Corporate social responsibility and development: An anthropological perspective

Abstract

Some analysts suggest that corporate-driven social responsibility initiatives offer a new, and potentially bright, prospect of addressing global poverty and underdevelopment effectively. There is a growing academic literature that examines this proposition critically, often rehearsing debates about the successes and failures of the international development programmes of the second half of the 20th century. From an anthropological perspective, however, the most useful question to ask is not whether corporate social responsibility (CSR) initiatives will succeed in fulfilling the promises made by their proponents but rather what the implications and consequences –often unintended – might be of expecting business corporations to become significant engines of development. This article examines current debates about CSR, and the emerging CSR discourse and apparatus, as a prelude to asking ‘What does CSR do?’.

1. Corporate social responsibility and anthropology

Corporate social responsibility (CSR), a global business trend that is becoming increasingly prevalent in South Africa's private sector, presents itself as an important topic for anthropological inquiry. As a branch of the social sciences more interested in process than product, and more curious about the broad significance of a phenomenon than its perceived ‘success’ or ‘failure’, anthropology is well suited to exploring the many dimensions of CSR in South Africa today. Using methods that view the issue from a variety of angles, anthropologists who study CSR investigate not only the purported aims and objectives of particular interventions but also the ideologies underlying these interventions, the discourses used to sell them to a range of ‘stakeholders’, and the power dynamics at all levels that make this component of current corporate practice simultaneously transforming and ineffectual. In short, anthropologists working on the phenomenon of CSR are able to focus not simply on whether companies are doing what they claim to be doing, and if not why not, but rather on the intended and unintended consequences of such endeavours, and on how these practices are related to other attempts at poverty alleviation, social improvement and economic development.

Anthropologists use a range of methods to obtain and analyse information. Using a combination of discourse analysis, ‘multisited ethnography’ and participant observation, they can explore CSR from a variety of perspectives, leading to a more critical understanding of this trend than can be achieved through conventional forms of interview-based inquiry. Indeed, it seems likely that by exploring CSR from different angles one might arrive at a conclusion that this concept covers a wide span of outreach efforts sponsored by companies, ranging from ostensibly contractual relations with non-employees perceived to have ‘rights’ to compensation for the damaging effects of the companies' activities to more traditional kinds of philanthropic activities. CSR might then start to look less like a single phenomenon and more like a rhetorical umbrella for a host of activities that do not fall within a company's ‘core business’. Such analysis has the potential to displace the dominant debate surrounding CSR's effectiveness and to open up a critical space for considering other outcomes and impacts of CSR.

2. Recent academic literature on CSR and development

Recent academic literature has moved very quickly to a critical evaluation of CSR but seems distinctly unclear about what its critique of the phenomenon is based on. Contributors to the recent discussion in International Affairs (2005), for instance, are unanimous in their judgement that CSR ‘in its present form’ does not meet the development goals spelt out, in 2000, in the United Nation's Global Compact (Kuper, 2004), but are awkwardly undecided about why this should be so. Is it because there is insufficient insight into what CSR is, or could be, and inadequate understanding of the impediments to a successful marriage between CSR and development, or is it because it is simply naïve to expect business corporations to undertake the task of developing the Third World? Similar questions are also posed in recent issues of Development (2004) and Development in Practice (2005).

Of the International Affairs contributors, Blowfield (2005) and Jenkins (2005) come closest to the latter argument, with Jenkins going so far as to say that there are ‘inherent features which limit CSR's ability to address poverty’, the most important of which is that ‘there is no reason to suppose that [there will be any relationship] between [a] company['s] actions to reduce poverty and [its] profitability’ (2005: 540). On the other hand, Frynas ends his devastating account of the seemingly wilful failure of Shell's community development programmes in Nigeria with a list of constraints on the implementation of these CSR initiatives – including ‘failure to involve the beneficiaries of CSR’, ‘lack of human resources’ and ‘failure to integrate CSR initiatives into a larger development plan’ (2005: 588) – as if paying due attention to overcoming these constraints would bring about an outcome in other places different from the one he observed in West Africa. Even more ambitiously, the editors of the collection suggest that analysts should consider approaches to CSR that go beyond the ‘business case’ (Blowfield & Frynas, 2005; see also Fox, 2004), as if it were possible for academics to craft an argument persuasive enough to redesign the fundamentals of capitalist logic in the interests of ‘real’ development.

The main difficulty here is one to which the anthropologist James Ferguson (1990) referred in his study of the rural development industry in the 1970s and 1980s. In that period, those who were for the idea of rural development as a viable enterprise and those who were against it were both guilty of conflating two distinct meanings of development – ‘development’ as a process of transition to a modern economy and ‘development’ as a moral injunction to fight poverty. As Ferguson noted, the pro-development camp saw no disjuncture between these meanings (although the reality of the disjuncture is all too apparent), while the anti-development camp perceived this disjuncture in existing, capitalist-driven development practice, but believed that both dimensions of development could be attained by alternative, non-capitalist means (if only it were possible to determine how to put these means into practice).

As Ferguson argued, the conflation of development as transition and development as moral imperative was linked to the fact that both schools of argument agreed that the key question to ask was how far rural development interventions met their own advertised objectives. In this respect, one might say, all the participants were party to a collective ‘misrecognition’ (Bourdieu, 1990, 1991) of the arbitrary character of this question. Working from a foundation of shared belief in this respect, they were unable to see that there were other questions that could more usefully have been asked. Academics in favour of development projects argued that they did meet these objectives in some measure, and that where they did not this was because of technical shortcomings in development planning and an insufficient understanding of the practical constraints on development in specific circumstances. Their opponents, who drew liberally on the ideas of neo-Marxist and dependency theorists, dismissed the whole idea of capitalist-driven development projects ever addressing rural poverty successfully as a contradiction in terms, implying that these projects were always bound to fail. The weakness of the whole debate, as Ferguson saw it, was that both sides used exactly the same findings about development interventions to arrive at different, and indeed opposed, conclusions. The presumptions with which both sides started their inquiry had a profound effect on their respective interpretations of the evidence they found.

For this reason, Ferguson characterised the academic debate about rural development in the 1980s as first and foremost an ideological debate, which could not be resolved in its own terms simply by scrutinising the outcomes of development projects. It is thus ironic that, 15 years later, the same ideological differences are apparent in the CSR debate, with the same attempt to suggest that these differences are amenable to arbitration on the basis of empirical evidence. Newell (2005: 586) declares forthrightly that ‘most CSR initiatives are not intended to tackle questions of poverty and social exclusion. They aim at less ambitious goals of performance enhancement and image management’ (emphasis added). If this is true, what is the point of the plea by Blowfield and Frynas (2005: 506) that more academic research is necessary to ‘explore whether CSR interventions are likely to fulfil their promise’?.

3. CSR discourse on development

Ferguson redirected analysis of rural development projects away from the question of whether or not they met the objectives advertised by the development industry itself. He did so because the question about whether these projects fulfilled their planners' promises was indeed naive, but not for the reasons given by their critics. Indeed, whether one is referring to development planners or CSR practitioners, it is surely misleading to suppose, along the lines Newell suggests, that they are systematically engaged in doing one thing – managing the corporate image – while claiming to be doing another – tackling poverty. This conclusion would render CSR a fraud, and its advocates and practitioners nothing more than camouflage experts.

A critical anthropological approach would take CSR discourse more seriously than this, while at the same time recognising that there is a world of difference between intentions and effects in any activities as complex as those associated with CSR. As Hamann & Kapelus (2004) note, CSR discourse in the form of the ‘business case for CSR’ is not simply a smokescreen that hides the ‘real’ interests of business corporations; rather, it is an orderly system of knowledge and practice that embodies particular ways of interpreting and acting on the world. For this reason the relationship between business corporations' interests, which may change rapidly in different contexts, and the business case for CSR, which has the stability and coherence of any system of knowledge and practice, is a complex one. Certainly one cannot read the changing interests of particular corporations directly into or off the CSR discourse.

Development discourse up to the 1980s constructed the contexts in which planners intervened in highly specific ways. In this discourse, development interventions always took place in areas in which the inhabitants were poor because they had not yet been incorporated into the modern economy, but which were nevertheless blessed with national governments that were, in principle, neutral and fair, even if they were lacking in capacity. The point of this characterisation was to establish that these areas were appropriate for the kinds of intervention development experts could make. They could build roads into isolated regions, teach subsistence farmers to produce for the market, and devise national development policies which they could advise governments in Africa and elsewhere to implement. The vision embodied in development discourse was important because it made certain kinds of interventions – those that development agencies and planners could undertake – not only possible but also, seemingly, vitally necessary. If the target areas for development had been construed in other ways – for instance, as areas that had already been incorporated into the modern economy on highly unequal terms – this would certainly have made more sense within the dominant academic discourse of the time, but would have rendered development planners incapable of meaningful intervention. It was never within their scope, as individuals or specific project teams, to address the political issues underlying global inequalities.

Since the late 1990s there has been a concerted move by international and national development agencies to add development to the list of business corporations' social responsibilities. In the process, some aspects of earlier development discourse have been carried over into the business case for CSR, and others have been radically modified. The concern to depict poverty as an original condition and a matter of where people happen to live in the world (in the South rather than the North) remains at least constant, and may even have intensified as business corporations take over from governments as the main agents of development. In this regard, Fig (2005) shows that South African businesses are more resistant to the actual concept of ‘corporate social responsibility’ than they are to the notion that they have a part to play in development. This is because they are averse to the implication that they are in any way responsible for exploitation and oppression – for making people poor – during the apartheid era.

On the other hand, however, precisely because responsibility for development is being shifted from governments to businesses, the CSR discourse has no need to rely, as did earlier development discourse, on the notion that poor people endure their poverty under generally benign and fair-minded, even if inefficient, governments. Development experts had to work with and through the institutions of the state, and therefore could not afford to see them as chronically corrupt. But business corporations make their CSR and development interventions where there are gaps in the state's efforts, and thus have reason to highlight the extent of government corruption.

Present CSR discourse also differs from earlier development discourse in the terms it uses to construct people's entitlement to be ‘developed’. Given the importance of the state as engine of development in development discourse, it was the state's citizens who fell into the category of beneficiaries of expert interventions. Even where specific development projects targeted only some citizens – those who lived in the immediate vicinity of the project, for instance – a noticeable aspect of the accompanying discourse was invariably that the project would serve an exemplary function, as a model of behaviour change and ostensible reward to be followed by others (Fischer, 1988). But CSR discourse deploys a new vocabulary of categorisation in which the notion of entitlement to development flowing from citizenship is largely replaced by ideas of ‘stakeholding’ and belonging to ‘host communities’.

Part of what this signifies is a shift in the moral basis of development interventions. Development acquired a new dimension of morality in the 1960s and 1970s, partly as a result of the dependency theorists' arguments about the causal link between ‘development’ in one part of the world and ‘underdevelopment’ in another (e.g. Frank, 1969). As noted above, this dimension was manifest, in practice, in calls for general social ‘upliftment’ and an all-out fight against poverty as the core of the transition to modern economies; but the new business discourse of development sets out to limit what can be expected of corporations. It does this not by jettisoning all notions of fighting poverty and uplifting the poor, but by adding a new, legalistic notion of who the poor that are entitled to development might be. They are the stakeholders or the members of ‘host communities’, whose entitlement stems from the fact that they are directly affected by, or in some way involved in, the core business of the corporations concerned.

The relationship envisaged in these terms is a quasi-contractual one. Those people who possess something, or some characteristic, that business wants are entitled to development in exchange. Host communities have oil or other minerals under their land, or timber growing on it; stakeholders possess some inherent quality – for instance, of being women, or disabled, or black – of which businesses want more, and are willing to acquire by paying a premium, in the form of general education or specific training. The new limits of this contractual morality are, of course, abundantly apparent. Those who are not members of host communities or stakeholders have no claim to recompense in any form. The fact of merely being a citizen is no longer a guarantee of being noticed for any kind of development.

CSR discourse is therefore in several significant respects a modification of earlier development discourse. The way it has shifted reflects a changing logic in conceptualising and practising development, although the character and significance of this shift is unlikely to be fully apparent to those involved in implementing CSR as an immediate task let alone, of course, to have been deliberately orchestrated by them. Discourse is, as Foucault (1979) explained, in large measure ‘authorless’, emerging out of a myriad encounters between intentions and structural constraints over which no one has final control. Yet the power of discourse is that it appears as systematic and coherent, specifying the only reasonable and possible way in which people can act in order to realise their intentions.

The intentions of CSR practitioners will surely vary across a wide spectrum, from ‘image management’ to wanting to ‘do one's bit’ to ‘make poverty history’. Newell's insistence that CSR is ‘not intended to tackle poverty and exclusion’ may therefore be accurate in some instances, but cannot be adduced as an iron(ic) law of CSR. On the other hand the characteristics of the CSR discourse uncovered above clearly make it impossible that any of this range of intentions could be realised in practice in straightforward fashion. Image managers may yet, despite their intentions, end up benefiting some women or some of the disabled, and anti-poverty warriors would have to contend with the fact that to develop stakeholders leaves non-stakeholders untouched.

A discourse is neither an ideological smokescreen that needs to be blown away nor a simple statement about what will happen that can be taken for granted. Since discourse mediates in complex ways between intentions and effects, anthropological inquiry would give careful attention to CSR discourse as a phenomenon in its own problematic light. The discussion above constitutes no more than a preliminary suggestion about how development discourse has been revised in the era of CSR. It is sufficient to provide some pointers to the character of the practices involved in CSR and the effects these practices might have.

4. Complex systems

The development industry of the 1980s constituted, as Ferguson explained, a relatively homogeneous ‘apparatus’. It was dominated by a handful of international and national agencies that provided most of the funding for development projects, and the personnel for these projects was a surprisingly small cadre of experts who moved from one project to the next. Small wonder, then, that particular categories of development undertaking – such as rural development projects across south-central Africa – were characterised by a marked degree of sameness. They were financed from the same sources, implemented by the same people and undertaken in accordance with the same set of ideas about how development should be conducted.

The business case for CSR may not yet have congealed into quite as rigid a form as the old development orthodoxy, but it provides an easily understandable route map for CSR practitioners to follow. On the other hand, however, other aspects of the emerging CSR apparatus are not now, and are unlikely ever to be, as tightly organised as their development counterparts were. Organisations such as the World Bank, USAID, the Canadian International Development Agency (CIDA) and Britain's Department for International Development (DFID) play an important role in linking development and CSR, but it will obviously never be as dominant a role in this field as it was when they were the main agents of development. This is precisely because a large part of the responsibility for development now rests on the shoulders of business corporations, which are far more numerous and more varied than the agencies themselves. Moreover, CSR is too new in its present form, for a cadre of experts to have emerged and be able to circulate among business corporations. In many instances, one gathers, company employees who were trained in totally different fields find themselves press-ganged into taking responsibility for CSR, and they bring many diverse kinds of competence to bear on their new tasks.

It is striking how little attention academics commenting on CSR pay to these issues (Moser, 2001 is a notable exception). It is as if, in their rush to find out whether or not CSR meets its advertised objectives, they feel they can afford to treat the question of how corporations try to meet these objectives as insignificant. This mistake springs from taking the notion of a corporation as a single legal personality too literally, thereby assuming that a corporation's many employees have a uniform interest.

Business corporations are powerful entities, at least in comparison to the people they intend to develop. But this power does not mean that their stated intentions are simply processed though their structures to achieve the desired results. Kapelus (2002) is one of the few who explicitly treats the internal structure of a multinational corporation as a problem for analysis, showing that, in the case of Rio Tinto Zinc's operations in South Africa, there was a significant difference between the definition of ‘stakeholder’ offered by the London-based directorate (‘all the people who are affected by our operations’) and the definition actually used by managers on the ground near Richards Bay (only the subjects of the chiefdom in which mining actually took place). Kapelus draws attention to this discrepancy in order to make a point about image management at the highest level of a multinational corporation, but he might as plausibly have shown that even the board of directors of RTZ does not have the power to compel straightforward compliance with its injunctions. Indeed, if one takes the whole CSR apparatus, including the development agencies and the business corporations, as well as governmental and non-governmental organisation (NGO) watchdogs into account, it is by no means obvious where the final power to propose and dispose actually lies. Saying this is not an attempt to go easy on multinational corporations, but rather a call to recognise the importance of considering the CSR apparatus as a ‘complex system’.

As Zabusky (2002) shows in her insightful analysis of the internal working of the European Space Agency (ESA), complex systems are the hallmark of emerging transnational endeavours. For scientists who design and build components for European spacecraft, the old way of understanding organisations – as a single authority issuing instructions from an unambiguous position of power – is clearly obsolete. The formal, hierarchical structures of the ESA can be seen as, at best, a series of points at which power is concentrated, usually only temporarily. Fixed notions of authority are continually obliterated and refashioned. In their efforts to ‘get things done’ scientists spend much of their time improvising stable vantage points where they can escape, briefly, from the flow of power. The sense they have of being submerged in this flow stems from the character of the ESA, which consists of numerous laboratories scattered across many states, the governments of which have ambiguous agendas with regard to space exploration. Since each government contributes revenue to the overall project, each wants an equitable return to national industries in the form of contracts for manufacturing high-tech components. On the other hand, the ESA serves as a flagship of European unity. Moreover, space exploration is a ‘big science’ project that no European country can afford to undertake alone.

The author would venture to suggest that some of Zabusky's analysis is pertinent to the case of the transnational, and fragmented, CSR apparatus. This is probably true not only with respect to the apparatus as a whole, in which each set of participants may follow a different, and ambiguous, agenda, but also within particular companies – which is the point that Kapelus explores in preliminary fashion with respect to the multinational RTZ. How do CSR practitioners actually work within these complex systems, where power is so diffused that there are times when nobody seems to have the final authority to do anything at all? Do they, like the ESA scientists, have to improvise stable vantage points from which to get things done, and do they too improvise these discursively – fashioning them by talking about them (in which case ‘the business case for CSR’ becomes a doubly crucial subject of study in its own right)?

Zabusky comments in some detail on the cynicism that appears to be a concomitant of the need to improvise in this manner. ‘Discourses of cynicism (among the scientists) can be understood as cultural expressions of the contradictions inherent in transnational processes, contradictions that arise as centres are continually undermined and power seems always to impinge on people from somewhere else as they travel in directions not of their own making’ (2002: 137). In this regard it is interesting that Frynas found ‘industry insiders’ volunteering a number of highly cynical comments to him about Shell's efforts in Nigeria (including the observation that ‘CSR is a waste of time’ – 2005: 582). But these should probably be interpreted as indicating not the ‘inside truth’ about public image management, but rather the fact that CSR practitioners are, like other people, able to work within two or more different, and even diametrically opposing, discourses at the same time, provided they keep separate the contexts in which each is deployed as a means to interpret the world. Since there are thousands of scientists who are also creationists, for example, and millions of Christians who also give credence to witchcraft, one must expect to find CSR practitioners who adhere to the business case for CSR and, at the same time, espouse arguments that are highly sceptical of it [indeed, Litvin's (2003) account of Shell's misadventures in Nigeria brings out admirably the ambivalence with which its employees approached the issue of development].

Recognising that the CSR apparatus is a complex system – more complex than its earlier development counterpart – is an essential first step towards analysing what CSR initiatives actually do, rather than continually rehearsing the old question of whether they do what they are claimed to do. As Zabusky observes, anthropological methods of field research, including participant observation (actually accompanying the scientists, or the CSR practitioners, as they pick their way through the complexities) are far more productive than questionnaires and formal interviews (the stock-in-trade of the existing literature) to make sense of the phenomenon.

5. What does CSR do?

Given that development discourse persisted in constructing south-central Africa as ‘less developed’ – a vast region of subsistence farmers whose productive activities were limited by their continuing adherence to ‘tradition’ – the projects this discourse spawned could never address the issue of rural poverty successfully. This was because, as Ferguson (1990) showed, this discourse may have been a way of getting certain things done, but it was wildly inaccurate in its depiction of the causes of rural poverty in the region.

On the other hand, development discourse had some profound effects, which worked themselves out ‘behind the backs’ of the experts whose commitment and sincerity Ferguson saw little reason to doubt. Rural development projects in south-central Africa had the unintended consequence of enhancing the exercise of state power and bureaucratic control in remote regions, and of ‘depoliticising’ both the exercise of this power and the development process itself.

No one has yet conducted an anthropological study of the CSR apparatus comparable to Ferguson's analysis of the development industry, so any propositions about what CSR actually achieves will have to be appropriately provisional. But we have already seen that CSR discourse preserves some important continuity with that of development, particularly as regards the causes of poverty in areas into which business corporations intrude. One could predict that the CSR apparatus is already deflected from fulfilling its stated promises by this fact alone, on the grounds that this apparatus works as the same kind of ‘anti-politics machine’ as the development industry did. On the other hand, however, the shifts evident in CSR discourse – from working with the state to working in the spaces in which it fails to deliver, and from defining ‘development’ beneficiaries as ‘citizens’ to defining them as ‘stakeholders’ – create opportunities for new unintended consequences to flow from CSR interventions.

Throughout south-central Africa rural development projects followed a similar path in their quest to increase food production and conquer poverty. They rationalised land use in target areas, setting aside dedicated lands for cultivation, grazing and residential purposes. Indeed, concentrating people into planned residential areas was the key mechanism for enhancing bureaucratic state control in the countryside, and was employed as much in Tanzania's ujamaa (villagisation) programme, for instance, as in South Africa's ‘betterment’ projects in the Bantustans. The unintended effect of increased bureaucratic control was, moreover, followed by other unintended consequences in the form of widespread resistance to development programmes wherever they were implemented. Since implementation of ‘development’ involved disrupting the lives of millions of people, the process provided these people with a basis of common experience from which to resort to joint action. With hindsight, one can see this as having been one of the weaknesses of the state's efforts, together with those of the development agencies, to bring ‘development’ to as many citizens as possible.

Seen in this light, the current reformulation of the basis of entitlement to development – by linking it to ‘stakeholding’ – has the effect of being a clever response to this earlier weakness. Stakeholders and host communities are to be treated differently from those who have no claim to involvement in particular interventions, and the CSR discourse happens to provide a legalistic, and therefore also a moral, justification for why this should be so. Business corporations have an obligation to all who are, or could be, affected by their production activities, but – by implication – to no one else, and there is no sense whatever in which the fulfilling of quasi-contractual obligations to stakeholders is represented as having a demonstration effect elsewhere (for instance, on less fortunate neighbours).

The possible ramifications of this are impressive. On one hand, CSR interventions have the effect of deflecting hostility away from the state, and of course from the business corporations, by turning non-stakeholders against stakeholders in a desperate scramble for resources. Frynas (2005) mentions a tragic example from Nigeria in which the residents of one village a short distance from an oil pipeline burnt another village to the ground, because it had been selected by Shell as a ‘host community’, entitled to community development, given that the pipeline happened to cross its residents' lands.

On the other hand, one must also take account of the fact that the tenets of CSR discourse turn out to be one particular refraction of the central tenets of capitalism in the current era. ‘Casino capitalism’, as Strange (1986) appropriately dubbed the system's current manifestation, turns on the notion that there is no conceivable connection between effort and reward, especially for the poor (see also Comaroff & Comaroff, 2001). Success within this system is a matter of good fortune, of chance rather than effort or achievement, and this is a notion that is presented repeatedly across numerous different registers – as, for instance, in the remarkably high-stakes competitions of pure chance that are associated with virtually every product one cares to think of, and that are advertised ad nauseam in all the communications media by precisely the business corporations that are also engaged in CSR activities. Is it in the least surprising that CSR discourse turns out to be part of a much broader discourse which, by dint of repetition in a variety of forms, establishes its interpretation of the world as one that can be questioned only with considerable difficulty?

One can argue, moreover, that the complex, fragmented organisation of the CSR apparatus will actually contribute both to this discourse and to its acceptance. Comprising a myriad business corporations and development agencies and NGOs and governments, this apparatus is likely to deliver a bewildering variety of benefits in unpredictable ways – a community development project here, an initiative aiming to advance the situation of one or other category of stakeholders there. If a settlement is not a ‘host community’ today, it may nevertheless turn out to be the custodian of some hitherto hidden resource desired by business tomorrow. And if some indigent people are not construed as stakeholders immediately, they may yet aspire to this happy condition at some point in the future, when the wheel of fortune has turned in their favour. The odds of this happening are probably, in the cold light of business reality, rather slim, but the beauty of the convoluted CSR apparatus is that it continually holds out the possibility that the poor may – one day – get lucky.

Additional information

Notes on contributors

John Sharp

Professor of Social Anthropology and Head, Department of Anthropology and Archaeology, University of Pretoria.