Dynamic de-responsibilization in business-government interactions
University of Cape Town Graduate School of Business
Uncorrected author version of a manuscript accepted for publication in Organization Studies.
I contribute to the “political CSR” debate on whether companies can or should fill public
governance gaps by exploring how related business-government interactions may result in corporate
social irresponsibility. I studied interactions between mining companies and the government in South
Africa between 2002 and 2012, focusing on processes that contributed to the deterioration of living
conditions in the Marikana area, site of an infamous massacre in 2012. The analysis resulted in a
process model of dynamic de-responsibilization, in which business-government interactions
progressively dissipate the adopted and enacted social responsibilities of both the government and
business. This extends extant critiques of political CSR and elaborates CSiR as not only a systemic,
but also a processual phenomenon, and it focuses related scholarly and practical attention on
problematizing the government’s role as custodian of democratic accountability.
Political corporate social responsibility, business-government interactions, corporate social
irresponsibility, process, de-responsibilization
Scholars argue that companies increasingly fulfil ostensibly government functions in the
provision of public goods and the establishment of commonly binding rules (e.g., Matten &
Crane, 2005; Palazzo & Scherer, 2008), and they propose a normative framework for why and
how companies might do so legitimately by engaging in deliberative “processes of political
will formation” (Scherer & Palazzo, 2007, p. 1109). But the risks of companies adopting such
governance roles under the rubric of “political corporate social responsibility (CSR)” have
also been highlighted, descriptively and normatively (e.g., Banerjee, 2008, 2010; Edward &
Wilmott, 2008; Fooks, Gilmore, Collin, Holden, & Lee, 2013; Hussain & Moriarty, 2018;
This debate points to a need to give more careful attention to business-government
interactions, building on recent processual analyses of political CSR (Djelic & Etchanchu,
2017; Levy, Reinecke, & Manning, 2016). We lack understanding of how such business-
government interactions may result in corporate social irresponsibility (CSiR), or
“organizational actions that cause harm to stakeholders” (Mena, Rintamäki, Fleming, &
Spicer, 2016, p. 720), and specifically CSiR as a “systemic phenomenon” (Whiteman &
Cooper, 2016, p. 118).
To consider this question, I studied interactions between mining companies and the
government in South Africa between 2002 and 2012. This is a strategic research site in that
hopes for more socially beneficial government-business interactions following the transition
to democracy in 1994 were largely disappointed (Bundy, 2016; Leibbrandt, Finn, & Woolard,
2012; Piketty, 2014). This was harshly evident in the infamous Marikana Massacre in 2012. A
specific empirical focus is hence on how Lonmin, a multinational platinum mining company,
came to renege on a formal commitment to build 5,500 houses in the Marikana area, a failure
that was widely seen (even by Lonmin managers) as contributing to conditions that led to the
massacre. My analysis results in a process model of dynamic de-responsibilization, in which
business-government interactions progressively dissipate the adopted and enacted social
responsibilities of both the government and business.
This model explains “political CSiR” as not just a systemic but also a processual
phenomenon, in which negotiated rule-making and expectations for business to provide public
goods – that is, key tenets of political CSR (Palazzo & Scherer, 2008) – contribute to a
weakening of governance over time and lead to corporate actions that harm vulnerable
stakeholders. This extends extant critiques of the notion of political CSR that focus on the
inherent attributes of corporations (e.g., Banerjee, 2008) by showing how companies’ self-
interests and actions are circumscribed by the processual dynamics of business-government
interactions. The implication is not that political CSR efforts are necessarily harmful, but that
more attention must be given to how and to whom such efforts are accountable, and
specifically to the role of the government as the often constrained custodian of democratic
Political CSR, government-business interactions, and CSiR
Political CSR and its critics
The political CSR conversation highlights the expanding role of corporations as
political actors, and as such has created a fruitful platform to bring politics and the state, and
diverse governance contexts, more centrally into scholarly CSR discussions (e.g., Frynas &
Stephens, 2015; Scherer, Rasche, Palazzo, & Spicer, 2016). Matten and Crane (2005) describe
how corporations are increasingly making efforts to promote and protect human rights in
situations, where governments are unable or unwilling to do so (see also the related work by
political scientists, e.g., Börzel & Risse, 2010; Mayntz & Scharpf, 1995). Meanwhile, Scherer
& Palazzo (2007) delineate a normative theory of political CSR for this “era of globalization”
when “law and the state apparatus are insufficient means for the integration of business
activities with societal concerns” (op cit., p. 1096), recommending that corporations
participate in deliberative multi-stakeholder “processes of political will formation” (op cit., p.
1109; Scherer & Palazzo, 2011).
However, the notion that corporations become participants, or even facilitators, of
deliberation on the common good has been criticised as paying insufficient attention to how
companies’ vested interests and political power likely contribute to the subversion of any
truly deliberative process (Banerjee, 2008, 2010; Edward & Wilmott, 2008; Fooks et al.,
2013; Whelan, 2012). “Corporations do not have the ability to take over the role of
governments in contributing to social welfare simply because their basic function (the rhetoric
of triple bottom line aside) is inherently driven by economic needs” (Banerjee, 2008, p. 74).
Hussain and Moriarty (2018) extend this critique by arguing that the normative model of
political CSR fails to distinguish between agents who are accountable and the “parties to
whom agents are accountable” (op cit., p. 525). Corporations cannot be the latter because they
do not “legitimately represent a group of citizens in social deliberation” (op cit., p. 527), and
hence they may contribute to deliberative processes as technical experts, but should not do so
as fully-fledged participants.
We therefore have contrasting descriptive and normative accounts of whether
corporations can or should effectively fill public governance gaps as political actors. This
impasse may rest, in part, on the fact that most analyses of political CSR focus on companies,
governments, and so on, as “things,” and their corresponding traits, rather than on the
processual dynamics of their interactions over time. It helps to pay more careful attention to
dynamically unfolding business-government interactions, as explained below.
A processual view of business-government interactions
Though scholars have attended to the role of the government in shaping companies’
CSR and related behaviours (e.g., Crane, 2011; Mäkinen & Kasanen, 2016; Schrempf-
Stirling, 2018; Vogel, 2010; Whelan, 2012), business-government interactions arguably have
been given insufficient or only partial attention in the political CSR literature. As much as
business-government interactions arise, they are mostly considered in a one-way direction.
For instance, numerous authors have focused on the important role of governments in
promulgating “hard” or “soft” laws influencing corporate behavior in general and CSR
specifically (e.g., Albareda, Lozano, & Ysa, 2007; Ruggie, 2014; Schrempf-Stirling, 2018).
Others have focused on the state’s absence and how this motivates companies to step in to
ensure a viable operating context (e.g., Börzel & Hamann, 2013; Börzel & Risse, 2010;
Matten & Crane, 2005). Meanwhile, the corporate political activity literature has focused on
when and how companies seek to influence the government’s rule-setting or allocation of
resources (e.g., Child & Tsai, 2005; den Hond, Rehbein, Bakker, & Lankveld, 2014) – though
notably some of these analyses have begun to consider “mutual influence” between
governments and companies (Child & Tsai, 2005, p. 96).
There is much scope in exploring such “mutual influence” as it unfolds over time.
Indeed, much of the current political CSR literature adopts mostly a static, substantive
approach, which focuses on business, states, governance gaps, and so on, as “things,” with
little explicit role for time and processual dynamics (Langley, Smallman, Tsoukas & Van de
Ven, 2013). Yet we may expect that processes of interaction and path creation (Djelic &
Quack, 2007) likely play an important role in how businesses assume and enact their roles,
responsible or otherwise, in diverse governance contexts.
Djelic and Etchanchu (2017, p. 642) demonstrate the power of a processual approach by
means of a “historical contextualization [that] shows that the frontier between economy and
polity has always been blurry and shifting and that firms have played for a very long time a
political role [which] co-evolved with shifts in dominant ideologies” (see also Mäkinen &
Kourula, 2012). Historical manifestations of CSR in the early 20th century were aligned by
their emphasis on the “principle of discretionary power,” which “takes for granted that private
actors voluntarily have the legitimacy to engage in political roles and responsibilities” (op cit.,
p. 656). But workers saw this paternalistic posture as a manipulative attempt to perpetuate
exploitative employment practices, and these “negative reactions” led to “the development of
the welfare state in Europe and the deployment of the New Deal in the United States” (op cit.,
p. 657). Djelic and Etchanchu (2017) thus describe a swing from private discretion to state
The opposite swing is described by Heath and Norman (2004, p. 256) for the case of
state-owned enterprises (SOEs): “The heady days of the 1960s, in which SOEs were
encouraged to pursue a variety of social objectives, were followed by a long period of
‘commercialization’”. In a related vein, Chua (1995) describes an oscillation between an
emphasis on free enterprise and “incentives for effort”, on the one hand, and an emphasis on
state intervention and nationalization to address negative social outcomes, on the other. An
initial model of business-government interactions thus suggest that social responsibilities are
allocated in an oscillation between state intervention and managerial discretion.
A second model of business-government interactions emerges from Levy, Reinecke,
and Manning’s (2016) emphasis to take “temporality seriously” (op cit., p. 365). Based on
their longitudinal study of interactions between coffee companies and civil society
organisations, they describe a dynamic process, in which NGOs’ challenge to the incumbent
companies’ value regimes leads to an accommodative dynamic as companies and NGOs make
successive strategic concessions. Incumbents and challengers adjust to each other’s strategies
and claims in a process of mutual accommodation, leading to a convergence in their value
regimes and increasing agreement on what business responsibilities look like. While Levy and
colleagues focus on civil society organizations as the challengers, it may be argued that
governments also play this role, and that a similar process of mutual accommodation takes
place as governments and business seek and find the “middle ground” in defining and
allocating social responsibilities, where the “overt ambition is neither to destroy capitalism
nor to… liberate it” (Gray, Owen, & Maunders, 1988, p. 8; see also Tinker, Neimark, &
CSiR as “systemic phenomenon”
The two models outlined above provide explanations of how social responsibilities are
continuously defined and allocated in the path-dependent interactions between business and
governments (or civil society organizations). They also begin to offer disparate clues as to
how such interactions may give rise to CSiR and negative social outcomes: In the oscillation
model, Djelic and Etchanchu (2017) suggest that an emphasis on managerial discretion gave
rise to negative social outcomes, while Heath and Norman (2004) argue that state intervention
and a broadening of managerial responsibilities in fact created greater social costs. In the
mutual accommodation model, the resulting “middle ground” may represent a compromise
that still furthers managers’ interests at the expense of stakeholders (Tinker et al., 1991).
But we lack a more focused analysis of how business-government interactions may
result in CSiR, that is, “organizational actions that cause harm to stakeholders” (Mena et al.,
2016, p. 720; see also Armstrong, 1977; Lin-Hi & Müller, 2013; Strike, Gao & Bansal, 2006),
including the “deception and manipulation of stakeholders” (Greenwood, 2007, p. 324) and a
resulting lack of the corporation’s reliability and trustworthiness (McMahon, 1999). Of
particular importance here is Whiteman and Cooper’s (2016) description of CSiR as a
“systemic phenomenon,” where CSiR is not a firm-level attribute but a consequence of
“diffuse actions performed by a variety of institutional actors” (op cit., p. 118).
Whiteman and Cooper’s (2016) case study thus focuses not only on the company in
question, but also on the role of the Forestry Stewardship Council. However, the government
remains largely in the backdrop even though it plays a prominent role in failing to address
egregious abuses of human rights. Mention is merely made of “its lack of governance
capacity and the high rate of corruption” (op cit., p. 116). Whiteman and Cooper (2016) also
do not explicate the process that gives rise to “multiparty decoupling” (op cit., p. 118). There
remains an important opportunity, therefore, in applying the temporal lens in political CSR
studies (Djelic & Etchanchu, 2017; Levy & colleagues, 2016) to explore how government-
business interactions may give rise to CSiR as a systemic phenomenon.
The South African mining sector is a strategic research setting to explore how business-
government interactions may give rise to CSiR. Following the transition from Apartheid and
the establishment of the first democratic government in 1994, expectations were high that
government-business interactions would give rise to more socially beneficial outcomes.
However, these expectations were largely disappointed (Bundy, 2016; Leibbrandt et al., 2012;
Piketty, 2014), and this was harshly manifest in the infamous Marikana Massacre in 2012,
when 34 striking mineworkers were shot and killed by police. The mineworkers were striking
for higher wages and also lamented their dismal living conditions in the slums around the
mines. Part of the motivation for this research was to better understand what contributed to
the underlying conditions that resulted in the massacre, especially the deleterious living
conditions around the mines. In so doing, I could draw upon data that I have been collecting
on CSR among mining companies in the area around Marikana since 2001.
The mineworkers striking at Marikana were employed by London-listed platinum
mining company Lonmin, which became implicated in creating the conditions that gave rise
to the unprotected strike action and the escalation in violence preceding the shooting
(Alexander, 2013; Chinguno, 2013). The judicial Commission of Inquiry concluded that
“Lonmin’s failure to comply with its housing obligations created an environment conducive
to the creation of tension, labour unrest, disunity among its employees or other harmful
conduct” (Farlam, Hemdraj & Tokota, 2015, p. 542). These housing commitments were made
in line with the Mining Charter, which was negotiated between companies and the state in
2002. The Commission highlighted that the company only “built three of the 5500 houses
which should have been built,” thus exacerbating a situation, in which “large numbers of
Lonmin workers live in squalid informal settlements surrounding the Lonmin mine shafts”
(op cit., p. 527). In one of the Commission’s hearings, a Lonmin executive conceded that the
“board and executive of Lonmin understood that the tragic events at Marikana were linked to
that [housing] shortage” (op cit., p. 527-528). In failing to live up to its commitment, the
company eroded its trustworthiness and contributed to harm to the company’s workers and
others living in the surrounding villages, and both of these aspects characterise this as an
instantiation of CSiR (Mena et al., 2016; McMahon, 1999).
My research thus sought to explore whether this was just another case of malfeasance
by a corporate “bad guy” (Scherer & Palazzo, 2007, p. 1109), or whether this case may help
advance theorising around the interplay between CSR/CSiR and governance (Moon, 2002),
and specifically the argument that corporations are playing and should play an explicitly
political role by making “rules and providing public goods” (Palazzo & Scherer, 2008, p.
774). This potential is manifest in Lonmin’s ostensibly prominent engagement in political
CSR. As one of the world’s three largest platinum mining companies at the time, it helped
negotiate new rules for the industry in the so-called “Mining Charter” and an associated
“scorecard” that stipulated requirements for mining companies in addressing the legacies of
apartheid, such as affirmative action in employment and procurement, as well as contributions
to socio-economic development in communities next to mines. These efforts clearly
represented the rule-making aspect of political CSR. Lonmin also had a reputation as a
“responsible” and “sustainable” company, and was recipient of numerous sustainability
reporting awards and a member in various responsible investment indices, thus contributing to
self-regulation and the provision of public goods, i.e., further aspects of political CSR. Yet
these apparent political CSR efforts clearly did not result in the positive outcomes hoped for
in much of the early political CSR literature (see Scherer & Palazzo, 2011; Scherer et al.,
2016). I thus embarked on a case study (Yin, 2003) focused on how business-government
interactions contributed to the irresponsible actions that led to the dismal social conditions
around the mines in the Marikana area.
My data are based on a combination of interviews, diverse documents, and direct
observations. This responds to recommendations for case study research (Yin, 2003) and
process studies (Langley, 1999; Langley, Smallman, Tsoukas, & Van de Ven, 2013), and it is
also aligned with prior work on dynamic interactions between business and other stakeholders
(Levy et al., 2016). The combination of sources situated in a named study area seeks to
respond to organisation scholars’ expectations of replicability as well as historiographers’
expectation of verifiability (Rowlinson, Hassard, & Decker, 2014, p. 258).
My primary data are interview transcripts based on three sets of interviews conducted
between 2001 and 2015. The first set of almost 100 interviews was conducted between 2001
and 2004 with platinum and chrome mining companies (including Lonmin), local
governments, traditional authorities, labour unions, and non-governmental organizations in
the study area, as well as business, government and civil society actors at the national level. A
key objective of these interviews was to assess companies’ narratives and practices in their
social and environmental context, including the effects of these corporate activities on the
social environment. A core topic in these interviews was how mining companies’ conception
of CSR was defined and implemented, and how this was influenced by state policy and
interactions between companies and the state. Another recurring theme related to the housing
of mine workers and specifically the shifting perceptions of what companies or other
stakeholders were – or should be – taking responsibility for. While my interviews were
guided by a protocol to ensure these key themes were addressed, my questions and
interviewing style adopted a narrative approach to allow the generation of in-depth, context-
specific information (Kvale & Brinkmann, 2009). This interviewing process included three
site visits to the study area (in 2002 and 2003), which provided important observational data
and impressions especially regarding people’s living condition around the mines.
My second set of interviews consisted of five interviews conducted in 2007, focused
more specifically on the opportunities and challenges faced by the mining companies, local
government, and other role-players to better collaborate to address the growing social
problems around the mines. This also included another site visit.
Then, prompted by the Marikana Massacre in 2012 and subsequent, continuing labour
unrest, a third set of 23 interviews was conducted in 2014 and 2015 to analyse the underlying
factors contributing to the violence that plagued labour and community relations in and
around Lonmin’s Marikana mine in 2012. These interviews again included an emphasis on
the relationship between mining companies and the local and national state. While the focus
was on Lonmin and the communities in the Marikana area, my initial findings suggested that
the social circumstances that contributed to the Marikana Massacre were not unique to
Lonmin, and so my data collection continued to include other mining companies in the
broader area. This helped corroborate data collected on business-government interactions in
the case of Lonmin, and it also helped in generalizing my resulting argument.
While many of the interviews in this third round were with new interviewees, I
purposefully sought out salient interviewees, whom I interviewed in my first round, as far as
possible. This helped create a temporal narrative across the roughly 13-year period between
the first and third phases. In the third round of interviews, I again intentionally interviewed
not only business and government representatives, but also trade unionists and members of
local community organisations and other civil society organisations. This helped develop a
richer picture of business-government interactions and how they gave rise to social
circumstances around the mines, and it also helped corroborate and elaborate data from
business and government interviewees and from archival and documentary sources. In total,
125 interviews were conducted, as outlined in Table 1. Almost all interviews were recorded
and transcribed with the permission of the participants.
In addition to the interview data, I collected documents in the form of archival material;
news articles; policy documents; public and internal reports from companies, government
agencies and non-governmental organizations; as well as extensive transcripts and the final
report of the Marikana Commission of Inquiry. These archival and document data helped to
augment and fill out the temporal “storyline” between the three interview phases, and they
also helped corroborate the retrospective sensemaking included in the interviews. An
overview of the archival and document data is provided in Table 1.
------------------------INSERT TABLE 1 SOMEWHERE HERE-----------------------
I adopted an inductive and iterative approach, identifying patterns and emerging themes
in the data (Glaser, 1998; Glaser & Strauss, 1967; Locke, 2011), and then “enfolding” the
literature to identify particularly promising themes and angles of inquiry, which were then
used to revisit the data (Denzin & Lincoln, 2005). This iterative back-and-forth involved
multiple rounds of coding (Miles, Huberman & Saldana, 2014). In the first round of analysis,
exploring the different manifestations and changing nature of the business-government
relationship became a primary focus. This helped crystalize the research question around how
business-government interactions over time contributed to CSiR. The subsequent analysis
thus focused on temporally situated practices and interactions. Initially, this involved
“temporal bracketing” (Langley et al., 2013, p. 7) in order to create an “analytically structured
history” (Rowlinson et al., 2014, p. 263) connecting actions and actors’ legitimating accounts
in particular temporal phases with specific beginnings and endings. I abandoned my focus on
identifying periods as I recognised how government and business practices were interacting
dynamically and overlapping in time. Analysing these dynamic interactions gave rise to a
model highlighting business-government interactions at national and local levels of analysis.
A third round of analysis was motivated by incisive peer review and, rather than
emphasise the geographical levels of analysis, it focused on the two primary dimensions of
governance: making “rules and providing public goods” (Palazzo & Scherer, 2008, p. 774; see
also Risse, 2011). This helped to revise in a more specific manner my elemental and pattern
coding (Miles et al., 2014), focusing on the temporally situated practices of government and
business in the process of making rules and in committing to and delivering on the provision
of public goods. The pattern coding contributed to a series of emerging network displays
(Miles et al., 2014) to align the key practices and their inter-relationships over time,
eventually resulting in the model of dynamic de-responsibilization described below.
My analysis results in a process model that explains how interactions between business
and government give rise to a dynamic of increasing de-responsibilization across both
business and government. By de-responsibilization I mean the progressive dilution or denial
of social responsibilities – a process that culminates in social irresponsibility and harm to
stakeholders. This model is illustrated schematically in Figure 1. It highlights sequential and
interacting practices of government and business in the two key dimensions of governance:
rule-making and the provision of public goods.
The process commences with the government engaging in negotiated rule-making
(marked A in Figure 1). This sets off a series of subsequent responses by business and
government, marked B and C in Figure 1. Simultaneously, government’s offer of negotiated
rule-making is connected to its expectation that business contributes to the provision of public
goods, marked D, creating a series of reactions marked E and F. The culmination of these two
inter-related series of reactions by business and government is that business reneges on its
commitment to provide public goods, marked G in the figure. I now describe these two series
of interactions in more detail, commencing with a focus on rule-making.
------------------------INSERT FIGURE 1 SOMEWHERE HERE-----------------------
Negotiated rule-making and resulting ambiguity and enforcement difficulties
The point of departure of the dynamic interaction model is the government’s
commitment to and engagement in negotiated rule-making, indicated by A in Figure 3. After
South Africa’s first democratic elections in 1994, the government embarked on an ambitious
policy and legislative reform programme, also in the mining industry, which had been at the
core of the country’s industrialisation and political economy since large deposits of diamonds
were found in the 1870s and of gold in the 1880s. A primary objective of the new Minerals
and Petroleum Development Act of 2002 and the associated negotiation of a “Mining
Charter” was the “transformation” of the industry to address apartheid legacies and to enable
black South Africans’ economic participation in the industry. At the height of this negotiation
process, the editors of the influential Sowetan newspaper complained: “The mining houses
still call the shots… government has been very slow in pushing the transformation of the
industry thanks to the need to bend over backwards and accommodate the concerns of the
industry” (The Sowetan, 2002). I subsequently put this concern to then Minister of Minerals
and Energy Phumzile Mlambo-Ngcuka (personal communication), who replied:
It’s shades of grey: business, government and others are deciding in terms of negotiation
and partnership… and we are happy with the outcome. The London investors also have a
right to be heard, it’s their money. So, we call the shots together; rather than ‘exercising
power over’ it’s ‘power with’ – it is an inclusive approach to governance. Companies
must comply with the charter, so it must work for them.
The Minister’s response highlights the government’s emphasis on “negotiation and
partnership,” expecting companies to contribute transparently to public deliberation on new
rules and that this will increase the rules’ legitimacy. This need for legitimacy seemed
obvious when a draft of a government proposal was leaked in July 2002, requiring that 51 per
cent of the industry be controlled by black-owned companies within ten years, and the
markets reacted with a drastic sell-off of shares (e.g., interviews 1A2, 1G1; various media
articles and investment analyst reports). This event invigorated negotiations between the
government and mining companies, resulting in a revised Mining Charter and an associated
“scorecard” with targets for, among other things, worker training, “employment equity”
(elsewhere referred to as affirmative action) for “historically disadvantaged South Africans,”
socio-economic development in mine-affected communities, and the provision of
mineworkers’ housing. The government’s offer of negotiated rule-making was thus premised
on an ideological commitment to negotiation – arguably emanating from the negotiated
transition to democracy a few years previously – as well as the practical concern to not scare
Business responded to the government’s offer of negotiated rule-making by similarly
extolling the need for negotiation and the recognition of shared interests, and this involved an
emphasis on the need for flexibility in the design and implementation of the rules (component
B in Figure 1). Notwithstanding the acrimony surrounding the leaked proposals, the revised
charter was celebrated in the rhetoric of some mining company leaders as being in business’s
enlightened self-interest. As argued by the then CEO of Anglo Gold, “I think [the revised
Charter] is a very good document and is going to make the South African industry more
competitive, not less, and lead to greater wealth creation, not less” (Anglo Gold, 2002, p. 9).
Some of this is likely to have been public relations, especially to assuage shareholders. But it
also represents business leaders’ emphasis on negotiation and shared interests as a means to
influence the rule-making process and its outcomes.
The negotiated rule-making process thus resulted in high degrees of ambiguity in the
rules themselves and in diverse actors’ interpretations of them, and numerous interviewees
involved in the process argued that this was at least in part due to companies’ involvement in
the negotiations (interviews 1A2, 1G3, 1G5, 1C2). For example, the Mining Charter
scorecard required that companies “cooperate in the formulation of integrated development
plans and… in the implementation of these plans” in mine-affected communities (Republic of
South Africa, 2002). It is apparent that any assessment or enforcement of compliance to this
obligation would be impossible. Another example of this ambiguity surrounded companies’
responsibility to offer employees “homeownership options” and this will be described in more
detail below. Consequently, a review conducted by the government in 2009 found that
inherent ambiguity in the Charter had contributed to “shocking levels of noncompliance” and
thus recommended its revision (Department of Mineral Resources, 2009, p. 22). This was
likely compounded by the significant administrative challenges faced by the implementing
authority in dealing with companies’ applications and compliance reports (e.g., interviews
3G1, 3M3). But the rules’ ambiguity was widely recognised as an underlying factor in
government’s systematic inability to enforce the negotiated rules – indicated by C in Figure 1.
High degrees of ambiguity and flexibility characterised not only the negotiated rule-
making process and the resulting rules, but also their implementation. The interpretation of
the Mining Charter targets at the operational level relied on the interaction between company
management and a range of consultants and state agents, in response to local and company-
specific circumstances. As argued by a government interviewee involved in the negotiations
of the charter (interview 1G3):
Government sets the framework – you don’t want to straightjacket anybody, so there is
flexibility in the implementation of the framework. It’s like this: say you want to get
[from Pretoria] to Cape Town – there is no flexibility in that you need to get there, but
there is in terms of how you get there.
The shared emphasis on flexibility thus increased room for manoeuvre in the interpretation
and implementation of established rules, expectations, and commitments. In sum, the
government’s and companies’ ideological and practical motivations to engage in negotiated
rule-making resulted in ambiguous rules and an emphasis on flexibility in their enforcement.
The implications became evident some years later in the contrasting views of companies’
responsibilities for providing public goods, as described in the following sub-sections.
The provision of public goods
After 1994, the new, democratic government built upon an established tradition of
companies making “corporate social investments” in support of health, education, and other
public goods, and it sought to codify and expand such requirements in the Mining Charter. As
noted, the negotiated rule-making process resulted in the requirement that companies
“cooperate in the formulation of integrated development plans” at the local level. The Mining
Charter also identified the provision of housing to mining employees as a company
obligation, though this was similarly ambiguous, requiring that companies “facilitate
homeownership options for all mine employees.” As indicated by process component D in
Figure 1, the state thus enjoined companies to contribute to the provision of public goods, but
left these expectations remarkably vague.
The business response was to vocally accept and advance such expectations that they
would become providers of public goods in mine affected communities (as indicated by E in
Figure 1). In fact, mining companies sought to out-do each other in acting, and being seen, as
public development agents. This was evident in almost all interviews, especially those with
company representatives, as well as in companies’ public reports, especially from 2002
onwards. This was partly motivated by the World Summit on Sustainable Development,
which was held in Johannesburg in 2002 and during which the role of mining companies in
providing public goods was universally recommended (e.g., interviews 1M11, 1M17, 1C5;
see also MMSD, 2002). It was also evident in companies’ bill-board advertising at the local
level in the area around Marikana during various site visits. Some mining company
interviewees identified this obligation to provide public goods as a moral obligation, or as
“blood money” for the social disruption created by mining and its migrant labour system
(interview 1M5; see also Bezuidenhout & Buhlungu, 2011). During the period of data
collection, however, an increasing number of mining company interviewees emphasised that
providing public goods had become part of the “social licence to operate,” and that the more
such public contributions were recognised, the better the company’s chances to receive
mining licences from the government. A third motivation was highlighted by some
interviewees and focused on the concern that the local municipality was clearly unable to
provide public goods and hence mining companies had to fill such gaps to maintain a viable
operating environment. As noted by one executive:
The reality is that if you look at how well local government is working, it is not. If you
look at areas like Mooinoi and Marikana, up to even Brits, where local government is
supposed to be looking after [public goods and services], we are doing refuge removal,
we are doing sewage plants, we are supplying quite a big portion of water (interview
Whatever the motivation, mining companies’ contributions to the provision of public
goods in the area around Marikana were significant. The annual budgets dedicated to
“corporate social investment” by the largest three companies in that area amounted to a figure
that was of a similar order of magnitude to the capital budget of the municipality. Yet these
contributions manifestly failed to make a significant difference to people’s living conditions
in the slums around the mines. This is likely due to a range of reasons, but for the sake of our
analysis an important governance factor was that the local government absconded from many
of its state responsibilities, and this was at least partly motivated by the government’s
expectation that mining companies would fill this gap (see component F in Figure 1). This is
illustrated in the following quote from a senior mining company manager (interview 3M3):
The municipality saw this as a way to deliver on their agenda; to get companies to put in
roads and so on. It should have been about augmenting what government does, but it
became so that companies became delivery agents for government. I was at a community
celebration… they had invited the mayor, we were launching a community project where
we had built a community hall, a road, quite a few facilities; and speaker after speaker,
including the [traditional] chief, were saying to this municipality guy, that every time
they go to the municipality for help… they get told, “you, you are the children of the
mine, whatever your needs are, you go and you ask the mine, don’t come here, we have
lots of other communities that don’t have mines that look after them.” This happens
within government, as well. I was told by [the municipality] in May, “we don’t have
rubbish bins and every time we go ask for budget for this, we are told go to the mines and
ask for budget from them.” The mayor gets told this when she goes to ask for budget
from the province.
Indeed, an interviewee from local government (interview 3G2) clearly distinguished between
areas of local government responsibility and those supported by the mines: “We are
concentrating on our own development [areas] as a local government, and the mines are
focusing on their areas.” This abrogation of state responsibilities was clearly unconstitutional.
It was also detrimental to mine-affected communities, because it undermined the kind of
collaborative effort between mining companies and government, which was necessary to
respond to the complex social development problems around the mines. Significantly, it was
also a self-reinforcing dynamic: As the state absconded from its responsibilities in mining
areas, mining companies were increasingly compelled to fill these gaps, which in turn
furthered the state’s rationale to concentrate elsewhere.
Given the self-reinforcing dynamic that increased expectations for companies to fill
governance gaps, and given the lacking ability or willingness of the local state to support such
efforts with vital planning and bulk infrastructure support, the companies became
overwhelmed and reneged on their commitments – as indicated by G in Figure 1. One senior
manager noted (interview 2M4), “There is no way we can do all these things people are
looking at us to do, by ourselves; the needs are just too great.” To conclude, companies had a
range of motivations to contribute public goods and services in neighbouring communities,
yet the local government responded by absconding from those areas and the resulting lack of
government investment and coordination contributed to companies reneging on their
commitments. The following sub-section describes in more detail how an absconding local
government and a lack of enforcement of ambiguous rules jointly contributed to companies
reneging on their commitments to provide public goods; i.e., how components C and F both
contributed to G in Figure 1.
Progressive de-responsibilization: A focus on Lonmin’s housing commitments
As mentioned, one of the most prominent and controversial issues surrounding
Lonmin’s role in creating the underlying conditions for the Marikana Massacre was its failure
to comply with its commitment to build 5,500 houses for workers. The company made this
commitment in its “Social and Labour Plan,” a statutory document that specified the
company’s responses to the Mining Charter scorecard at the local level. One way to explain
this failure would be to characterise the company as an inherently untrustworthy actor that is
prone to reneging on commitments whenever possible. But this would underexpose the
“systemic” nature of this example of CSiR (Whiteman & Cooper, 2016) and specifically the
way, in which the dynamic governance interactions between business and government made
this reneging of a formal commitment possible.
The negotiated rule-making process and its emphasis on flexibility gave rise to vague
and ambiguous responsibilities, allowing for differential appraisals of what the company’s
responsibility actually was. This was evident in how different actors advanced contrasting
responses to the Marikana Commission’s finding that one of Lonmin’s key commitments had
been blatantly ignored. On the one hand, an academic activist described this as follows:
We know that, despite committing to building 5,500 houses in its Social and Labour Plan
(a commitment that becomes legally binding once [it] is approved by the department), it
in fact managed to build only three. Adding insult to injury, in 2009 Lonmin unilaterally
adjusted its commitment to three houses and proceeded to audaciously give itself a 100%
compliance rating. Quite simply, this is illegal (Chamberlain, 2015).
On the other hand, a Lonmin manager (interview 3M2) emphasised the overarching flexibility
in responding to Mining Charter commitments, noting that the mining charter “is a scorecard,
not a single measurement,” and so even if the company failed to fulfil its housing obligations,
its performance in other areas would qualify it for sufficient points to pass muster. Moreover,
much of this flexibility and ambiguity corresponded to the Mining Charter’s vague
requirement that companies “facilitate homeownership options for all mine employees.” The
imprecise language used by the drafters allowed the company an interpretation that was at
odds with that of the government, critical NGOs, and indeed the judicial Commission. The
potential for these claims to be tested in court is limited by the vagueness and ambiguity of
the Charter’s phrasing. Specifically, it is difficult to reliably confirm to what extent “options”
have been “facilitated.” This is particularly so because of the complexity of homeownership
in the context of many workers being migrants from other parts of the country. Lonmin
reports, for instance:
Among the different aspects of the housing challenge are… a lack of interest in home
ownership [among employees]… Many employees still opt to stay in hostels or rentals as
they are considered to be more affordable and employees from distant labour-sending
areas often do not want to invest in property at Marikana when they already have a home
and community base elsewhere” (Lonmin, 2013a, p. 12).
The need for flexibility in rule-making was motivated by reference to the need for adaptability
to local contingencies and complexities. These local contingencies and complexities were
then highlighted again to explain why initial commitments could not be abided by.
This emphasis on adaptability and flexibility was also evident in the company’s public
reports. In an NGO-sponsored study, Lonmin reports from 2003 and 2012 were analysed to
“hold the company to account against its own promises and stated goals, and against
regulations and laws as Lonmin itself cites them” (Benchmarks Foundation, 2013, p. 2). It
Targets are set, missed, explained or abandoned. Apologies are made, new targets set.
Things are conflated together and then separated off and confusion reigns. When we try
to measure what Lonmin has done to provide decent accommodation or what it has
actually spent in communities over the years, it is difficult to separate fact from fiction…
Lonmin is, by its own admission and on the basis of its own documentation, clearly
violating its commitments in the Mining Charter (op cit., pp. x, 39).
The company’s managers, on the other hand, again point to the need for flexibility in
identifying and responding to development needs around the mines, as well as the need to
recognise “the learning journey” (interview 3M2) the company was on with regard to the
practice of sustainability reporting. The company’s media release responding to the NGO
report states accordingly:
We are concerned that the underlying assumption… is that changes in measurement,
reporting or commitments ought to be taken as evidence of failure or deception.
Sustainability reporting is not an exact science and has changed a great deal in the last
decade, as have environmental and other standards (Lonmin, 2013b).
Engaging in and responding to negotiated rule-making and commitments to transparency –
whether in response to national rules such as the Mining Charter or international expectations
such as sustainability reporting – thus went hand in hand with an emphasis on flexibility in
implementation and rule enforcement. Crucially, this wilful ambiguity in the rules stipulating
business responsibilities exacerbated the government’s limited capacity to effectively monitor
and enforce companies’ performance relative to the negotiated rules. This is how business-
government interactions culminated in Lonmin’s ability to renege on its commitment to build
a specified number of houses.
In addition, Lonmin justified its departure from the stated commitment to build houses
by referring to the lack of local government support. The local state was manifestly unable to
respond to the complex social and ecological problems arising due to the growing mining
industry and associated trends, especially immigration. The local government admitted, “it
has not provided a significant number of government-subsidised affordable housing [and
associated bulk infrastructure] in recent years” (Rustenburg Local Municipality, 2015, p. 60).
As noted by a local government official (interview 3G1), “government is overloaded in
providing houses for people, so shacks are mushrooming, and then issues of service delivery
arise… government becomes overloaded and struggles to provide those services.” Local
government’s capacity constraints were compounded by organisational problems. Auditors’
reports repeatedly gave the local municipality a qualified audit, and they also identified a
variety of strategic, procurement, and information management deficiencies: “The
municipality did not have and maintain effective, efficient and transparent systems of
financial and risk management and internal controls” (Office of the Auditor-General, 2012, p.
6-7). These concerns became even more serious in the context of corruption-related murder
(e.g., City Press, 2010).
These budgetary, human resources, and management problems constrained
municipalities’ effectiveness in providing housing and municipal services. They also
constrained mining companies’ ability to implement their housing and other social
programmes. One manager argued (3M4), “We must not underestimate some of the serious
capacity constraints within the municipalities, and the extent to which this constrains even the
ability of mining companies to deliver on our own Social and Labour Plan obligations.”
Mining companies’ challenges resulted from local government were multifaceted. Delays and
difficulties faced by local government in providing land use planning and bulk infrastructure
constrained housing and other social projects. Even though local government enjoyed
constitutional and electoral legitimation, councillors often struggled to provide clear
representation for communities in interactions with mining companies and other development
proponents, and conflicts between local government and traditional authorities were
commonplace (as described in many interviews and documented in various media reports; see
also Capps & Mnwana, 2015).
These severe governance challenges faced by the local government contributed to
government officials referring to mine affected communities as “children of the mine” (see
quote from interview 3M3 above), and then focusing their limited resources elsewhere. This
exacerbated the lacking attention by local government to mine affected communities, which
in turn created the planning and bulk infrastructure deficits that Lonmin pointed to when
justifying abandoning its commitment to build 5,500 houses. The overarching outcome was
that neither the government nor the company lived up to their statutory or public
commitments to provide public goods in mine-affected communities, as outlined in Figure 1,
and this helped create the underlying conditions that contributed to the Marikana massacre in
Figure 2 is an adapted version of this model, with the extent of adoption and enactment
of social responsibility on the y-axis. This is to highlight how the adopted and enacted social
responsibility of both government and business dissipate over time, as business and
government respond to each other’s posture and practices in interacting governance processes
of rule-making and the provision of public goods.
------------------------INSERT FIGURE 2 SOMEWHERE HERE-----------------------
Key elements of this model resonate with prior scholarship and broader trends. The
process commences with government engaging in negotiated rule-making, emphasising
shared interests between private and public realms and the need to enhance the legitimacy of
rules. This corresponds to broader trends that are well established in the literature – indeed, a
core tenet of “new governance” and political CSR is that governments involve business in the
process of negotiating rules (Bingham, Nabatchi, & O'Leary, 2005; Moon, 2002; Scherer &
Palazzo, 2007, 2011). A common premise is that the resulting rules need to respond to the
needs of business, to maintain or enhance the necessary incentives for economic activity
(Chang et al., 2017; Djelic & Etchanchu, 2017). Prior scholarship also reflects the way in
which business, in my case analysis, responded to this opportunity by emphasising the need
for flexibility and introducing high degrees of ambiguity (Whelan, 2018). Finally, others have
pointed to how the state may abscond from its responsibilities to provide public goods not
only because of its lacking willingness or capacity, but because it diverts attention away from
areas or communities receiving business attention (Börzel & Hamann, 2013; Ilunga, 2011).
The process model developed here thus builds on a specific empirical case study to connect
and elaborate prior findings in other settings into a temporal dynamic illustrated in Figure 2.
This contributes to the scholarly conversations at the intersection between political CSR,
business-government interactions, and CSiR.
The primary contribution of this model is to highlight the processual character of
“political CSiR,” that is, to explicate how business-government interactions moulded by key
elements of political CSR – negotiated rule-making and expectations for business to provide
public goods – can in fact weaken public governance over time and lead to corporate actions
that harm vulnerable stakeholders. It is the combination of ambiguous and difficult-to-enforce
rules emanating from negotiated rule-making, on the one hand, and the retreat of the state
from areas or domains ostensibly addressed by companies’ political CSR activities, which
together culminate in CSiR as not only a systemic but also a processual phenomenon.
This extends extant critiques of the notion of political CSR that focus on the inherent
attributes of corporations, arguing that corporations cannot contribute “social welfare simply
because their basic function… is inherently driven by economic needs” (Banerjee, 2008, p.
74). Companies’ self-interests are clearly part of the problem in the case study described
above and the resulting model, but how they manifest is circumscribed by the processual
dynamics of business-government interactions. This is theoretically important because it
widens our analytical frame and directs attention to a broader set of actors and dynamics. It
also important practical implications for both private- and public-sector actors (to which I
This critical analysis of political CSiR makes auxiliary contributions to research on
business-government interactions and on CSiR. Previous analyses of business-government
interactions have in common that social responsibilities are allocated and enacted in some
way or other (Albareda et al., 2007; Börzel & Hamann, 2013; Chang, Hevia, & Loayza, 2018;
Chua, 1995; Djelic & Etchanchu, 2017; Levy et al., 2016; Matten & Crane, 2005; Schrempf-
Stirling, 2018). The model resulting from this study, however, highlights not the allocation of
responsibility, but its dissipation. This dissipation is the result of dynamic interactions – an
iterative back-and-forth – between government practices and business practices. This
“temporal interconnectedness of interactions” (Levy et al., 2016, p. 395) gives rise not to
some middle ground of assumed and enacted responsibilities, but its erosion.
I thus also contribute to our emerging understanding of CSiR as a “systemic
phenomenon” (Whiteman & Cooper, 2016), beyond a reliance on “corporate bad guys”
(Scherer & Palazzo, 2007, p. 1103; see also Lange & Washburn, 2012). Whiteman and
Cooper’s (2016) analysis retains a focus on actors’ strategic and more or less conscious
choices to engage in decoupling to respond “to incompatible or conflicting demands” (op cit.,
p. 120). The “political CSiR” model developed here does not ignore such strategic and self-
interested behaviour by business (or government) to “avoid responsibility” (op cit., p. 120),
but it situates this within dynamics of path creation that contextualize and shape actors’
perception and response to their strategic interests, making possible the avoidance and
dissipation of social responsibilities.
To clarify, Lonmin’s failure to live up to its commitment to build houses was obviously
related to its economic self-interest. But it is more interesting than just that: It was possible
for the company to ignore and unilaterally reverse its commitment because of the ambiguity
of the rules that it and other companies had helped shape, and because of the government’s
failure to maintain its role as primary custodian of democratic accountability in the area
around the mines.
In other words, this case illustrates the systemic character of CSiR, in that the
manifestation of CSiR involved more than company-specific motivations, characteristics, and
actions, but instead involved a broader range of role-players. My analysis thus corresponds in
important ways with Whiteman and Cooper’s (2016) study, where the character and
implementation of FSC rules and the connivance of both the company and the national
government allowed the operation to be declared fully compliant, even exemplary, despite
repeated claims of egregious human rights and environmental abuses in its area. However, the
case analysis and resulting model described here does not rely on government complicity in
covering up or even defending human rights abuses, as in Whiteman and Cooper’s (2016)
study. Instead, even an ostensibly well-meaning government contributed to the processual
dissipation of responsibilities. This is why it is important to recognise CSiR as not only a
systemic but also a processual phenomenon.
The process of de-responsibilization relies on companies’ assumed role in negotiated
rule-making and the provision of public goods and services, and it is perpetuated by and in
turn perpetuates the government’s limited ability or willingness to uphold its statutory
responsibilities. My study thus suggests empirical support to Hussain and Moriarty’s (2018,
pp. 530-31) normative argument to delimit corporations’ role in negotiated rule-making to a
contribution as “technical experts” as opposed to “supervising authorities”. There are likely
practical challenges in drawing a clear line between such roles, however. The more immediate
implication is to highlight the need for the government to maintain and indeed strengthen its
role as custodian of democratic accountability in the establishment of commonly binding rules
and their enforcement, and in the provision of public goods. In other words, my argument
does not suggest that companies’ participation in negotiated rule-making and contributions to
public goods are necessarily harmful. Rather, it emphasises that the government must insist
on clear and enforceable rules emanating from such negotiated rule-making processes, and it
must ensure that expectations for companies to make private contributions to public goods are
clearly laid out and monitored. The retreat of government attention to areas supported by such
private contributions must be rigorously opposed.
There are also implications for companies. This model of political CSiR shows that a
dynamic process involving various role-players’ diverse actions and reactions, each of which
might seem innocuous on their own, progressively lead to severe harm to vulnerable
stakeholders and corresponding reputation and other damage to the firm. Recognising this
negative dynamic may help in revising the managers’ posture and (re)actions with a broader,
longer-term view of economic self-interest. It also points to a need for private actors to insist
that their contributions to public goods and services maintain, rather than displace, the
government and the state as the primary governance agents. There is some evidence of private
actors institutionalising such insistence, for example in cross-sector partnerships (Powell,
Hamann, Bitzer, & Baker, 2017).
Further research should clarify the boundary conditions of the dynamic process model
outlined in Figure 2. As noted above, key elements of the proposed model – negotiated rule-
making, managerial emphasis on rule ambiguity, and government retreat from areas
ostensibly “covered” by CSR – have been identified in different contexts in the literature.
However, the case study discussed here and many of the other studies referred to above share
two important contextual features: A focus on the natural resources sector in an area of
“limited statehood,” that is, where the state struggles to enforce commonly binding rules and
to provide public goods (Risse, 2011; Börzel & Risse, 2010). For instance, it is characteristic
of the coffee industry focused on by Levy and colleagues (2016), and it is the context of the
egregious goings-on in Whiteman and Cooper’s (2016) study.
Indeed, there are elements of the proposed model that may be fundamentally influenced
by such a context. For a start, natural resources companies often operate in areas, where the
state is particularly weak and where infrastructure is limited. This may motivate companies to
initially assume and advertise an apparent willingness to provide public goods (that is, item E
in the model); it may influence the local government’s incentives to abscond from areas or
communities supported by business (F); and it may exacerbate the government’s overarching
constraints in enforcing rules (C). Further research should thus explore whether dynamic de-
responsibilization occurs – perhaps in different forms – also in other sectoral and statehood
A second area for further research consists of micro- and meso-level analyses that
explicate in more detail how managers’ decisions are circumscribed by the processual
dynamic in Figures 1 and 2. Building on scholars’ growing interest in micro-level
perspectives on CSR (Aguinis & Glavas, 2012; Kourula & Delalieux, 2016; Risi & Wickert,
2017), the model developed in this paper suggests specific questions for further research, such
as: How do managers engage in negotiated rule-making and in so doing shape the resulting
rules? How do they respond to the state’s expectations that they contribute to public goods
and services? How do they identify and evaluate their options when the state absconds from
its governance role in their area? Over and above the sectoral and statehood contexts
mentioned above, it is possible that different organisational and individual factors give rise to
variance in how, or indeed whether, dynamic de-responsibilization occurs.
This study sought to build on prior work on political CSR and business-government
interactions by exploring how such processual interactions give rise to CSiR as a “systemic
phenomenon” (Whiteman & Cooper, 2016, p. 118). The empirical analysis focused on
business-government interactions in rule-making and the provision of public goods in the area
surrounding Marikana in South Africa, site of the infamous Marikana Massacre in 2012. The
resulting model highlights the dynamic interactions, through which business and government
progressively dissipate their social responsibilities, with deleterious consequences for
vulnerable stakeholders. CSiR is thus not only a systemic, but also a processual phenomenon.
Further work will help confirm whether – or under what conditions – this is a patterned
process of some broader relevance, complementing and in some ways challenging extant
approaches to political CSR, business-governance interactions, and CSiR. In any case, I argue
that scholars, government officials, and business managers should attend to the important
risks associated with ostensibly helpful political CSR efforts, and to the need to strengthen the
government’s role as custodian of democratic accountability.
Aguinis, H., & Glavas, A. (2012). What we know and don’t know about corporate social
responsibility: A review and research agenda. Journal of Management, 38, 932–968.
Albareda, L., Lozano, J., & Ysa, T. (2007). Public policies on Corporate Social
Responsibility: The role of governments in Europe. Journal of Business Ethics, 74, 391-
Anglo Gold (2002). Anglo Gold Annual Report 2002. Johannesburg: Anglo Gold.
Alexander, P. (2013). Marikana, turning point in South African history. Review of African
Political Economy, 40, 605–619.
Armstrong, J. S. (1977). Social irresponsibility in management. Journal of Business Research,
Banerjee, S. B. (2008). Corporate Social Responsibility: The Good, the Bad and the Ugly.
Critical Sociology, 34, 51–79.
Banerjee, S. B. (2010). Governing the global corporation: A critical perspective. Business
Ethics Quarterly, 20, 265–274.
Bezuidenhout, A., & Buhlungu, S. (2011). From compounded to fragmented labour:
Mineworkers and the demise of compounds in South Africa. Antipode, 43, 237–263.
Bingham, L. B., Nabatchi, T., & O'Leary, R. (2005). The new governance: Practices and
processes for stakeholder and citizen participation in the work of government. Public
Administration Review, 65, 547-558.
Börzel, T.A. & Hamann, R. (eds.) (2013). Business contributions to climate change
governance: South Africa in comparative perspective. Houndsmill: Palgrave
Börzel, T. A., & Risse, T. (2010). Governance without a state: Can it work? Regulation &
Governance, 4, 113–134.
Bundy, C. (2016). Poverty in South Africa: Past and present. Jacana.
Capps, G., & Mnwana, S. (2015). Claims from below: platinum and the politics of land in the
Bakgatla-ba-Kgafela traditional authority area. Review of African Political
Economy, 42, 606-624.
Chamberlain, L. (2015). Lonmin has broken law by dodging housing obligation. Business
Day, 1 July 2015.
Chang, R., Hevia, C., & Loayza, N. (2018). Privatization and Nationalization
Cycles. Macroeconomic Dynamics, 22, 331-361
Chua, A. L. (1995). The privatization-nationalization cycle: The link between markets and
ethnicity in developing countries. Columbia Law Review, 95, 223-303.
Child, J., & Tsai, T. (2005). The dynamic between firms’ environmental strategies and
institutional constraints in emerging economies: Evidence from China and
Taiwan. Journal of Management Studies, 42, 95-125.
Chinguno, C. (2013). Marikana and the post-apartheid workplace order. Johannesburg:
Sociology, Work and Development Unit.
City Press, 2010. Who killed Councillor Phakoe?, 24 October 2010.
Crane, A. (2011). From governance to Governance: On Blurring Boundaries. Journal of
Business Ethics, 94, 17–19.
den Hond, F., Rehbein, K. A., Bakker, F. G., & Lankveld, H. K. V. (2014). Playing on two
chessboards: Reputation effects between corporate social responsibility (CSR) and
corporate political activity (CPA). Journal of Management Studies, 51, 790-813.
Denzin, N. K. & Lincoln, Y. S. (eds.) (2005). The Sage Handbook of Qualitative Research,
2nd edition. Thousand Oaks, CA: Sage.
Department of Mineral Resources, Republic of South Africa, (2009). Mining Charter impact
report. Pretoria: Department of Mineral Resources.
Djelic, M. L., & Etchanchu, H. (2017). Contextualizing corporate political responsibilities:
Neoliberal CSR in historical perspective. Journal of Business Ethics, 142, 641-661.
Djelic, M. L., & Quack, S. (2007). Overcoming path dependency: Path generation in open
systems. Theory and Society, 36, 161-186.
Edward, P., & Wilmott, H. (2008). Corporate citizenship: Rise or demise of a myth? Academy
of Management Review, 33, 771–774.
Farlam, I. G., Hemdraj, P. D., & Tokota, B. R. (2015). Marikana Commission of Inquiry:
Report on Matters of Public, National, and International Concern Arising out of the
Tragic Incidents at the Lonmin Mine in Marikana, in the North-West Province.
Fooks, G., Gilmore, A., Collin, J., Holden, C., & Lee, K. (2013). The limits of corporate
social responsibility: Techniques of neutralization, stakeholder management and
political CSR. Journal of Business Ethics, 112, 283–299.
Frynas, J. G., & Stephens, S. (2015). Political corporate social responsibility: Reviewing
theories and setting new agendas. International Journal of Management Reviews, 17,
Glaser, B. G. (1998). Doing grounded theory: Issues and discussions. Mill Valley, CA:
Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for
qualitative research. Aldine Publishing Company.
Gray, R., Owen, D., & Maunders, K. (1988). Corporate social reporting: Emerging trends in
accountability and the social contract. Accounting, Auditing & Accountability Journal,
Greenwood, M. (2007). Stakeholder engagement: Beyond the myth of corporate
responsibility. Journal of Business Ethics, 74, 315–327.
Heath, J., & Norman, W. (2004). Stakeholder theory, corporate governance and public
management: What can the history of state-run enterprises teach us in the post-Enron
era? Journal of Business Ethics, 53, 247–265.
Hussain, W., & Moriarty, J. (2018). Accountable to whom? Rethinking the role of
corporations in Political CSR. Journal of Business Ethics, 149, 519–534.
Ilunga, Y. (2011). Why and how do companies contribute to socioeconomic development and
the provision of public goods and services in areas of limited statehood: The case of
mining companies in the Democratic Republic of Congo. Unpublished Master’s
research report, University of Cape Town Graduate School of Business.
Kourula, A., & Delalieux, G. (2016). The micro-level foundations and dynamics of political
corporate social responsibility: Hegemony and passive revolution through civil society.
Journal of Business Ethics, 135, 769–785.
Kvale, S., & Brinkmann, S. (2009). Interviews: Learning the craft of qualitative research
interviewing. London: Sage.
Lange, D., & Washburn, N. T. (2012). Understanding attributions of corporate social
irresponsibility. Academy of Management Review, 37, 300–326.
Langley, A. (1999). Strategies for theorizing from process data. Academy of Management
Review, 24, 691–710.
Langley, A., Smallman, C., Tsoukas, H., & Van de Ven, A. H. (2013). Process studies of
change in organization and management: Unveiling temporality, activity, and flow.
Academy of Management Journal, 56, 1–13.
Leibbrandt, M., Finn, A., & Woolard, I. (2012). Describing and decomposing post-apartheid
income inequality in South Africa. Development Southern Africa, 29, 19-34.
Levy, D., Reinecke, J., & Manning, S. (2016). The political dynamics of sustainable coffee:
Contested value regimes and the transformation of sustainability. Journal of
Management Studies, 53, 364–401.
Lin-Hi, N., & Müller, K. (2013). The CSR bottom line: Preventing corporate social
irresponsibility. Journal of Business Research, 66, 1928–1936.
Locke, K. (2011). Grounded Theory in management research. Thousand Oaks, CA: SAGE
Lonmin (2013a). Lonmin Sustainable Development Summary Report 2013.
Lonmin (2013b). Response to the Bench Marks Foundation Policy Gap 7 Report – Lonmin
Media Release, 31 October 2013.
Mäkinen, J., & Kasanen, E. (2016). Boundaries between business and politics: A study on the
division of moral labor. Journal of Business Ethics, 134, 103-116.
Mäkinen, J., & Kourula, A. (2012). Pluralism in political corporate social responsibility.
Business Ethics Quarterly, 22, 649–678.
Matten, D., & Crane, A. (2005). Corporate citizenship: Toward an extended theoretical
conceptualization. Academy of Management Review, 30, 166–179.
Mayntz, R., & Scharpf, F.W. (eds.) (1995). Gesellschaftliche Selbstregulierung und politische
Steuerung. Frankfurt am Main, Germany: Campus.
McMahon, T. F. (1999). From social irresponsibility to social responsiveness: The
Chrysler/Kenosha plant closing. Journal of Business Ethics, 20, 101–111.
Mena, S., Rintamäki, J., Fleming, P., & Spicer, A. (2016). On the forgetting of Corporate
Social Responsibility. Academy of Management Review, 41, 720–738.
Miles, M. B., Huberman, A. M., & Saldana, J. (2014). Qualitative Data Analysis: A methods
sourcebook (3rd edition). Sage.
Moon, J. (2002). The social responsibility of business and new governance. Government and
Opposition, 37, 385–408.
Office of the Auditor-General (2012). Report of the Auditor-General to the North-West
Provincial Legislature and the Council of the Rustenburg Local Municipality, 2012.
Pretoria: Office of the Auditor-General.
Palazzo, G., & Scherer, A. G. (2008). Corporate social responsibility, democracy, and the
politicization of the corporation. Academy of Management Review, 33, 773–776.
Piketty, T. (2014). Capital in the 21st Century. Cambridge: Harvard University Press.
Powell, E. E., Hamann, R., Bitzer, V., & Baker, T. (2017). Bringing the elephant into the
room? Enacting conflict in collective prosocial organizing. Journal of Business
Venturing [online first].
Risi, D., & Wickert, C. (2017). Reconsidering the ‘symmetry’ between institutionalization
and professionalization: The case of corporate social responsibility managers. Journal
of Management Studies, 54, 613–646.
Risse, T. (2011). Governance in areas of limited statehood: Introduction and overview. In T.
Risse (ed.), Governance without a state? Policies and politics in areas of limited
statehood. Columbia University Press.
Rowlinson, M., Hassard, J., & Decker, S. (2014). Research strategies for organizational
history: A dialogue between historical theory and organization theory. Academy of
Management Review, 39, 250–274.
Ruggie, J. G. (2014). Global governance and “new governance theory”: Lessons from
business and human rights. Global Governance, 20, 5–17.
Rustenburg Local Municipality (2015). Integrated Development Plan Review 2014-15.
Rustenburg Local Municipality.
Scherer, A. G., & Palazzo, G. (2007). Toward a political conception of corporate
responsibility: Business and society seen from a Habermasian perspective. Academy of
Management Review, 32, 1096–1120.
Scherer, A. G., & Palazzo, G. (2011). The new political role of business in a globalized
world: A review of a new perspective on CSR and its implications for the firm,
governance, and democracy. Journal of Management Studies, 48, 899–931.
Scherer, A. G., Rasche, A., Palazzo, G., & Spicer, A. (2016). Managing for Political
Corporate Social Responsibility: New Challenges and Directions for PCSR 2.0. Journal
of Management Studies, 53, 273–298.
Schrempf-Stirling, J. (2018). State power: Rethinking the role of the state in political
corporate social responsibility. Journal of Business Ethics, 150, 1-14.
Strike, V. M., Gao, J., & Bansal, P. (2006). Being good while being bad: Social responsibility
and the international diversification of US firms. Journal of International Business
Studies, 37, 850–862.
The Sowetan (2002). Editorial, 16 August 2002, p. 11.
Tinker, T., Neimark, M., & Lehmann, C. (1991). Falling down the hole in the middle of the
road: Political quietism in corporate social reporting. Accounting, Auditing &
Accountability Journal, 4, 28-54.
Whelan, G. (2012). The political perspective of corporate social responsibility: A critical
research agenda. Business Ethics Quarterly, 22, 709–737.
Whelan, G. (2018). Political CSR: The corporation as a political actor. In A. Rasche, M.
Morsing, & J. Moon (Eds.), Corporate Social Responsibility: Strategy, Communication,
Governance. Cambridge University Press.
Whiteman, G., & Cooper, W. H. (2016). Decoupling rape. Academy of Management
Discoveries, 2, 115–154.
Yin, R. K. (2003). Case study research: Design and methods (3rd ed.). Thousand Oaks, CA:
I am grateful for helpful guidance and comments from Christopher Wickert and the
other guest editors, the anonymous reviewers, as well as participants in a track at the 2016
colloquium of the European Group for Organization Studies (chaired by Steen Vallentin,
Guido Palazzo and Andreas Georg Scherer). I have benefited from financial and other support
from the African Climate and Development Initiative at the University of Cape Town (UCT),
the UCT Graduate School of Business, and the South African National Research Foundation.
I also thank Angus Kingon and Brown University for some of my time spent on this paper as
Pearson Visiting Professor of Entrepreneurship at Brown University.
Figure 1: Dynamic de-responsibilization of business and government
Figure 2: The dissipation of responsibility in business-government interactions
Table 1: Overview of data sources
# of interviews
Characteristics and focus
Phase 1 (2001-2004)
Mining companies (M)
Semi-structured interviews, lasting between 45 minutes and two hours
(approx. 2500 pages of text - verbatim transcriptions from audiotape).
Focus was on companies’ approach to CSR and social conditions
around the mines.
Business associations (A)
Consultants / academics (C)
Government (national / local) (G)
Traditional authority (T)
Phase 2 (2007)
Mining companies (M)
Semi-structured interviews, lasting 30-60 minutes (approx. 20 pages
of text - verbatim transcriptions from audiotape). Focus was on
opportunities and challenges of collaboration between companies,
government, and other actors.
Consultants / academics (C)
Government (local) (G)
Phase 3 (2014-2015)
Mining companies (M)
Semi-structured interviews, lasting 45-60 minutes (approx. 200 pages
of text - verbatim transcriptions from audiotape). Focus was on
conditions contributing to the Marikana Massacre, including
mineworkers’ housing and social conditions around mines.
Consultants / academics (C)
Government (local) (G)
4 visits to the study
area (2002, 2003, and
These visits involved face-to-face interviews, informal discussions;
observation in some meetings; site visits to mines and surrounding
settlements. They advanced insights into employer-worker
interactions and workers’ living conditions.
* In the text, interviews are labelled by the interview phase, the interviewees’ background, and a unique identifying number
within this set. E.g., 2M1 represents the first mining company representative interviewed in phase 2.
** Number in brackets indicates interviewees who were interviewed also in previous phases.
Use in Analysis
Historical accounts of the
9 books and 11 articles
Insights into the historical context of the
including white papers,
draft policies, and
promulgated policies and
Approx. 500 pages of text, focusing mostly on the process
leading up to the promulgation of the Mining Charter in 2002
Analysis of business and government
practices and interactions in negotiated
36 newspaper or magazine articles focused on the process
leading up to the promulgation of the Mining Charter, as well
as subsequent controversies or agreements related to
implementation of or amendments to the Charter; 22 articles
focused on labour conflict and social conditions around the
mines in the study area
Analysis of business and government
practices and interactions in negotiated
rule-making and in provision of public
Final report and
transcripts of the
Marikana Commission of
Final report and transcripts (approx. 500 pages) focused on
labour relations and worker housing, focused on the lead-up to
the Marikana Massacre
Insights into labour relations in the
mining industry, worker housing, and
workers’ living conditions in informal
CSR/Sustainability Reports from Anglo Platinum, Impala, and
Lonmin, from 1996 onwards; focus on Lonmin reports from
2002 to 2012 (approx. 1000 pages)
Insight into companies’ perspectives on
and approach to social issues around the
6 reports (approximately 150 pages of text), with a focus on
NGO reports focused on Lonmin before and after the Marikan
Insights into NGO analysis of business
practices and apparent irresponsibility
11 reports (approx. 500 pages of text)
Insights into social conditions around
the mines in the study area
*** Where applicable, these are referenced in the text and included in the reference list.