and those who challenge it. Politicians selling “hope” must persuade citizens either that inequality is irrelevant or that they will bring greater inclusion. Politicians selling “anger” focus on the injustices of the status quo. South Africa’s turn in the 2010s from hope to anger comprised, in part, a necessary effort to further reckon with the country’s difficult history. However, as has been the case in many countries in recent years, political entrepreneurship also was at play, often with more than a little self-interested opportunism in the background.

Reckoning with the ideational legacy. Hirschman’s analysis of the tunnel effect provides a useful point of departure for understanding why, in accounting for a turn from hope to anger, it is not enough to focus on the economic dimensions of inequality. Hirschman understood that the turn to an angry phase of the tunnel effect was driven by more than the reaction of those left behind economically:

The upwardly mobile do not necessarily turn into pillars of society all at once, but may on the contrary be disaffected for a considerable time. . . . They may have risen along one of the dimensions of social status, such as wealth, [but] find that a number of obstacles, rigidities and discriminatory practices still block their ascent . . . and consequently feel that in spite of all their efforts and achievements, they are not really ‘making it.’110

Unsurprisingly, given the country’s fraught history, there was a distinctly racial flavor to South Africa’s angry turn.

Racial tropes have long been central to South Africa’s political discourse. The promise (or perhaps illusion) of the “rainbow miracle” was that racist tropes had been left behind. Although the country’s racist past needed to be addressed, such as by a Truth and Reconciliation Commission, this factor did not need to derail the cooperative endeavor as a whole. One all-too-easy way of addressing this—an approach unsurprisingly embraced by many white South Africans—was through a form of amnesia, taking the view that South Africa’s transition to democracy had washed the country’s history clean. Another more challenging but also ultimately hopeful approach to resolving the seeming contradiction between a racist past and an inclusive new beginning was captured in Stephen Biko’s classic dictum from the 1970s: “Whites must be made to realize that they are only human, not superior. Same with blacks. They must be made to realize that they are also human, not inferior.”111

For the first two decades of democracy, the amnesia approach seemed largely to have won out, especially among elites. But in 2015, after two decades of quietism, South Africa’s university campuses erupted with the emergence of a “Fallist” (“Rhodes Must Fall”; “Fees Must Fall”) student movement. The movement began at the University of Cape Town (UCT), the country’s premier university—admission into which seemingly offered a ticket into the elite, including for black South Africans.112 The movement was precipitated by a controversy over the pride of place given on the campus to a statue of arch-imperialist Cecil John Rhodes. This protest resulted in the removal of the Rhodes statue in April of that year, then quickly morphed into a national “Fees Must Fall” student movement, protesting the continuing economic and social obstacles that confronted the supposed next-generation black elite. This movement also had a successful outcome, with an agreement by government in 2018 to radically expand support, including the elimination of all university fees (plus some grant support for living expenses) for households with family incomes below approximately $25,000.

The Fallist movement combined an assault on continuing inequities (at least within those strata of young people who aspired to post–high school education) with an assault on the ideational structures associated with these inequities. Student activist Rekgotsofetse Chikane, the son of Frank Chikane (an eminent figure in the struggle against apartheid and subsequently director general in Mbeki’s presidential office) vividly captures the spirit of the early Fallist movement:

Chumani Maxwele, for some, will be remembered as the man who threw shit on a statue. . . . Though protests had taken place at UCT before, none had ever been as visceral as Chumani’s. Standing shirtless in front of the statue of Rhodes, which he had desecrated with human faeces . . . Chumani held up a sign that read: ‘White arrogance @ UCT. . . .’

Many, within the university and across the country, had fallen into the trap of believing that the protest was about the statue and not what it was actually about, namely the pervasive nature of institutionalized forms of racism on campus. . . . We had become orators of our circumstances and warriors of its change. We were disciples of the message of decolonization and teachers of its practice. We learned how to love ourselves as black children of South Africa.113

Ethnopopulist political entrepreneurship. In parallel with the Fallist movement, a variety of South African political entrepreneurs became more inclined to embrace a newly racialized discourse. Before turning to the details of this discourse, it is helpful to step back and consider more generally how populist discourses take hold.

A useful point of departure is Sharun Mukand and Dani Rodrik’s distinction between two dimensions of ideational ethnopolitics. In the “worldview” dimension, political entrepreneurs work to “shape the electorate’s understanding of how the world works, which in turn alters its perceptions of the proposed policies and their outcomes.”114 In the “identity” dimension, advocates focus on “shaping or priming voters’ self-identity. . . . Individuals have a multiplicity of identities—revolving around ethnicity, race, religion or nationality.”115 The identity and worldview “memes” reinforce each other:

Voters who share an identity with the political challenger [may] want to support him because of the tug of shared identity. However, supporting the policies of the political challenger could economically hurt the low-income voter . . . recalcitrant low-income voters can be targeted using a combination of policy and identity memes (where neither would have been sufficient to switch voter allegiance on its own).116

The political discourse embraced by South Africa’s Economic Freedom Fighters (EFF) is a textbook example of ideational politics in action. The 2012 expulsion from the ANC of Julius Malema, the firebrand head of its Youth League, and his creation the following year of a new political party, the EFF, was a turning point in the emergence of South Africa’s contemporary ethnopolitics—both for its direct effects and for its subsequent influence on the mode of discourse employed in the ANC’s internecine conflicts. At the core of the EFF’s strategy for mobilizing political support was an angry rhetoric that targeted both the country’s continuing inequalities and their ethnic correlates. Malema outlined the EFF’s economic vision as follows: “Soldiers and the police are not our enemy; our enemy is white monopoly capital and their political co-optees. . . . [The party’s] plan includes the non-negotiable principles of land expropriation and nationalization of mines, both without compensation. The EFF seeks to move away from a discourse of reconciliation to one of justice.”117

The EFF has become the third-largest political party in South African politics. In 2014, it won 6.4 percent of all votes cast; in 2016 (in municipal elections, held nationally), it won 8.2 percent; in 2019, 10.8 percent. Figures 6 and 7 track the use in three major media outlets (none aligned with the EFF) of two EFF memes that combine worldview and identity politics:

  • White monopoly capital (WMC), which fuses the long-standing radical critique that the continuing dominance of “monopoly capital” is the root cause of South Africa’s poverty and inequality, with the proposition that capital is “white,” and thus part of the colonial past that needs to be expunged.
  • Expropriation without compensation (EWC), which identifies the colonial conquest of land as the country’s original sin and calls for Zimbabwe-style expropriation of land without compensation. Although this call is seemingly narrowly focused on farm land, given that two-thirds of South Africa’s population is urban—and that the movement pushing for EWC is urban, not rural—EWC is widely interpreted as a proxy for a broader questioning of constitutionally protected property rights.

Although both memes were part of the EFF’s economic manifesto, the manifesto was not the driver of their usage in the period covered by the tables.

The WMC meme was fueled by a British public relations consulting company, Bell Pottinger. The following summarizes some of the findings of news investigations of this “strange bedfellows” partnership:

Bell Pottinger took on the Gupta family as clients in 2016 to try to improve their image, and the chosen strategy was to target white business leaders as a distraction from serious allegations of state capture.

One of the strategies was apparently to drive a predominantly social media narrative that ‘white monopoly capital’, the SA Communist Party and National Treasury have been standing in the way of transforming the South African economy.

The Bell Pottinger plan reportedly involved using, among other things, Twitter bots involved in a fake news campaign to support messages critical of white monopoly capital and be defensive of the Guptas.118

In the wake of the exposure of the Bell Pottinger campaign, dissemination of the WMC meme slowed but did not end entirely. In November 2017, in the run-up to the ANC’s elective conference, Zuma said the following about WMC:

It is a fact, it’s not manufactured. We are telling the truth. We fought, they took everything, political power which we now have. They took economic power and land. Let us take the mines, companies who dominate in mines, few are benefiting. You will find the same companies in charge. That means they are dominating. They are monopolising, they are not black. . . . You have companies that are white. . . . They are monopolies, they should be called by what they practice. There is monopoly capital and in SA it is white. Because of our history it does have a colour and it is white.119

In 2017, the EWC meme accelerated as well. In February, Malema proposed a constitutional amendment to allow for EWC. In March, Zuma added his voice to the call.120 In subsequent months, EWC was furiously debated within the ANC. A motion to support EWC was added to the roster of items for consideration at the ANC’s December elective conference; it passed, following heated debate.121 The ANC continues to struggle with how to square the circle of a renewed commitment to reconciliation, with a continuing push for EWC by some of its influential members.

In sum, Zuma rode to power on a wave of discontent with the able but imperious Mbeki. Various groups with differing motives articulated their anger and frustration over large continuing inequalities during the Zuma era with slogans such as WMC/EMC, “Rhodes Must Fall,” and “Fees Must Fall.” These slogans represented real grievances, but they also were used and abused by political entrepreneurs within and outside government as a populist vehicle for furthering political and personal ambitions. In particular, some political leaders used them as cover for a regime that was no longer focused on building a pact for growth and redistribution but had been redesigned to channel patronage to an increasingly greedy machine. The outcome was a steady decline in trust between economic and political elites and a steadily deteriorating economy. Since 2014, every year has seen a decline in the real average per capita income of South Africans. The current government, hampered by indecision and the coronavirus pandemic, has not yet been able to reverse the downward drift.

Patterns and Implications

In December 2017, Ramaphosa narrowly defeated Zuma’s preferred candidate as his successor as ANC president. In February 2018, Ramaphosa became president of South Africa. His victory helped to press a pause button on what might have seemed an inexorable, continuing spiral down from South Africa’s extravagant hopes in the first decade of democracy.

What were the strengths that put a halt to accelerating state capture? What can be learned from the experience that can help craft a way forward to a more thriving, inclusive hopeful trajectory? What lessons from South Africa might be more broadly relevant? Before considering any of these questions, it is helpful to summarize the findings to this point as to what went wrong in South Africa after the high expectations at the dawn of democracy in the early 1990s.

Unmet Aspirations and Their Consequences

At the root of South Africa’s reversal was a failure to address adequately the challenges of inequality and exclusion. This section provides an overview of the channels through which continuing high levels of inequality can influence development outcomes.

Certain aspects of inequality influence the shape and robustness of a country’s “elite bargain,” and others influence the mode of incorporation of nonelites. If South Africa was to realize its aspiration at the outset of democracy to a cooperative way forward, each of these aspects needed to be addressed in a “good enough” way. Achieving this goal turned out to be beyond the reach even of the visionary leadership with which South Africa was endowed in the first dozen or so years of democracy. For South Africa’s putative elite bargain to be robust, three distinct subbargains all needed to work effectively:

  • A subbargain between the established (overwhelmingly white) economic elite and the country’s new political leadership. This involved four sets of commitments: (1) an overarching commitment to cooperation (with political power in the hands of the electorally dominant ANC and economic power continuing to be overwhelmingly in the hands of the white elite); (2) a commitment to sustaining the rule of law, including the protection of private property; (3) a commitment to ongoing economic transformation (including via BEE); and (4) given that white elites comprised the predominant source of public revenues, a commitment to support pro-poor fiscal restructuring.
  • A subbargain among the new political elites within the ANC. Although political power had shifted to the ANC, the questions of how it would be exercised (and who among the new elites would benefit) remained unresolved. Not only does the ANC explicitly describe itself as an alliance, including the Congress of South African Trade Unions (COSATU) and the South African Communist Party, the ANC itself is a broad tent encompassing many ideological proclivities; degrees of public-spiritedness; and regional, ethnic, and economic interests. The ANC’s implicit promise was that its formal structures, plus the structures of government, would channel this diversity toward a shared national purpose.
  • As part of the subbargain within the ANC, a commitment to protect the interests of new (predominantly black) middle-class insiders. This is highlighted as a third subbargain because of the distinctive economic and political dynamics involved. This insider middle class is defined broadly. It ranges from white-collar middle managers and skilled production workers within government to unionized blue-collar workers to formal sector employees earning somewhat above a formally mandated minimum wage.

In the initial years of democracy, the ANC’s credibility as a broad-based liberation movement enabled it to incorporate nonelites into the bargain. Indeed, one of the attractions to apartheid-era elites of negotiating with the ANC was that its certain accession to power via a democratic election would confer legitimacy across society for a new, negotiated constitutional order that promised equal rights to all. Beyond the promise of equal rights, two sets of promises to nonelites, inherent in the ANC’s electoral slogan of “a better life for all,” were key:

  • A promise of upward mobility. For those on the cusp of middle-class status (the “vulnerable middle” and the “transient poor” as per table 11), a combination of education, job creation, and an end to racial discrimination would open up a readily accessible pathway into the middle class.
  • A promise to reduce extreme poverty. A post-minority-rule redirection of public resources and services would benefit the population as a whole.

With the exception of the commitment to reduce extreme poverty, which was unequivocally pursued at least into the 2010s, each of the subbargains and promises identified above was less than realistic. Each is better described as an aspirational commitment—anchored, to be sure, within a carefully negotiated democratic constitution and strong formal institutions but falling short of being a robust deal. The ANC’s implicit commitment to align its contending interests around a shared national purpose came more from faith than from experience governing a country. The promise of upward mobility and the commitment to protect the interests of existing middle-class insiders were in direct conflict with one another. The BEE initiatives to transform the ethnic complexion of the economic elite, both ownership and managerial, offered less than it might have seemed on the surface. The initiatives did not address in any systematic ways the country’s extreme concentration of wealth and income.

Had South Africa been able to enjoy East Asian rates of rapid economic growth, averaging 6 to 10 percent annually for a sustained period of time, the expansion of opportunity throughout society might have trumped the limitations of each of the various aspirational commitments. In reality, an annual rate of 5 percent was reached only briefly from 2005 to 2008. Even so, for the first fifteen years of democracy, these aspirational commitments, however vague, seemed to provide a “good enough” platform for a hopeful way forward. The ideational dimension of South Africa’s “rainbow miracle” remained intact. Indeed, as of the latter 2000s, it still seemed possible that a collective aspiration to build a better society through cooperation could hold. There was hope that Zuma’s accession to the presidency could help reignite confidence and set in motion a renewed cycle of growth. However, as the 2010s progressed, it became evident that the trend was going in the opposite direction. In the absence of rapid growth, the limitations of the promises made for addressing the various dimensions of inequality became apparent. New tensions surfaced and fed on one another. Four successive turns set in motion what looked to be an accelerating downward spiral of decline.

The first turn in the spiral was rising pressure on institutions. This pressure was sparked by the continuing ambiguities and unresolved tensions within the elite bargain, and Zuma’s personalized and sometimes predatory approach to leadership added fuel to the fire. Within the upper echelons of the ANC, various subgroups engaged in increasingly relentless, uncompromising conflicts for control over resources and positions of influence, as seen in the vicious internal strife of the ANC’s 2007 conference at Polokwane.

Decisions as to who would participate in BEE initiatives became part of the ANC’s interelite conflict. Private companies began making decisions as to who would be their BEE partners on the basis of their political influence networks. Around this time, SOEs emerged as an especially lucrative area for the most powerful and well-connected individuals and factions to engage in systematic looting. Control over key political and management positions within provincial and local government became a means both of responding to the promise made to middle-class insiders and of building personalized networks (and opportunities for self-enrichment) across the country. As per-capita income growth entered negative territory, insiders within the ANC became enmeshed in zero-sum contestation over positions and resources at all levels of politics and government, whether national, provincial, or local. Institutional and governance decay undermined confidence in the cooperative vision of the future, lessening the possibility of a renewal of growth.

The second and third turns in the direction of a downward spiral are the ones highlighted by Hirschman in his “tunnel effect” metaphor: a rising tide of disillusion, followed by a sudden ideational turn to anger. Disillusion took hold across a broad swathe of society. The losers in elite conflicts within the ANC felt aggrieved. Those on the cusp of the formal economy found themselves unable to consolidate middle-class status; job creation in the formal sector was anemic, except for a few years in the mid-2000s.122 Unemployment steadily increased. This economic malaise fueled dissatisfaction not only among those who felt directly thwarted despite their best efforts but also among the population in general. Meanwhile, income and wealth remained massively concentrated, disproportionately in the hands of the white community. BEE involved no systematic changes in tax structure; indeed, throughout the first quarter-century of democracy, South Africa’s taxes on assets remained below those of the United States. Even on its own terms, the gains from BEE were limited: as of 2018, 69.6 percent of top management within the private sector was white (down from 73.7 percent in 2008), as was 60 percent of senior management (down from 67.9 percent in 2008).123

Disillusion resulted in the third turn in the spiral, an ideational turn toward anger. The angry turn was catalyzed by both genuine grievance and political opportunism. In the face of thwarted opportunity, an increasing number of South Africa’s population came to see the privilege enjoyed by the mostly white economic elite—and the tide of apparent corruption that seemed to be the only way that new elites could share in that privilege—as a provocation. In turn, opportunistic ethnopopulist political entrepreneurs sought to use the disillusion to strengthen their position within interelite political struggles.

As of 2017, while formal institutions remained mostly functional, they were under increasing strain. All the elements seemed to be in place for a fourth downward turn in the spiral: a rapidly accelerating cumulative slide, with weakened economic performance, institutional decay, and the ideational shift toward anger and ethnopopulism feeding on one another in an accelerating race to the bottom. Then came the election of Cyril Ramaphosa as president.

South Africa’s Surprising Strengths

Going into the ANC’s December 2017 elective conference, Zuma was confident that he was in control of events, and that his ex-wife Nkosazana Dlamini Zuma would be elected as his successor. In the event, Ramaphosa won by the slimmest of margins. The blow-by-blow details of how he achieved victory are not relevant here, though it is worth noting that his victory was secured by the defection of one of Zuma’s more senior allies, David Mabuza—and that Mabuza is now South Africa’s deputy president. Of broader interest is the way in which the ground for Ramaphosa’s success was laid by the determined pushback of a combination of civil society, including the media, and those official checks-and-balances institutions, including the judiciary, that largely remained uncaptured during the Zuma era.

Underlying most of South Africa’s fail-safe mechanisms as it entered a spiral of self-destruction was the far-sighted constitution that was agreed to in the South African Parliament on May 8, 1996. Through its effective separation of powers and protection of freedoms, the constitution provided the space in which critics and rebels could challenge the government. Even though Zuma defanged the police and prosecutorial system by inserting his supporters into the leading positions, he was not able to do significant damage to the judiciary and to some other constitutionally created and protected checks-and-balances institutions.

To begin with the judiciary, when Mogoeng Mogoeng was proposed by the Judicial Service Commission, confirmed by parliament, and appointed in 2011 as chief justice by Zuma, some feared that this socially conservative self-professed religious Christian might be too timid to pursue justice with vigor.124 This fear was unfounded; the Constitutional Court and the other key courts largely retained their integrity throughout the Zuma era and beyond. Prosecutions and judicial enquiries were able to proceed, albeit within the limitations of a damaged prosecutorial system.

Among the other formal checks-and-balances institutions, the most important was the Office of the Public Protector, which was set up by South Africa’s constitution to investigate complaints against government agencies or officials and to propose remedial actions. The force of this office depends on the courage and stamina of the public protector. During this vulnerable period, South Africa was fortunate to have its toughest and most effective public protector, advocate Thuli Madonsela. Of her many important investigations and reports, the best known is the State of Capture report late in 2016, shortly before the end of her term, which drew on whistleblowers’ testimonies and a cache of private emails to reveal the abuse of state institutions for personal gain. The Auditor General’s office also played an important role reporting on abuses of state funds and assets.

The role of media was critical, amplified, and supported by a variety of nongovernmental activist organizations. Some of the findings of the Office of the Public Protector were anticipated by investigative reports in the media. According to the Reporters Without Borders’ Media Freedom Index, South Africa has one of the freest media environments in the world.125 It is second in Africa, behind Namibia, but ahead of “free” countries such as France, Italy, the United Kingdom, and the United States. South Africa also has a tradition of courage in journalism, stemming back to the apartheid era when reporters risked their lives and personal freedoms to uncover stories of oppression and some brave editors carried their stories. Courageous whistleblowers helped the media and other investigators with personal testimonies, documents, and memory sticks loaded with confidential emails and other data. Although the commercial media is in decline, investigative reporters banded together in in-depth reporting groups to secure funding from foundations supporting freedom of information. Many of the illegalities and transgressions of the increasingly corrupt government were first exposed in the print and internet-based media. Academics dug deeper and verified and expanded disclosures made in the media. Civil society organizations provided vehicles for mobilizing against encroaching state capture.

Parliament itself was an important site of protest. Opposition parties used the freedom that parliament allowed to voice objections vigorously and theatrically. The populist EFF disrupted several State of the Nation addresses by Zuma and forced the speaker of the house to call on parliament’s security officials to expel them. Their protests were broadcast on national television. Opposition parties could also pose challenging parliamentary questions and make speeches that publicized government abuses.

What finally proved to be decisive was Zuma’s progressive loss of support from some key factions within the ANC. For a long while, the ANC alliance remained firmly behind Zuma, with some exceptions. Former COSATU general secretary Zwelinzima Vavi, then a powerful ally of the ANC and instrumental in Zuma’s rise to power, broke early on with Zuma over ethical issues. He was forced out of COSATU and started a rival trade union grouping. A group of ANC veterans of stature, if not political power, banded together and spoke out strongly. After many revelations, including the State of Capture report, some leaders within the Communist Party (part of the ANC alliance), the parliamentary ANC, and even the remaining rump of COSATU became openly critical. As the rise of David Mabuza signals, it would be a mistake to characterize the narrow election of Ramaphosa as ANC president in December 2017 as an unambiguous triumph of principle over capture. Even so, there is no doubt that the cumulative clamor against Zuma-era corruption and the continuing reputation that Ramaphosa generally enjoyed as a committed constitutionalist contributed to the outcome.

Crafting a Way Forward

Notwithstanding Ramaphosa’s accession to South Africa’s presidency, the country is not out of the woods. To date, the number of prosecutions of corrupt insiders has been limited, though there are signs that court actions are beginning to accelerate.126 Ramaphosa has not yet been able to move beyond a promise to stop the rot and offer a renewed positive vision as to how South Africa might move forward. The country’s deep-seated, underlying challenges remain.

Beyond the immediate tasks facing South Africa—including how to respond to the COVID-19 crisis—it is worth exploring three divergent scenarios as to what the country’s future might hold. The most positive scenario would be an inclusive development option, anchored in a renewed societal commitment to cooperation and the pursuit of win-win outcomes. A second, less optimistic scenario would involve more muddling through. A third scenario could be a renewed ethnopopulist push. The inclusive scenario is framed along the lines that Hirschman suggested were key to Latin America turning the corner from anger to renewal: “efforts at catching up on the part of lagging sectors and regions, at social reforms to improve the welfare and position of groups that have been neglected and squeezed, and at redistribution of wealth and income in general.”127 (Appendix B details some key elements of a putative South African agenda for fostering inclusive reform.)

Table 12 uses simple arithmetic to depict the three scenarios schematically, in terms of changes in aggregate income and in the distribution of that income. The bottom row of the table suggests, for each scenario, the potential magnitude of real income changes (against a baseline set at 100 points) over a five- to ten-year period; the middle rows depict associated changes in the distribution of that income. In the muddling-through scenario, aggregate income rises by 20 percent. In the ethnopopulist scenario, it falls by 20 percent. In the inclusive scenario, it rises by a hypothetical 40 percent.

The muddling-through scenario is the path of least resistance for established elites, and unsurprisingly has emerged as the preferred option of many. They have advocated for the country to return to “the basics,” to reembrace the trajectory of the Mandela and Mbeki presidencies. Such a back-to-the-future approach would give renewed priority in decisionmaking to pro-growth economic policies and to improved public sector performance, while addressing other social and political concerns on the margins.

However, this muddling-through scenario is unlikely to have the broad-based political support needed for it to be sustainable over the medium term. To be successful, a pro-growth strategy needs credible commitment that a business- and market-friendly policy regime is sustainable. Credible commitment requires that a critical mass of elites in the private, government, and political realms support and voluntarily comply with the rules of the game, with compliance underpinned by credible mechanisms of enforcement. The contested reality of South Africa’s fraying elite bargain falls far short of any robust commitment to such mutually agreed rules. Muddling through also does not address the question of how nonelites are to be more effectively incorporated into the economic and political order. There is no shared understanding among elites—let alone agreement or acquiescence among nonelites—as to how such incorporation is to be achieved.

In 1994, the halo effect of a negotiated constitutional democracy, the sweeping electoral victory of the ANC, and the promise of “a better life for all” were sufficient to paper over the limitations in both the elite bargain and the promises made to nonelites and to usher in a season of hope. Yet by the mid-2010s, the limitations of the aspirational promises of 1994 had become all too evident. With Ramaphosa’s accession to the presidency, relief at disaster averted might perhaps have provided space in the short term for a narrow growth-focused approach. It might have been good enough to paper over, for a few years, the widening cracks in the country’s political, institutional, and social foundations. But it would only be a matter of time before a deepening downward spiral reasserted itself—with the COVID-19 crisis adding to the risk of decline.

The second scenario in table 12 is ethnopopulism. In this scenario, opportunistic political entrepreneurs leverage the elite/nonelite relationship as a weapon in interelite contestation. In the resulting economic crash, the losses fall disproportionately on the preexisting predominantly white elites. The poor and vulnerable also lose as a consequence of declining fiscal revenues, which have been key to supporting incomes in the bottom half of the distribution. However—and this is the economic basis for the strategy’s attractiveness among ethnopopulist political entrepreneurs—in such a scenario, it is plausible that black elites could reap large real gains in income. Depending on the details of implementation, the effect on incomes of the black “insider” middle class could be quite modest. (This ethnopopulist scenario is a fairly good approximation of what has transpired in Zimbabwe since the early 2000s.)

The right-hand column of table 12 summarizes a third, more optimistic scenario—one that renews a platform of cooperation with a commitment to shared benefits. The outcomes associated with this scenario include the following:

  • High growth, with real income rising by 40 percent over the period.
  • Substantial declines in absolute poverty, with a near-doubling of real incomes for the poorest 50–60 percent of the population.
  • Very modest gains in the real incomes of white South Africans—marginally above zero as a way of encouraging cooperation but proportionately less than for any other subgroups, implying a substantial decline (from 37 percent to 28 percent) in their share of total income.
  • Major gains in upward mobility on the part of the approximately one-quarter of the population who are at the cusp of middle-class status or struggle to hold onto it.

The centerpiece of a next-generation inclusive development strategy for South Africa is a credible promise of upward mobility. Such a promise would, of course, be crucial for those already on the vulnerable cusp of the middle class. Beyond that, a credible promise of upward mobility would offer a vision of hope and possibility for better lives across society as a whole. In its absence, anger and despair could metastasize. But if promises of upward mobility were seen to be real (and if a minimum threshold of basic needs was met), then perceptions could be renewed as to the legitimacy of the social and economic order.

A crippling weakness of South Africa’s development trajectory over the past quarter-century has been a failure to address the distributional cliff described earlier. In the first fifteen or so years of democracy, the elimination of racial barriers and the country’s accelerating growth concealed the reality that the economy was not geared to support broad-based upward mobility. However, once the low-hanging fruit of the opportunity opened up by the end of apartheid’s racial privileges was gone, the limited economic prospects of those outside the elite became evident.

It is easier, of course, to write down the optimistic arithmetic of an inclusive development strategy than it is to spell out what such a strategy might entail. As described here, a strategy capable of delivering on a promise of upward mobility would have four interdependent aspects:

  • Improved public governance as key to renewing citizens’ perceptions of the legitimacy of state institutions. In the absence of improvements in public sector governance, none of the tasks set for the public sector would be achievable, nor would the cooperation necessary for compliance with a higher tax burden be forthcoming.
  • Bold initiatives to strengthen the supply side of upward mobility. Along with expanded investment in learning (including in early childhood and postsecondary education), such efforts could include a capital endowment grant, payable at adulthood, as a way to address historical legacies of massively unequal wealth and to ease the constraints on upward mobility that can come from precarious financial situations.
  • Bold initiatives to strengthen the demand side of upward mobility, with job creation as central, by private and public actors and by small as well as large firms.
  • Growth-compatible redress and revenue-raising, which are key to providing the additional revenue needed to deliver on the expanded public initiatives and to strengthen the platform of legitimacy (and fairness) of the economic order. This approach could be a preferred substitute for BEE and its unintended corrosive consequences. Relative to comparators, South Africa’s taxation of assets is not disproportionately high.

Appendix B provides some additional detail on each of these aspects.

Some Broader Implications

South Africa’s mid-1990s political settlement reconfigured radically the space occupied by political elites and the ways in which they interacted, but its inherited extreme inequalities were addressed only on the margins. The country’s experience over the subsequent quarter-century thus may offer useful insights for the many countries struggling to maintain a positive social, political, and economic trajectory in the face of a declining tolerance for high or rising inequality. Four connected lessons, applicable beyond the South African case, are of broad interest.

First, the trajectory of change is a knife-edge. Subsequent to the initial settlement, the country’s trajectory of change has been shaped by interactions among ideas, institutions, and distributional concerns. There is the potential to set in motion virtuous circles of positive interactions. But at the same time, there is a substantial risk that unaddressed distributional imbalances can result in a reversal, setting in motion a cumulative downward spiral of decline.

South Africa embarked on its democratic journey with a bold political and institutional realignment and a strong platform of formal institutions but also with massive misalignments between the political dimension and the economic/distributional dimension. Resources remained disproportionately concentrated within established white elite communities. For a while, though, a potent ideational platform seemed sufficient to paper over the continuing distributional disconnects.

Second, ideas matter. A hopeful vision of change, when combined with a “good enough” response to distributional concerns, can be sufficient to launch a positive trajectory.

The South African experience suggests that a positive trajectory can be sustained as long as influential interests and citizens more broadly remain hopeful that preexisting legacies of inequality and injustice are being addressed. In retrospect, the promise to turn around these legacies in South Africa was more aspirational than realistic. Even so, at least for a while, the promise to address them was sufficient to set in motion a season of hope. Not everything needs to be addressed all at once.

However, positive narratives can also be lost. Within a decade, the cracks in the foundations of South Africa’s democracy began to be evident. Over the course of the 2010s, things seemed to be falling apart, at an accelerating pace.

Third, in different ways, both ideas and institutions can be shields against adversity—but only up to a point. Hopeful ideas can evoke positive agency and help mobilize for collective action. Institutions can function as shock absorbers. However, both need reinforcement, including ongoing attention to festering imbalances.

South Africa’s reversal mirrors the ideational turn depicted in Hirschman’s tunnel effect: an initial season of hope; rising disillusion in the face of continuing (and in some aspect growing) imbalances; and then an ideational shift, an eruption of anger. In the South African case, pressure on institutions came first and the ideational shift came later. Plausibly, in other contexts, a turn to anger could precede, indeed catalyze, a subsequent attack on institutions.

Once a positive trajectory has been lost, what does it take to achieve renewal? As noted earlier, while Ramaphosa began the task of rebuilding the damaged state early in his tenure, he chose an incremental and legalistic path to reform. Unlike, for example, U.S. president Theodore Roosevelt’s assault on corruption and abuse of power in the United States nearly 120 years ago, Ramaphosa did not choose to ignore party bosses and rely on public support for reforms.128 He chose instead an incremental path, relying largely on hastily repaired prosecutorial authorities to clean up the state. The ANC remains divided, with the formerly Zuma-led faction led by Ace Magashule, who has been able to control the party machinery through his position as general secretary. Resistance remains strong in many regions, cities, and towns, and Ramaphosa’s hold on power remains conditional. As seen in figure 6, in the two (pre-COVID-19) years subsequent to Ramaphosa becoming president, there was no economic growth recovery. Table 13 summarizes what has changed in South Africa’s political economy during 2020, the coronavirus year.

Muddling through, without addressing festering imbalances, can fuel disillusion and cannot continue indefinitely. Ramaphosa has sought to govern in a high-minded way and tried to rekindle the aspirations of the 1990s, but more is needed than nostalgia for past promises that may once have offered inspiration but have passed their sell-by date and are broadly seen by citizens as inadequate to address contemporary challenges.

Fourth, initiating a new cycle of renewal requires a set of ideas that address in a “good enough” way the imbalances that had resulted in derailment—including a readiness to challenge incumbent interests that had blocked change.

It is imperative that South Africa strengthen the governance and effectiveness of the public sector. The country needs to raise the revenues to finance a credible commitment to upward mobility and to confront the rigidities that block mobility. It needs to renew a sense of fairness and of opportunity. Leadership needs to take the risk of mobilizing new coalitions, capable of trumping the vested interests that stymie inclusive change. Can South Africa’s leadership—and can leadership in other countries, where a similar sense of disillusion has taken hold—summon the necessary boldness to rise to this challenge?

Appendix A: Sector-Level Trends in Public Service Provision

This appendix reports some empirical patterns with regard to trends in the provision of four sets of public services: education, healthcare, local road maintenance, and water and sanitation. The first two are implemented by provincial governments; the latter two are local government responsibilities.

In terms of performance, South Africa’s education system is an extreme negative outlier globally in measured learning outcomes relative to expectations. However, using a global standardized measure, the Trends in International Mathematics and Science Study (TIMSS), Table A1 shows that South African students have made major gains in performance since 2003 across almost all provinces. Indeed, the gains achieved (from an extraordinarily low baseline) were unprecedentedly rapid.

Incremental gains in other sectors of provincial service provision must also be viewed in a wider context of long-term net underperformance, resulting from a disproportionate failure to expand the frontline highly skilled segment of the public sector workforce at the expense of a burgeoning managerial class. This is evident in only a modest growth in the nurse-to-population ratio in South Africa between 1998 and 2018, including a stable rise between 2009 and 2017.135 Moreover, any net benefit for the nursing workforce in the public sector would have been diminished by the doubling of nurses working in private healthcare between 1994 and 2014. Gains from this incremental rise in the nursing workforce are also offset by South Africa’s considerable disease burden. Despite impressive rates of antiretroviral rollouts, South Africa’s improvement in the nurse-to-population ratio is marginal at best compared with other MICs, taking life expectancy at birth into account.

Figure 8 summarizes trends in performance across two sets of key local government functions: local road maintenance and water and sanitation services. The data on road maintenance, from the perspective of citizens, remain generally negative and highly variable across provinces. Nearly 57 percent of respondents to a 2004 Afrobarometer survey perceived the quality of local road maintenance by municipalities to be “very bad/fairly bad.”136 Eleven years later, the prevailing view remained negative but moderated somewhat to 53 percent. At the provincial level, in two provinces local governments were viewed as having significantly improved road maintenance between 2008 and 2015, and in two provinces local road maintenance declined; there was no trend evident in the majority (five) of provinces. According to the South African National Roads Agency, a state entity created to manage the country’s national road network, it has assumed responsibility for some provincial and municipal roads, including in the Eastern Cape, Mpumalanga, Limpopo, and North West provinces.137 The latter are among the provinces that showed either a decline or no trend in improved public perceptions of local road maintenance, underscoring the level of deterioration in municipal management.

The water and sanitation services data in figure 9 show a consistent expansion in the number of billed household units being provided with these services. Despite steady gains in expanding access or connectivity, other perspectives reveal a less sanguine picture of municipal performance. The nationally representative Afrobarometer survey found an increase in the percentage of respondents saying that the government was handling the delivery of water and sanitation “very/fairly badly,” from 37 percent in 2002/2003 to 53 percent by 2018.138

One explanation for the discrepancy between aggregate expansion in the municipal delivery of water and sanitation and the public opinion response is that broader measures of household access encompass settlement patterns (as well as multiple unit connections, a byproduct of South Africa’s long-standing crisis of urban housing). They show virtually no long-term change in urban access, which hovers just below 90 percent, and only a marginal rise for rural access, which remains well below urban areas.139 Moreover, the rural subsample of respondents from the Afrobarometer survey was consistently more critical of water and sanitation delivery. Finally, despite increasing (audited) expenditure on repairs and maintenance of water and sanitation infrastructure between 2008/2009 and 2016/2017—an important indicator for gauging the qualitative impact of improved access—employee-related costs as a proportion of overall municipal spending on infrastructure repairs and maintenance were highly uneven between 2007/2008 and 2018/2019 and consistently were outpaced by contracted services.140 These facts intersect with two problematic dynamics discussed earlier: the significant growth in the managerial segment of the local government workforce at the expense of nonmanagerial staff (which includes technical officials involved in infrastructure maintenance) and the heightened political contestation over an increasingly large contracted-out budget for public service delivery in a context of increasing financial mismanagement.

Appendix B: Fostering Inclusive Development

Four elements are central to a strategy capable of credibly fostering inclusive development, focused especially on proactive support for upward mobility: (1) improved public sector governance, (2) a strengthened supply side of upward mobility, (3) a strengthened demand side of upward mobility, and (4) growth-compatible redress.

Improved public sector governance. This is a foundational concern; without improved governance, the public sector would not be able to deliver on any proposed development strategy, nor would the public be willing to comply with the higher tax burden required to support these improvements. One useful innovation would be to move beyond the hierarchical (“government should deliver”) perspective on the public sector common to most South Africans and instead focus on “active citizenship” in governance processes. This revised approach could be key to renewing citizens’ perceptions of the legitimacy of state institutions. As South Africa’s 2012 National Development Plan puts it, “To build an inclusive nation, the country needs to find ways to promote a positive cycle, where an effective state, inspirational leadership across all levels of society, and active citizens reinforce and strengthen each other.”141

A strengthened supply side for upward mobility. This consideration is principally about filling key gaps in South Africa’s currently uneven ladders of opportunity. Improved education and financial support can help to create an environment where upward mobility is possible.

The quality of basic education is an obvious area of continuing focus for South Africa. Schooling resources are more or less in line with what is available in other MICs, and any criticism of South African students’ performance must keep the historical legacies of apartheid in mind. Even so, governance remains a critical challenge. Careful theoretical and empirical analysis of data from countries around the world has indicated that expanding investment in high-quality early childhood development (ECD) may have the highest return of all development investments. However, a much lower proportion of South African children under the age of six are in ECD programs compared with other MICs.

Postsecondary-school adolescents and young adults also need expanded investment in higher education and vocational training. The proportion of South Africans between ages fifteen and twenty-four who are in tertiary or vocational education is below that of comparator countries. In 2017, in response to the “Fees Must Fall” student movement, South Africa made some steps to address these shortfalls, but there is much more to be done in terms of financial and quality improvements. Innovation is needed.

A few countries have considered direct financial supports such as “capital endowments” and “baby bonds” (payable at, say, age eighteen) as ways to simultaneously address historical legacies of massively unequal wealth and ease the constraints on upward mobility that can come from being one paycheck away from disaster. As Anthony Atkinson put it in his 2014 book Inequality: What Can Be Done?, “There should be a capital endowment paid to all at adulthood. . . . A case can be made for imposing a degree of ‘prudence’ on its use. . . . Possible permitted uses could include education or training . . . down payments on houses or flats, or the establishment of a small business.”142

A strengthened demand side for upward mobility. This is an arena with which South African policymakers and academics have long been preoccupied.143 The demand side of upward mobility is overwhelmingly about job creation—by private as well as public actors, in the informal and formal sectors. Survivalist informal sector jobs, and temporary employment of the kind provided by South Africa’s Expanded Public Works Programme, barely qualify as opportunities for upward mobility. A major push for infrastructure could be a potent and productivity-enhancing way of creating jobs and is likely to require major additional public resources.

South Africa also has been experimenting with a variety of modalities for subsidizing wages, as a way of helping to turn borderline-subsistence employment into a viable pathway to upward mobility. This should be expanded. Furthermore, in the context of a shifting global economy, especially including a changing role for China in global supply chains, South Africa may have an opportunity to insert itself in the global economy in a new way. Efforts should be made to accelerate implementation of export zones, including by streamlining regulation within the zones, in coastal regions where there is good port infrastructure and high unemployment.

Growth-compatible redress and revenue-raising. Additional resources will be needed to follow through on the bold initiatives to accelerate upward mobility. Especially in the South African context, increased taxation is about more than meeting fiscal needs. In the face of South Africa’s legacy of extreme historical injustice, with the honeymoon of the country’s “rainbow miracle” long over and with growth having slowed, the continued concentration of income and wealth at the top of the distribution fuels a sense of unfairness, of salt on a still-festering wound. For similar historical reasons, redress is likely to have to include continuing attention to the difficult challenges of land redistribution, an area where the Ramaphosa government has been moving forward deliberately, and with due sensitivity to the hazards.

Strikingly, notwithstanding almost three decades of effort to reverse South Africa’s legacy of extreme historical injustice, relative to other middle- and high-income countries, as of 2018 the country’s taxation levels were in a middle range. At 38 percent, revenue collection as a percentage of GDP was above the United States (31 percent), similar to Australia (37 percent), and below Brazil (41 percent), Germany (46 percent), and Sweden (51 percent). South Africa’s marginal tax rates currently are not disproportionately high, and taxes on wealth are relatively low. In its 2018 budget, South Africa increased its maximum inheritance tax on wealth from 20 percent to 25 percent. By contrast, from the mid-1930s until the early 2000s, the top inheritance tax rate in the United States consistently was over 50 percent. South Africa’s property taxes currently are below those in the United States. Evidently, South Africa has ample room for revenue-raising—all the more so given the country’s history of wealth accumulation and exploitation.

Although the hazards of increased taxation on assets are obvious, there could also be a win-win dimension politically. In the first two decades of democracy, pro-poor fiscal reallocation and BEE efforts were the principal ways of addressing South Africa’s historical legacy of injustice. The combination of heightened attention to upward mobility along lines laid out above, plus increased taxation to fund opportunities, add up to a major new, sufficient, and growth-compatible set of initiatives to achieve redress for the country’s historical legacy—including perhaps as a preferred substitute for BEE and its unintentionally corrosive consequences.


ANC African National Congress
ARC African Rainbow Capital
ARM African Rainbow Minerals
BB-BEE Broad-based Black Economic Empowerment
BEE Black Economic Empowerment (programme)
CEO chief executive officer
COSATU Congress of South African Trade Unions
CSR China South Rail
ECD early childhood development
EFF Economic Freedom Fighters
EWC expropriation without compensation
GDP gross domestic product
JSE Johannesburg Stock Exchange
MIC middle-income country
MPRDA Mineral and Petroleum Resources Development Act
NAIL New African Investments Limited
OECD Organisation for Economic Co-operation and Development
RDP Reconstruction and Development Programme
SOE state-owned enterprise
TIMSS Trends in Mathematics and Science Survey
UC University of Cape Town
WGI Worldwide Governance Indicator
WMC white monopoly capital


This paper is a product of a research collaboration between the Carnegie Endowment for International Peace and the Nelson Mandela School of Public Governance at the University of Cape Town, focused on governance-development interactions in Africa. The authors would like to thank Carnegie for its financial support; Tom Carothers, Faizel Ismail, and Carlos Lopes for their useful inputs; Kevin Foster and Thomas Van Huyssteen for research assistance; and Lori Merritt and Amy Mellon for their excellent copy editing and design of the paper.

About the Authors

Brian Levy teaches at the School of Advanced International Studies, Johns Hopkins University. He was the founding academic director of the Nelson Mandela School of Public Governance at the University of Cape Town. He worked at the World Bank from 1989 to 2012, including as head of the secretariat responsible for the implementation of the World Bank Group’s governance and anticorruption strategy. He has published widely on the interactions among institutions, political economy, and development policy, including as the author of Working With the Grain (Oxford University Press, 2014) and as the lead editor and a contributing author of The Governance and Politics of Basic Education: A Tale of Two South African Provinces (Oxford University Press, 2018). He completed his PhD in economics at Harvard University in 1983.

Alan Hirsch is a professor at the Nelson Mandela School of Public Governance at the University of Cape Town (UCT) and was the school’s founding director from 2011 to 2019. Born in Cape Town and educated at UCT, the University of Witwatersrand, and Columbia, he first taught at UCT and then joined the South Africa Department of Trade and Industry in 1995. From 2002 to 2011, he managed economic policy in the South African presidency, represented the president in the G20, and was co-chair of the G20 Development Working Group. Hirsch serves on the Board of the European Centre for Development Policy Management, the International Advisory Board of the New Development Bank, and President Cyril Ramaphosa’s Economic Advisory Council. He was a visiting scholar at the Harvard Business School, a regular visiting professor at Maastricht University, a research director for the International Growth Centre in Zambia, a member of the OECD Inclusive Growth Advisory Panel, and a Bradlow fellow at the South Africa Institute for International Affairs. His published work includes Season of Hope: Economic Reform Under Mandela and Mbeki (University of KwaZulu-Natal Press, 2005) and The Oxford Companion to the Economics of South Africa (Oxford University Press, 2015).

Vinothan Naidoo teaches public policy and public administration at the University of Cape Town. He has a multi-disciplinary background with degrees in history, political studies, and public administration and has published on a variety of topics associated with South Africa’s public service. These topics include public sector reform, political-administrative relations, anticorruption, policy implementation, monitoring and evaluation, and intergovernmental relations. He has participated in many applied research projects for South African and international organizations over the course of a twenty-year career spanning private sector consultancy, public sector think tank, and university roles. He completed his PhD in public administration at the University of Cape Town in 2008.

Musa Nxele teaches political economy of development at the Nelson Mandela School of Public Governance at the University of Cape Town. He is doing a joint PhD in economics and development policy and practice at Université de Paris 1 Panthéon-Sorbonne and at the University of Cape Town, respectively. He also holds masters degrees in economic development and globalization from both universities. His research focuses on the political economy of industrial development and investment. Before moving to tertiary education teaching in 2017, Nxele worked in the private sector in the fields of investment banking and industrial development consulting. He dedicates his life to the transformation of people and societies, driven by hope and the incredible story of his hometown, Soweto.


1 Larry Diamond, The Spirit of Democracy: The Struggle to Build Free Democracies Throughout the World (New York: Times Books, 2008), 372. The data are for countries with populations above 1 million.

2 Larry Diamond, Ill Winds: Saving Democracy From Russian Rage, Chinese Ambition, and American Complacency (New York: Penguin Press, 2019), 54.

3 In this paper, the lowercasing of both “white” and “black” reflects current style preferences for racial terms within South Africa.

4 Albert Hirschman and Michael Rothschild, “The Changing Tolerance for Income Inequality in the Course of Economic Development,” Quarterly Journal of Economics 87, no. 4 (1973): 545.

5 Ibid., 552.

6 Albert Hirschman, “The Turn to Authoritarianism in Latin America and the Search for Its Economic Determinants,” in David Collier, ed., The New Authoritarianism in Latin America (Princeton, NJ: Princeton University Press, 1980), 88.

7 John Maynard Keynes, “The General Theory of Employment,” Quarterly Journal of Economics 51, no. 2 (1937): 209.

8 Ibid., 210.

9 See, for example, Dani Rodrik, “When Ideas Trump Interests: Preferences, Worldviews, and Policy Innovations,” Journal of Economic Perspectives 28, no. 1 (2014): 189–208; Sharun Mukand and Dani Rodrik, “The Political Economy of Ideas: On Ideas Versus Interests in Policymaking,” NBER Working Paper Series, no. 24467 (2018): 1–40; Douglass North, Understanding the Process of Economic Change (Princeton, NJ: Princeton University Press, 2005); and World Bank, World Development Report 2015: Mind, Society, and Behavior (Washington DC: World Bank, 2015).

10 Rodrik, “When Ideas Trump Interests,” 189–190.

11 African National Congress, The Reconstruction and Development Programme: A Policy Framework (Johannesburg: Umayamo Publishers, 1994).

12 Douglass North, Institutions, Institutional Change and Economic Performance (New York: Cambridge University Press, 1991), 5.

13 Italy was in the 78th percentile, and France was in the 86th percentile.

14 World Bank, Worldwide Governance Indicators,

15 Fernando Bizzarro, Allen Hicken, and Darin Self, “The V-Dem Party Institutionalization Index: A New Global Indicator (1900–2015) (May 1, 2017),” V-Dem Working Paper 2017, 48, or

16 Tim Kelsall, Nicolai Schulz, William Ferguson, Matthias vom Hau, Sam Hickey, and Brian Levy, Understanding Development: The Promise of Political Settlements (Oxford: Oxford University Press, 2021, forthcoming); and Tim Kelsall and Matthias vom Hau, “Beyond Institutions: Political Settlements Analysis and Development,” IBEI Working Paper 56, University of Barcelona, 2020, Influential contributions include Douglass North, John Wallis, and Barry Weingast, Violence and Social Orders (New York: Cambridge University Press, 2009); Douglass North, John Wallis, Barry Weingast, and Stephen Webb (eds.), In the Shadow of Violence—the Problem of Development in Limited Access Order Societies (New York: Cambridge University Press, 2013); Mushtaq Khan, “Political Settlements and the Governance of Growth-Enhancing Institutions,” 2010,; Daron Acemoglu and James A. Robinson, Why Nations Fail (New York: Crown Books, 2012); William Ferguson, The Political Economy of Collective Action, Inequality, and Development (Palo Alto, CA: Stanford University Press, 2020); World Bank, World Development Report 2017: Governance and the Law (Washington, DC: World Bank, 2017); and Brian Levy, Working With the Grain: Integrating Governance and Growth in Development Strategies (New York: Oxford University Press, 2014).

17 By way of comparison, the equivalent 2015 figure for the United States was 77 percent.

18 Pádraig Carmody (2002), quoted in Martin Plaut and Paul Holden, Who Rules South Africa? (London: Biteback Publishing, 2012), 256–257.

19 For a detailed analysis, see Brian Levy, Alan Hirsch, and Ingrid Woolard, “Governance and Inequality: Benchmarking and Interpreting South Africa’s Evolving Political Settlement,” Effective States and Inclusive Development, no.51 (2015).

20 For a definitive analysis of the roots of inequality in South Africa’s apartheid economy, see Jeremy Seekings and Nicoli Nattrass, Class, Race, and Inequality in South Africa (London: Yale University Press, 2005). Jeremy Seekings and Nicoli Nattrass explore the evolving balance between continuity and change in South Africa’s class structure in the first two decades of democracy; see Jeremy Seekings and Nicoli Nattrass, Policy, Politics and Poverty in South Africa (London: Palgrave Macmillan, 2015).

21 Note that in Hirschman’s (1973) framing, asymmetries between political and economic power are not necessarily evident at the outset of the cycle but are fueled by a subsequent period of rapid unbalanced growth. In South Africa, by contrast, stark asymmetries were present from the start of the democratic political settlement.

22 Alan Hirsch, Season of Hope: Economic Reform Under Mandela and Mbeki (Pietermaritzburg: University of KwaZulu-Natal Press, 2005, 1st Edition).

23 Hirsch, Season of Hope: Economic Reform Under Mandela and Mbeki, 1.

24 Oliver E. Williamson, “The New Institutional Economics: Taking Stock, Looking Ahead,” Journal of Economic Literature 38, no.3 (2000): 595–613.

25 Elinor Ostrom, “Collective Action and the Evolution of Social Norms,” Journal of Economic Perspectives 14, no. 3 (2000): 137–158.

26 Policy Co-ordination and Advisory Services, “Towards a 10 Year Review: Synthesis Report on Implementation of Government Programmes,” The Presidency, Republic of South Africa, October 2003,

27 Alan Hirsch and Brian Levy, “Elaborate Scaffolding, Weak Foundations: Business–Government Relations and Economic Reform in Democratic South Africa,” Effective States and Inclusive Development, no.105 (2018): 111.

28 North et al., Violence and Social Orders; Merilee S. Grindle, “Good Governance, R.I.P.: A Critique and an Alternative,” Governance 30, no.1 (2016),; and Lant Pritchett, Kunal Sen, and Eric Werker, Deals and Development: The Political Dynamics of Growth Episodes (Oxford: Oxford University Press 2018).

29 Paul Taylor, “A Black Success Story Brewing in South Africa,” Washington Post, July 21, 1993,

30 Hirsch, Season of Hope: Economic Reform Under Mandela and Mbeki.

31 Ewan Sutherland, “MTN: A South African Mobile Telecommunications Group in Africa and Asia,” South African Journal for Communication Theory and Research 41, no.4 (2015): 471–505.

32 Securities and Exchange Commission, “Annual Report Pursuant to Section 13 of the Securities Exchange Act of 1934, for the Fiscal Year Ended June 30, 2004,” Form 20-F, Commission File Number 001-31545, 2004,

33 “AngloGold Announcement Expected on Tuesday,” Mail & Guardian, November 25, 1997.

34 Susan Adams, “The Prince of Mines,” Forbes, March 6, 2008,

35 “Motsepe Says He Lost R500m on Modikwa Platinum Mine,” IOL, October 10, 2004,

36 “Empowerment Hits Gold,” IOL, May 17, 2002,

37 Adams, “The Prince of Mines.”

38 “Annual Report 2008,” African Rainbow Minerals, 2008,

39 “Sustainability Report 2011,” African Rainbow Minerals, 2011,

40 Dewald Van Rensburg, “JSE Listing—a Little Motsepe Goes a Long Way,” City Press, August 27, 2017,

41 Matt Andrews, “Is BEE a South African Growth Catalyst? (Or Could It Be . . .),” produced for the South African Government as part of the International Growth Panel initiative, September 2007,

42 African National Congress, “National Empowerment Policy,” 50th National Conference of the African National Congress, Mafikeng, South Africa, 1997; and African Communist, “Inequality Is No Longer Just About Race,” Politicsweb, February 16, 2017,

43 Roger Tangri and Roger Southall, “The Politics of Black Economic Empowerment in South Africa,” Journal of Southern African Studies 34, no.3 (2008): 699–716.

44 For details, see Hirsch, Season of Hope: Economic Reform Under Mandela and Mbeki; Tangri and Southall, “The Politics of Black Economic Empowerment in South Africa”; Jabulile Mpanza, “B-BBEE and Its Impact on the South African Construction Industry,” University of Cape Town, 2016,; Plaut and Holden, Who Rules South Africa?; and Ross Harvey, “Between a Rock and a Hard Place: State-Business Relations in the Mining Sector,” New South Africa Review 5: Beyond Marikana, Chapter 5, Witwatersrand University Press, 2015.

45 Hirsch, Season of Hope: Economic Reform Under Mandela and Mbeki, 221. The chair of the commission was senior ANC figure and business leader Cyril Ramaphosa, who later would become president.

46 The quote, the information from Citigroup, and additional details of the challenges of implementing the MPRDA can be found in Harvey, “Between a Rock and a Hard Place: State-Business Relations in the Mining Sector.”

47 “South Africa High Court Ruling: Once Empowered, Always Empowered,” Mining Review Africa, April 5, 2018,

48 Giada Masina, Allan Reid, and Ben Cripps, “Minister Publishes Draft Amendments to MPRDA Regulations,” Mining and Minerals Alert, Cliffe Dekker Hofmeyr, December 4, 2019,

49 Almost all of the core negotiators who were not already senior industry executives subsequently were appointed as nonexecutive directors on the boards of leading construction companies.

50 Mpanza, “B-BBEE and Its Impact on the South African Construction Industry.”

51 “South African Building Firms Fined $140m for Price-Fixing,” Global Construction Review, June 26, 2013,

52 Open Secrets & SWI, “Arms Deal: The BAE Corruption Bombshell,” Daily Maverick, July 29, 2020,

53 North Gauteng High Court, “In the High Court of South Africa/Es (North Gauteng High Court, Pretoria),” CASE no. 2263/06, July 30, 2009,

54 Public Protector South Africa, “Report on an Investigation Into an Allegation of Improper Conduct by the Former Chairperson of the Board of Directors of Eskom Holdings Limited, Mr. V. Moosa, Relating to the Awarding of a Contract,” Report no. 30 of 2008/9, Public Protector South Africa, 2009,

55 Simpiwe Piliso, “How ANC Bigwigs Jostled for Telkom,” Sunday Times, October 15, 2006; and Simpiwe Piliso, “ANC Defends State Role in Telkom Deal,” Sunday Times, October 15, 2006.

56 “Vodacom: ANC’s Elephant Hunt,” Mail & Guardian, May 25, 2009,

57 Zweli Mokgata, “Vodacom Goes Ahead With Empowerment Deal,” Sowetan Live, April 1, 2008,

58 Hirsch and Levy, “Elaborate Scaffolding, Weak Foundations,” 26.

59 Carnegie Corporation of New York, “Carnegie Corporation in South Africa: A Difficult Past Leads to a Commitment to Change,” Carnegie Results, Winter 2004.

60 Given South Africa’s massive inequality, the mean per capita income is in the 20th to 25th percentile of the distribution, well above the median. South Africa also is an MIC, with a per capita income about one-fifth that of high-income counterparts. Its mean per capita income is well below what would be perceived, in high-income countries, to be middle class.

61 In constant 2010 dollars. In current 1994 dollars (at prevailing exchange rates), 1995 per capita income was closer to $4,000.

62 Measuring middle-class expansion in this way yields a magnitude similar to the estimates of Roger Southall; see Roger Southall, “The Black Middle Class and Democracy in South Africa,” Journal of Modern African Studies 52, no. 4 (2014): 647–670.

63 This was referred to as a “consistent upward drift in the distribution of personnel across the grades” in National Treasury, “Medium Term Budget Policy Statement 2019.”

64 National Treasury, “Medium Term Budget Policy Statement 2019.”

65 The salaries of highly skilled frontline public sector workers are typically between levels 6 and 8—so the increase in the share of employment in levels 9–11 generally is consistent with an enlargement of the managerial ranks.

66 Department of Public Service and Administration; South African government’s personnel salary database; and Public Service Commission, “State of the Public Service 2006: Assessing the Capacity of the State to Deliver,” Pretoria, 2006. The dramatic increase in the number of officials employed at salary levels 9–12 (highly skilled supervision) between 2006 and 2009 is also evident in data presented in National Treasury, “Medium Term Budget Policy Statement 2019.”

67 National Treasury, Budget Review, various years; Medium Term Budget Policy Statement, 2018 (compensation data); and OECD, “OECD Economic Surveys: South Africa,” 2020.

68 International Monetary Fund, International Financial Statistics online database, various years.

69 As reported in OECD, “OECD Economic Surveys: South Africa,” 2020.

70 National Treasury, “Medium Term Budget Policy Statement 2019,” October 30, 2019. Data showed a continuation of this trend into 2017/18. The OECD also cited wage increases rather than an increase in employment as the primary driver of government spending; see OECD, “OECD Economic Surveys: South Africa.”

71 Robert Cameron, “New Public Management Reforms in the South African Public Service: 1999–2009,” Journal of Public Administration 44, no. 4.1 (2009): 910–942; and V. Naidoo, “Changing Conceptions of Public ‘Management’ and Public Sector Reform in South Africa,” International Public Management Review 16, no. 1 (2015): 23–42.

72 Joel Pearson and Thatshisiwe Ndlovu, “‘I’m No Longer Applying Straight’: Human Resource Practices in State Institutions: Perspectives of Public Servants,” Working Paper 5, Public Affairs Research Institute, 2018; and Public Service Commission, “Recruitment, Retention, Career Pathing and Utilization of Senior Management Service Members’ Expertise and Skills in the Public Service,” Public Service Commission, 2017.

73 Between 1996 and 2017, the WGI measures for Brazil and Mexico also declined; those for Thailand and Turkey remained broadly stable.

74 Parliamentary Monitoring Group, “Public Service Commission: Building a Professional Public Service; Management of Service Terminations and Pensions Pay-outs; Public Service Regulations 2015,” transcript of a meeting of the Public Service and Administration, Performance Monitoring and Evaluation Portfolio Committee, and National Assembly, November 9, 2016,

75 Ryan Brunette, “Position Paper on Appointment and Removal in the Public Service and Municipalities,” Position Papers on State Reform, Public Affairs Research Institute, April 2020,, 12.

76 Auditor-General South Africa, “Auditor-General Kimi Makwetu Announces National and Provincial Audit Results,” November 21, 2018,

77 Brunette, “Position Paper on Appointment and Removal in the Public Service and Municipalities,” 15.

78 V. Naidoo, “Transitional Politics and Machinery of Government Change in South Africa,” Journal of Southern African Studies 45, no. 3 (2019): 575–595.

79 G. Van Onselen, “Political Musical Chairs: Turnover in the National Executive and Administration Since 2009,” South African Institute of Race Relations, September 27, 2017.

80 Brunette, “Position Paper on Appointment and Removal in the Public Service and Municipalities.” The phenomenon of “state capture,” referring to corrupt relationships between politicians, senior civil servants, and business leaders, emerged under the Zuma presidency and stemmed from an investigation by South Africa’s Public Protector in 2016. It is currently the subject of a Presidential Commission of Inquiry.

81 National Treasury, “Medium Term Budget Policy Statement 2019—Based on Consolidated Expenditure for National and Provincial Departments, 2018/19”; and National Treasury, “Aggregated/Consolidated Municipal 2018 MTREF Information.”

82 Department of Basic Education, “Report of the Ministerial Task Team Appointed by Minister Angie Motshekga to Investigate Allegations Into the Selling of Posts of Educators by Members of Teacher Unions and Departmental Officials in Provincial Education Departments: Final Report 2016.”

83 Hirschman and Rothschild, “The Changing Tolerance for Income Inequality in the Course of Economic Development,” 551.

84 Hein Marais, South Africa Pushed to the Limit: The Political Economy of Change (London: Zed Books, 2011), 363–64, 375, 378–79.

85 Thabo Mbeki had been forced by the now-dominant Zuma faction of the ANC to resign from the country’s presidency in September 2008. Kgalema Motlanthe served as president from the time of Mbeki’s resignation until, subsequent to the 2009 election, Zuma acceded to office.

86 Brian Levy, “Commanding Heights: The Governance of State-Owned Enterprises in South Africa,” in Haroon Bhorat, et al., The Oxford Companion to the Economics of South Africa (Oxford: Oxford University Press, 2014).

87 World Bank, “Held by the Visible Hand: The Challenge of SOE Corporate Governance for Emerging Markets,” Washington, DC, May 1, 2006, 9, 13–14, 16, 21.

88 Republic of South Africa, Public Protector State of Capture, “Report no. 6 of 2016/17,” 2016,, 91.

89 Its ten authors comprise something of a “who’s who” of respected South African academics active in the public policy domain. An edited, book-length version was published in 2018, authored by Ivor Chipkin and Mark Swilling “with” the other authors. Other eminent academics among the ten include Haroon Bhorat, Mbongiseni Buthelezi, and Mzukisi Qobo. The quotes and references in the paper are from Ivor Chipkin and Mark Swilling, with H. Bhorat, M. Buthelezi, S. Duma, N. Prins, L. Mondi, P. Camaren, M. Qobo, H. Friedenstein, Shadow State: The Politics of State Capture (Johannesburg: Wits University Press, 2018).

90 Chipkin and Swilling, Shadow State, 31–32, 36–37.

91 The “Zupta” network is the focus here because it has been the most thoroughly investigated, has received the greatest attention in the media, and generally is perceived to have had the greatest impact. However, the evidence also suggests that—even setting aside individual acts of corruption—other networks also were operating, separately from the Gupta family. An important loose network, one which was not wholly disconnected from the Guptas, was the “premier league”: the premiers of the Free State (Ace Magashule), North West (Supra Mahumapelo), and Mpumalanga’s premier David Mabuza. Former Gauteng Premier and Cabinet Minister Tokyo Sexwale’s Mvelaphanda Group comprise another network.

92 Chipkin and Swilling, Shadow State, 25, 92. This paper reports most transaction values in South African Rand; between 2010 and 2017, the exchange rate has been in the range of R8 to R17 per dollar. In October 2016, Minister of Finance Pravin Gordhan filed an affidavit showing how R6.8 billion in “suspicious and unusual” transactions might have contributed to the decision by [South Africa’s Big Four] banks to close accounts associated with the Gupta family.

93 Other well-documented examples include the Passenger Rail Authority of South Africa, where the R51 billion procurement contract for new passenger coaches from a consortium led by a Spanish company resulted in the delivery of thirteen coaches configured in a way that made them unusable on the South African rail system (Chipkin and Swilling, Shadow State, 68–71); South African Airways, where a cost competitive purchase/lease of twenty Airbus A320s was abandoned at the last moment in order to proceed with an alternative arrangement that increased the cost to South Africa by R1.9 billion (Chipkin and Swilling, Shadow State, 67–68); and Denel, where it was alleged that during controversial negotiations with the Gupta-linked Denel Asia joint venture (Chipkin and Swilling, Shadow State, 88–90), the Guptas tried to sell Denel’s intellectual property to India while watering the company’s stake down by half.

94 Details of the Transnet saga are from Chipkin and Swilling’s Shadow State, plus additional research assistance conducted by Thomas Van Huyssteen under the supervision of Musa Nxele.

95 For the previous two years, the position of Transnet CEO had been filled in an acting capacity, following the resignation of Mbeki loyalist Maria Ramos.

96 Gama was in fact Zuma’s preferred choice of CEO in 2010/2011; indeed, the immediate cause of Barbara Hogan’s resignation was her refusal to agree to his appointment.

97 In parallel to Transnet’s locomotives deal, the Passenger Rail Authority of South Africa was engaged in a R123 billion procurement of a fleet of passenger trains. While the international partner in that deal turned out to be a Spanish company, in 2012 the CEO, Lucky Montana, sent a letter to his board reporting that “in numerous meetings, the Guptas, Duduzane Zuma, and their associates had allegedly pressured then Minister of Transport Ben Martins and Montana to favor [CSR]”; see Chipkin and Swilling, Shadow State, 68. The full deal for the remaining locomotives was split between four companies: CSR, Bombardier, General Electric, and China North Rail.

98 Chipkin and Swilling, Shadow State, 74. In 2016, VR Laser was centrally involved as a joint venture partner, in a subsequently aborted (after critical media exposure) initiative by South Africa’s defense contractor SOE, Denel, to restructure the way it did business in Asia. The deal between VR Laser and Denel gained momentum shortly after then minister of public enterprises Lynne Brown radically restructured the Denel board, and the new board quickly suspended Denel’s respected CEO, chief financial officer, and company secretary (Chipkin and Swilling, Shadow State, 89).

99 This material is drawn principally from Catrina Goudinho and Lauren Hermanus, “(Re)conceptualising State Capture—With a Case Study of South African Power Company Eskom,” prepared for the Public Affairs Research Institute’s research conference on State Capture and Its Aftermath: Building Responsiveness Through State Reform, Johannesburg, October 22–24, 2018; and Anton Eberhard and Catrina Goudinho, “Eskom Inquiry Reference Book: A Resource for Parliament’s Public Enterprises Inquiry, Civil Society, Journalists and Engaged Citizens,” State Capacity Research Project, Graduate School of Business, University of Cape Town and Public Affairs Research Institute, August 2017. The material also is informed by the analysis of the political economy of electricity sector decisionmaking in Nchimunya Hamukoma and Brian Levy, “When the Quest for Electricity Reform and the Need for Investment Collide: South Africa, 1998–2004,” in Nuno Gil, Anne Stafford, and Innocent Musonda (eds.), Duality by Design: The Global Race to Build Africa's Infrastructure (Cambridge: Cambridge University Press, 2019), 69–96.

100 Goudinho and Hermanus, “(Re)conceptualising State Capture,” 16.

101 Eberhard and Goudinho, “Eskom Inquiry Reference Book,” 12.

102 Details are from Chipkin and Swilling, Shadow State, 84–87; also Eberhard and Goudinho, “Eskom Inquiry Reference Book,” 14–19.

103 Eberhard and Goudinho, “Eskom Inquiry Reference Book,” 16.

104 Goudinho and Hermanus, “(Re)conceptualising State Capture,” 22.

105 Chipkin and Swilling, Shadow State, 86.

106 Perhaps not entirely coincidentally, one of the first major investments of the Gupta family, jointly with Duduzane Zuma, was the purchase in 2010 of the hitherto-Canadian-owned Uranium One mine.

107 Andrew Weiss and Eugene Rumer, “Nuclear Enrichment: Russia’s Ill-Fated Influence Campaign in South Africa,” Carnegie Endowment for International Peace, December 16, 2019,, 11.

108 Republic of South Africa, Public Protector State of Capture, “Report no. 6 of 2016/17.”

109 Sharun Mukand and Dani Rodrik, “The Divided Public Heart,” Aeon, June 6, 2018,

110 Hirschman and Rothschild, “The Changing Tolerance for Income Inequality in the Course of Economic Development,” 557.

111 Thabo Mbeki’s “I Am an African” May 1996 speech to parliament on the occasion of South Africa’s passage of a new constitution is another powerful statement, in a similar vein.

112 As of 2015, as reported in the University of Cape Town’s 2016 Teaching and Learning Report, the ethnic composition of the undergraduate student body at UCT was 27 percent “SA (South African) Black,” 14.4 percent “SA Coloured,” 7.4 percent “SA Indian,” 29 percent “SA White,” 14.5 percent “International,” and 7.6 percent “unknown.”

113 Rekgotsofetse Chikane, Breaking a Rainbow, Building a Nation: The Politics Behind #MustFall Movements (Johannesburg: Picador Africa, 2019), 99, 105, 110.

114 Mukand and Rodrik, “The Political Economy of Ideas,” 2.

115 Ibid.

116 Mukand and Rodrik, “The Divided Public Heart.” Per Mukand and Rodrik, “a meme is some combination of cues, narratives, symbols or indeed any choice of communication disseminated by the political entrepreneur, such that exposure to it either shifts views about how the world works or makes an identity salient. Memes are the concrete vehicle that channels ideas developed by the politician for the political marketplace.”

117 “Founding Manifesto of the Economic Freedom Fighters,” Politicsweb, July 25, 2013,; and “Malema: It’s an EFF Revolution,” News24, July 11, 2013,

118 “Alleged Full Report of Bell Pottinger’s Gupta PR Plan,” The Citizen, April 4, 2017,

119 Bernice Maune and Batandwa Malingo, “Zuma Tells Us Why ‘White Monopoly Capital Is a Fact and Not an Insult,’” The Citizen, November 13, 2017,

120 Andisiwe Makinana, “Land Reform: Zuma Moves for Expropriation With No Compensation,” City Press, March 31, 2017,

121 Marianne Merten, “#ANCdecides2017: Land Expropriation Without Compensation Makes Grand Entrance,” Daily Maverick, December 21, 2017,

122 One of multiple reasons for anemic job creation was the way in which politically influential unionized workers protected their positions through restrictive labor market policies.

123 Commission for Employment Equity, multiple years.

124 Richard Calland, The Zuma Years: South Africa’s Changing Face of Power (Cape Town: Random House Struik, 2013), chapter 11. “The judges” (loc 5903-6401) lays out the fears of many at the time of chief justice Mogoeng’s appointment.

125 Reporters Without Borders, “2020 World Press Freedom Index,”, accessed July 9, 2020.

126 Alexander Winning and Emma Rumney, “South African Police Issue Arrest Warrant for Top ANC Official Magashule,” Reuters World News, November 10, 2020,; and Pieter du Toit, “Top UN Lawyer, Ex-Scorpion Among Senior NPA Appointments Announced by Ramaphosa,” News 24, January 14, 2021,

127 Hirschman, “The Turn to Authoritarianism in Latin America and the Search for Its Economic Determinants,” 88.

128 Doris Kearns Goodwin, Leadership: In Turbulent Times (New York: Simon & Schuster, 2018).

129 Our World in Data, “South Africa: Coronavirus Country Profile,” accessed February 21, 2021,

130 Siphelele Dludla, “Government Debt Escalates to R3.7 Trillion, Says Reserve Bank,” IOL, December 16, 2020,

131 Ibid.

132 Bureau for Economic Research, “Beeld Consensus: Monthly Survey of Leading South African Economists Who Forecast Key Macroeconomic Variables,” Faculty of Economics and Management Sciences, University of Stellenbosch, South Africa, December 2020,

133 Ibid.

134 Republic of South Africa, “Electoral Commission on Results of By-Elections Held on 9 December 2020,” December 10, 2020,

135 From forty-one to forty-nine nurses per 10,000 population. See SalimahValiani, “Public Health Care Spending in South Africa and the Impact on Nurses: 25 Years of Democracy?” Agenda 33, no.4 (2019): 67–78.

136 This was in contrast to answering “fairly well/very well.”

137 SANRAL, “Frequently Asked Questions About SANRAL,” October 15, 2019,

138 Round 2 South Africa survey in 2002/2003, and round 7 in 2016/2018,

139 Statistics South Africa, General Household Survey, data for 2002–2017.

140 National Treasury local government expenditure tables, various years. Expenditure crested for water in 2014/2015 and displayed a more modest pattern of increase for sanitation. The gap was particularly large between 2010/2011 and 2013/2014 before reversing in favor of a marginal increase in employee-related costs compared to contracted services.

141 Brian Levy, Robert Cameron, Ursula Hoadley, and Vinothan Naidoo (eds.), The Politics and Governance of Basic Education: A Tale of Two South African Provinces (Oxford: Oxford University Press, 2018), 281. For details, see National Planning Commission, “National Development Plan, 2030: Our Future—Making It Work,” Republic of South Africa, The Presidency, 2012.

142 Anthony Atkinson, Inequality: What Can Be Done? (Cambridge: Harvard University Press, 2015), 170–172.

143 For some recent contributions, see Anthony Black (ed.), Towards Employment-Intensive Growth in South Africa (Cape Town: University of Cape Town Press, 2016); and Nicoli Nattrass and Jeremy Seekings, Inclusive Dualism: Labour-Intensive Development, Decent Work and Surplus Labour in South Africa (Oxford: Oxford University Press, 2019).