Abstract
This paper seeks to apply a critical realist lens to the analysis of discourse. Our central argument is that the most fruitful way to make sense of discourse is to treat it as an empirical object of inquiry alongside extra-discursive phenomena. In making this argument, we consider two recent discourses, the ‘knowledge-based economy’ (KBE) and ‘shareholder value’ (SV) and demonstrate that, while the KBE discourse gained much more attention, SV discourse had much more significant material outcomes. We suggest that research on the potential causal powers of discourse should move away from an exclusive focus on the discursive and pay attention to actors and institutions and to extra-discursive conditions and constraints.
Introduction: Discourse Analysis and Critical Realism
In the past decade there has been growing interest in discourse analysis (DA), particularly from within the fields of management and organizational studies (see Grant, Hardy, Oswick, & Putnam, 2004). This is an approach utilized principally by post-structuralists and informed by social constructionist ontology and epistemology. Phillips and Hardy (2002, pp. 2–6) sum up what is distinctive about this form of DA as follows: ‘without discourse there is no social reality…discourse analysis views discourse as constitutive of the social world’.
DA has received a good deal of attention, particularly within the field of critical management studies, for giving primacy to the discursive (Fleetwood, 2005; Thompson, 2004). What Fairclough (2005) refers to as extreme versions of social constructionism tend to explain both causes and effects of discourse with reference to other discourses, resulting in any contingent effects of texts being collapsed into the all-encompassing theoretical categories of discourse and thus rendered opaque. Critics of DA have often proceeded from critical realist (CR) assumptions. In contrast to the social constructionist foundations of DA, CR holds that there is indeed an extra-discursive social reality that exists independently of knowledge of it.
Debates between advocates of CR and DA have often become bogged down in apparently intractable disputes conducted at a high level of theoretical abstraction. One unfortunate by-product of the general debate is the widely held assumption that DA deals exclusively with the discursive and CR exclusively with the extra-discursive. At its most basic, the discursive comprises talk and text, and a specific discourse can be defined as ‘an interrelated set of texts, and the practices of their production, dissemination, and reception’ (Phillips & Hardy, 2002, p. 3). This definition can be broadened to include all ‘linguistic and other semiotic elements…of the social’ (Fairclough, 2005, p. 916), which might encompass media transmissions, visual symbols, body language and so on.
Contrary to what one might suppose based on the sometimes acrimonious debates between CR and DA, the former does not a priori privilege the non-discursive or imply that discourse is in any sense less than real. Fleetwood (2005, p. 200) argues that ‘although critical realists claim there is more to the world than discourse, this should not be taken to suggest that they think discourse is irrelevant: far from it’. Within a CR framework, the discursive is categorized as being ‘ideally real’, by which is meant that it is real to the extent that it has causal efficacy, regardless of the fact that it has no structural properties (Fleetwood, 2005, p. 199). If we conceptualize discourse in this way, it is possible to see discourses as ‘generative mechanisms’ with ‘performative potential’ – the potential to shape extra-discursive outcomes (Reed, 2000, pp. 528–9). To argue that discursive phenomena are real, however, is not the same as the argument that everything that is real (or even important) is discursive. In referring to extra-discursive we mean those social phenomena which are real, but do not reside purely in the discursive realm, including social structures, systems of economic regulation, labour markets and processes of government. From a critical realist perspective the non-discursive encompasses those phenomena which can best be classified as ‘socially real’ (Fleetwood, 2005, pp. 201–2).
We recognize that the boundary between the discursive and the extra-discursive is not a clear one, particularly in an empirical sense. Analytically, however, the maintenance of such a distinction is important in that, unless discourse is defined so broadly as to encompass everything, the very field of DA must rely on this distinction. Indeed it is difficult to see how, without such a distinction, we can explain processes of social and economic change. Fairclough (2005) makes this point very effectively when he argues that, unless we distinguish between the discursive and the extra-discursive, we are restricted to analyses which explore relations between and within discourses, but which cannot engage with relations between discourse on one hand and structures and agents on the other. These points are consistent with more materialist readings of Michel Foucault that reject ‘discourse determinism’ in accounts of social change, seek to examine relations with the extra-discursive within broader discursive formation, and recognize that an understanding of how the discursive and the extra-discursive interact is central to developing an account of social change (see Law 1994; Pearce & Woodiwiss, 2001).
If we are to develop understandings of the role of discourse consistent with a critical realist view, we need to move from the lofty heights of meta-theoretical debates and to look more closely at discourses in particular contexts. There are a number of ways this could be approached, but we need to make clear that any approach that focuses purely or mainly on the internal relations and dynamics within texts fails to meet the criteria of examining the interactions between the discursive and extra-discursive.
One compatible approach draws on institutional theories to focus on how discourse or language shapes the reproduction of practices and structures within social and organizational fields (Suddaby & Greenwood, 2005). Given that practices are seen as emerging and persisting because they provide legitimacy and resources, such frameworks are useful both in demonstrating the potential of discourse to produce change as well as enabling more careful and contested accounts of language, purposeful actors and their interests. Of particular relevance to this paper is Green, Babb, & Alpaslan’s (2008) study of the use of rival rhetorical strategies (with rhetoric acting as a specific, purposeful and persuasive resource) in a clash of institutional – stakeholder versus shareholder – logics connected to the control of the modern corporation.
However, given our aim of contributing to the limited amount of recent work that has sought to bring together insights from CR to bear on the analysis of discourse, our starting point is the attempts by Fairclough (2005) and Jessop (2004) to draw on insights from CR in developing a combined framework of critical discourse analysis (CDA) and cultural political economy (CPE). In particular, they explore the role of discourse in organizational and system-level change, and apply this framework to the supposed emergence of the knowledge-based economy (KBE) as a ‘hegemonic discourse’.
We regard their work as important and fruitful, particularly in developing a systematic way of analysing the emergence, diffusion and sustainability of discourses – but we argue that it has limitations that reveal some useful general lessons. Our view is that Fairclough and Jessop overestimate the significance of KBE, in part, because of excessive focus on DA. In contrast, a focus on the analysis of discourse in terms of the production and use of texts in particular social contexts (Jones, 2004, p. 97) indicates that, during the 1990s and 2000s, a much less obvious discourse – that of shareholder value (SV) – was more significant and provided vocabularies of motive and mechanisms for the diffusion of a range of business practices that led to a major restructuring of capitalism. This in turn had very significant impacts on the organization of production, on working lives and on the distribution of wealth, and legitimized practices which were central to the global financial crisis. The primary objective of the paper is to develop a more credible account of the interplay of the discursive and extra-discursive, paying particular attention to the definition and mobilization of interests among corporate agents and the capacity to embed new practices in key socioeconomic spheres. The second objective is to contribute to substantive understandings of dynamics and shifts in capitalist political economy, and to the growing number of empirical considerations of the nature of financialized capitalism.
Critical discourse analysis and cultural political economy
Fairclough is exclusively a discourse analyst, best known for his work on the languages of New Labour in the UK (Fairclough, 2000). In his more recent focus on organizations, he begins by setting out a critique of strong social constructionist perspectives. The key point of his critique is that social constructionists work within a ‘flat ontology’ that does not distinguish between processes and underlying structures. This could be put another way – that the only context considered for micro-level discourse is meta-level discourse. In the place of post-structuralist intertextualities, Fairclough (2005, p. 924) puts ‘interdiscursivity’, which focuses on the tensions within and between linguistic and semiotic systems (orders of discourse) that generate ‘texts’ (the discursive element of social events), in the wider network of social practices.
However, some of the same problems recur in Fairclough’s own work. In a detailed and perceptive critique, Jones (2004) argues that Fairclough, while formally correct in seeking to explore the dialectical relations between discourse and social structures, fails to resolve the problem of the movement from the ‘constructive’ to the ‘constitutive’ effect of discourse. Like Jones, we believe that the CDA/CPE approach is predisposed to overemphasize the discursive relative to the extra-discursive. Though concerned with relations between discourse and extra-discursive elements of the social (2005, p. 924), Fairclough is far more comfortable with the former than the latter. This is, perhaps, not surprising. As Jones (2004, p. 98) notes, CDA saw itself as part of the turn to discourse in the social sciences. While the framework contains formal acknowledgement of the non-discursive conditions of possibility and the requirement for discourses to meet the strategic intent and interests of social agents, these remain underdeveloped. Fairclough (2005, p. 916) refers to the doubly relational character of DA – dealing with relations between discourse and other elements of the social and between discourse and linguistic elements of social events. However, as we shall argue, the emphasis is primarily on the latter, on semiosis and interdiscursivity in a way that privileges discourse and tends to assume that if a discourse is hegemonic it has causal powers.
Jones illustrates the problem with respect to Fairclough’s (1992) claims concerning the growth and pervasiveness of ‘enterprise culture’. The extension of market discourses is taken as coterminous with changes in market relations, yet commodification in discourse may be little more than ephemeral fads whose ideological function is to obscure actual changes. Similar problems can be found in some of the brief illustrations Fairclough (2005) uses to provide an alternative account of organizational change to post-structuralist theorists. A good example of the weakness is Fairclough’s view that Boltanski and Chiapello’s (1999) identification of a ‘new spirit of capitalism’, in discourse analytical terms, has produced a change in orders of discourse within business organizations, leading to changes in the styles and practices of managers. This view of a self-organized, mobile, though less secure network society is derived from an analysis of management texts – a sample that is ‘relatively small, and does not distinguish between local or translated works, or discuss relative sales or penetration’, and more importantly, ‘no strong evidence is advanced for the influence of this literature in French society at large’ (Budgeon, 2000, p. 156). Similar objections could be made to his well-known account of New Labour (Fairclough, 2000), in which the discourse of the ‘third way’ is incorporated into a hegemonic strategy for change, recontextualized from the domestic to the international sphere, and operationalized through policy formulation and implementation. However, all the available evidence indicates that this discourse was never successfully articulated, was quickly dispensed with and had little influence on New Labour’s technocratic and pro-market policy and practice (N. Thompson, 2002).
However, it is not or at least it is no longer true, as Jones (2004) asserts, that Fairclough and CDA analyse discourse ‘without having to take the trouble to look concretely at the economic and political context of discourse (Jones, 2004, p. 120). The focus is on how interdiscursivity can produce relatively durable discourses which mediate the relations between process and structure (Fairclough, 2005, p. 920). In order to operationalize these general arguments, emphasis is placed on ‘the operation of evolutionary mechanisms of variation, selection and retention that shape the relationships between semiosis and social structuration’ (Fairclough, Jessop, & Sayer, 2002, p. 7). Fairclough and colleagues do set out a framework and set of propositions derived, in part from CPE, that suggests a ‘methodology’ consisting of stages through which the origins and operation of discourses can be understood. Combining the propositions and stages, we might get the following framework. In the extra-discursive realm particular organizational or societal structures are hegemonic, but come into crisis for internal and external reasons. New discourses emerge that re-articulate elements of existing discourse (texturing). Fairclough (2005, p. 932) provides little guidance on the role of the extra-discursive context, other than: ‘new discourses which may contribute to changes in organisational structures have their own conditions of possibility in the structures of organisations, the strategies of social agents, the habitus of social agents, and so forth’.
The next stage is when a particular discourse becomes hegemonic. This happens when emergent phenomena in social processes and texts can be connected to the identities and strategies of groups of social agents trying to develop a new ‘fix’. The successful selection of a new dominant discourse necessarily results in the selecting or filtering out of other, competing discourses that are less able to ‘capture and encapsulate the experiences of social agents’ (Fairclough, 2005, p. 933). An important role may be played in this process by the articulation of themes or genres that may become encapsulated in policy formulations (Fairclough et al., 2002, p. 7). Strategies thus have a partly discursive character that can construct narratives or ‘imaginaries’ and that can be ‘recontextualized’ across organizations and other structures. ‘External’ discourses are ‘internalized’ within these structures, colonized but also actively appropriated by them. Finally, a hegemonic discourse must be operationalized or enacted. To have transformative effects on structures (and cease being ‘merely imaginaries’) the implementation of strategy must become embedded in new practices, inculcating new identities, styles and systems. Relevant social groups and organizations are then likely to recruit and retain social agents who talk and act in their own image. Fairclough (2005, p. 934) gives the example of the hegemonic discourse of ‘new public management’ that produces changes in ‘the identities of public service managers and workers, including changes in their communicative styles’. It is unfair to use illustrative examples – however (un)illuminating – from the article as a test of the framework. For this task we turn to Jessop’s (2004) account of the emergence of the knowledge-based economy as a ‘master narrative’.
Jessop and the knowledge-based economy
A key concern for Jessop (2004, p. 160) is to ‘redirect’ the cultural turn, with its emphasis on discourse, towards CPE. The KBE is then used as an extended case of ‘a distinctive semiotic order that (re)articulates various genres, discourses and styles around a novel economic strategy, state project, and hegemonic vision that affects diverse institutional orders and the lifeworld’. Jessop broadly follows the framework of examining discourses in terms of emergence, selection, hegemony, re-contexualization and operationalization/enactment, though the language differs a little and there is no coherent developmental account provided. We can, however, reconstruct one from the discussion. Jessop’s background in neo-Marxian regulation theory appears to be reflected in how the discussion of the KBE as the ‘hegemonic economic imaginary’ is framed. KBE is initially presented as a response to the crisis of Atlantic Fordism and its competitive challenges from East Asia and Latin America. That economic imaginary must be able to invoke substantive and wide-ranging institutional innovation that reorganizes an ‘entire social formation’, including technologies, labour processes, enterprise forms and forms of competition. From Jessop’s perspective, the KBE is the semiotic glue which binds together the elements of a new post-Fordist social and economic order.
The paper then sets out a more general argument about the conditions under which master narratives emerge and develop. Crises prompt alternative visions, many of which invoke, repeat or re-articulate established discourses. Others insert themselves in a new discursive space opened up for more radical visions. In the struggle for hegemony, the plausibility of narratives and their associated strategies and projects depends on their resonance (including a capacity to reinterpret and mobilize) with the personal or shared narratives of significant classes and groups. There is a useful reference to the requirement of a narrative being ‘capable of translation into a specific set of material, social, and spatio-temporal fixes that jointly underpin a relative structured coherence to support capital accumulation’ (Jessop, 2004, p. 167).
Jessop certainly puts more emphasis than Fairclough on the extra-discursive conditions of narrative appeal. This permits a return to the argument of KBE as ‘an effective solution to the search for a meaningful post-Fordist macro-economic order in an increasingly globalised market’ (2004, p. 168). It was able to meet the preconditions for an economic imaginary that can inform and shape economic and political strategies across a variety of territorial and functional ‘scales’. In fact, the more scales it can address, ‘the more resonant and influential’ a new economic imaginary will be (2004, p. 168). The success of the KBE can be explained by the fact that it has demonstrated high translation capacity. In other words, It has been translated into many visions and strategies, including smart machines and expert systems, creative industries, the increasing centrality of IP, lifelong learning, the information society, and the rise of cybercommunities, and can be inflected in neo-liberal, neo-corporatist, neo-statist and neo-communitarian ways. (Jessop, 2004, p. 168)
As a result, the KBE has played a key role in consolidating a relatively stable post-Fordist accumulation regime and corresponding mode of regulation. Not only was the KBE selected from among many competing post-Fordist discourses, it passes the sustainability test because it has become embedded in a ‘complex and heterogeneous network of practices across diverse systems and scales of action’ (2004, p. 169). Jessop is careful to put a little distance from the actual descriptive claims of the narrative, but it nevertheless ‘does correspond in significant ways to changes in core technologies, labour processes, enterprise forms, modes of regulation and economic identity politics’ (2004, p. 169).
Oddly enough, only towards the end of the paper do we get any argument about the discursive origins of the master narrative – that its material and ideological roots lie in 1960s post-industrialism. We then leap to the claim that the KBE gained momentum as American capitalists and state managers sought an effective reply to competitive threats. It was then translated into successful legal and social campaigns and norms. From there it proceeds up various scales and is warmly embraced as a master narrative by virtually any international body of note – the OECD, WTO, IMF, World Bank, UNCTAD, EU, APEC, ASEAN and NAFTA. This kind of ‘institutional materiality’ means a potentially rapid move from discourse to practice, with Jessop claiming that ‘It can also be used to guide economic and political strategies at all levels from the labour process through the accumulation regime and its mode of regulation, to an all-embracing mode of societalization’ (Jessop, 2004, p. 170).
In hindsight, we would argue that this reading appears to be highly problematic. We are in the middle of a systemic crisis of financialized capitalism and the term ‘knowledge economy’ appears to have dropped off the radar almost completely. Of course, hindsight is, as they say, a wonderful thing. We have, however, previously questioned the accuracy of knowledge economy claims (Fleming, Harley, & Sewell, 2004; Thompson, Warhurst, & Callaghan, 2001). What accounts for the flaws? Jessop’s framework for understanding the emergence and development of discourse is basically sound and formally takes extra-discursive influences seriously. On the surface it appears to be a misreading of the evidence and, in CR terms, a failure to consider rival structures and mechanisms that could exert stronger causal powers. As we saw earlier, he argues that the KBE emerged as a response to the crisis of Atlantic Fordism, dominating the available ‘discursive space’, before diffusing across a variety of institutional and geographic territories due to its high translation capacity, eventually becoming the solution for a stable, post-Fordist macro-economic order.
Jessop ends on a partial disclaimer – there is the possibility for ‘counter-hegemonic versions of the KBE’ (Jessop, 2004, p. 170). This is, however, a highly circumscribed consideration of alternatives. If we return to the emergence of the KBE as an influential discourse, the analysis is strangely silent on any other response, for example, that of lean production, or even the versions of post-Fordism prior to the KBE that Jessop and other regulation school members once discussed (Jessop, 1992).The KBE was indeed a, arguably the, master economic narrative in public policy from the mid-1990s, but Jessop and others have vastly overstated its scope, coherence and effectiveness. In our view, these mistakes reflect theoretical influences as well as empirical judgements. The tendency to view discourses and practices as constituting an entire social formation is typical of regulationist perspectives that have a tendency to shoehorn complex and varied levels of social structure into single, overarching explanations. The tendency to overestimate the reach of master narratives is part of the conceptual baggage of CPE, which appears to be predisposed to privilege DA over analysis of discourse.
The next section gives an alternative reading of the development and sustainability of the KBE discourse. We go on to argue that a rival SV discourse, though subordinate as a narrative, ultimately exerted more influence in key areas of practice.
The Knowledge-Based Economy: The Real(ist) Story
As previously indicated, we have chosen to work within the ‘methodology’ set out by Fairclough and colleagues and utilized by Jessop in the above analysis. This focuses on identifying the interplay between discursive and extra-discursive elements of social change within a framework of emergence, selection/hegemony, recontextualization and enactment. It is worth noting that Jessop does not engage in DA – there is no close attention to texts or the internal dynamics of discourse. Given our focus on analysis of discourse in context, we have no complaint. Our concern is with the negative effects of CPE, with its tendency to distort the prominence and performativity assigned to hegemonic discourses. To avoid such problems we put additional focus on the mobilizing capacity of discourse with respect to agential interests and identities, as well as to the existence of contrasting themes (or ‘genres’) that have differential capacity to influence key areas of practice.
The KBE discourse does indeed have ‘its roots in 1960s post-industrialism’. As observers have noted (Warhurst & Thompson, 2006; Webster, 2002), the centrality of knowledge to claims about new, postindustrial or postcapitalist economies had been around for some time, notably in the writings of Drucker (1969) and Bell (1973). Each of these writers made contingent arguments that linked this centrality to changed socioeconomic conditions. If the KBE becomes hegemonic because there is some real correspondence between the discourse and the conditions, then these earlier claims must be either wrong or, at best, based on premature reading of trends.
By the 1980s, the postindustrial franchise had been passed to theories of flexible specialization and post-Fordism. Given that Jessop was an exponent of post-Fordism, we are never given an explanation of how and why it was ‘selected out’ of competing discourses. In fact, post-Fordism did become the master, big-picture social science narrative, but it largely failed to escape its academic confines.
The KBE re-emerged strongly in the early 1990s. Among the foundational texts, McKeen and Staples (2003, p. 21) refer to the ‘instant legitimation’ received from Drucker’s (1993) book Post-Capitalist Society. Instant or not, the book is notable for re-articulating the central claim of KBE discourse that knowledge rather than capital, natural resources or labour is the central economic resource. This claim is at the core of the first and general theme of the discourse – one that asserts a macro-economic logic. However, as we noted above, this claim has been made a number of times before. We can only understand what made it more persuasive in that period by identifying the other major themes.
In our view, the key text was Robert Reich’s (1991) The Work of Nations. His arguments about the competitive advantage of advanced economies resting on the ‘skills and insights’ of its growing number of knowledge workers was pitched as a policy response to globalization and a shift to a service economy (Blackler, 1995, p. 1027). Success would depend, in future, on high value-added activities, specialized knowledge and services. The fact that Reich was part of President Clinton’s economic transition team, before becoming Labor Secretary from 1993 to 1997, added considerable weight to these views. The knowledge work(er) theme of the discourse is associated primarily with supply-side skill formation and training policies underpinned by uncritical adoption of human capital concepts (Keep & Mayhew, 2010).
This became a central policy theme of the New Democrats and then New Labour. The dual knowledge economy/work themes became linked to a wider hyping of the ‘new economy’, with a significant role played by policy entrepreneurs, ‘gonzo journalists’, ‘futurists’ and popular business writers (e.g. Kelly, 1999; Leadbetter, 1999). Academically, the major influence legitimizing and extending KBE discourse was Castells’ The Rise of the Network Society (1996), in which an informational mode of development directed towards accumulation of knowledge and intellectual property, combined with self-programmable labour in flat, networked organizations, becomes the fundamental source of productivity and power. The ‘cult of Castells’ (Crabtree, 2002) had a significant influence on intellectual and policy debate, facilitating translation from theory to policy.
Other themes of the discourse emerged in the same period. One focused on knowledge-creating companies (Nonaka, 1991) and fed into existing policy debates on innovation and high-tech clusters. This was particularly influential with respect to high-profile areas such as Silicon Valley (Finegold, 1999), but influenced regional and economic development policy in a variety of other locales (e.g. Scottish Enterprise Network, 1999). A second theme focuses on knowledge management. Again, initially this was largely putting a gloss on existing discourses of organizational learning situated within human resource development/soft HRM domains (see Neef, 1999). The knowledge management angle was strengthened by ideas and techniques geared towards leveraging expertise from the ‘new knowledge worker’ (Ruggles, 1998, p. 81). This theme has tended to explicitly disavow the significance of technological tools as means of capturing knowledge. But as Scarbrough and Swan (2001) show, IT specialists have appropriated aspects of the discourse to develop such tools, thus feeding into more general trends in the growth of ICT-enabled surveillance applications within and across organizations.
This discussion shows that the discourse emerged as a serious force and gained intellectual and practical momentum, by combining aspects of the various themes described above in the 1990s. Dissemination and policy translation was extremely rapid. In the UK, the New Labour administration was a particularly quick and enthusiastic adopter and this could be seen in early policy statements on skills, training and industrial clusters (Department of Trade and Industry, 1998). Despite limited evidence of the effectiveness of supply-side policies (Keep, Mayhew, & Payne, 2006), such policy frameworks were continued by the Treasury (HM Treasury, 2002, 2004); the Department of Education and Skills (DfES, 2005) and the Leitch Review of Skills (2005). The discursive resources were initially largely provided by policy advisers from think-tanks such as Demos and (post)Marxist writers (and promoters of post-Fordism) turned policy entrepreneurs such as Charles Leadbetter (1999). They were later reinforced through bodies such as the Work Foundation that linked private and public organizations.
Internationally, the OECD (1996) quickly developed an index of innovation in knowledge and the Lisbon Accord signed by the EU Council of Ministers set targets for research and development, innovation and a learning economy. Together with the World Bank (2002), these international bodies adopted the KBE as a contemporary discursive resource for framing policy packages on competitiveness (Warhurst & Thompson, 2006). With respect to the EC, this was influenced by Castells (1996), who had promoted an active role for the state in creating conditions for competitiveness in the new global economy. What is also interesting about this period is the way in which KBE discourses could be appropriated and modified in the public policy of right-of-centre governments such as the National Party in New Zealand towards the end of its third term of office (Roberts, 2004; and see Peters, 2001). Despite this relatively rapid policy diffusion, the economic arguments of the discourse suffered to some extent through its association with the dotcom bubble (Frank, 2000; Henwood, 2003). Hyping the new economy significantly depended on dubious assertions about knowledge flows and falling costs of information that would facilitate an escape from the business cycle and the economics of scarcity; providing, as one advocate typically put it, ‘ever increasing returns’, ‘without cost-cutting’ (Neef, 1999, pp. 5, 8). The appearance of a viable new economy business model without a cost recovery requirement was always dubious.
The crash was inevitable because the hope that new companies could turn untried and rapidly changing digital technologies into profitable mass-market products represented not a judgement about cost recovery, but a suspension of disbelief in a gold rush which involved sinking mines without a geological survey (Feng, Froud, Johal, Haslam, & Williams, 2001, p. 48). The elements of the discourse that survived and continued to flourish did so in somewhat narrower and more specialized domains. On skills, a policy nexus focused on politicians, state managers, think-tanks and educational institutions had been consolidated. For the former, the KBE provided a rationale for not enacting demand-side interventions focused on macro- economic conditions and industrial strategy. As Keep and Mayhew (2010, p. 570) note, this constituted ‘a powerful form of displacement activity, whereby the structurally-rooted causes of economic and social policy problems are set to one side, and the “solution” reduced to a government-funded skills initiative’. Universities too, found KBE rhetoric advantageous in their own competitive struggle for scarce resources, given the emphasis on expanding the supply of graduates despite evidence that the number of graduate jobs was diminishing (Green & Zhou, 2008). In another strand, knowledge management practices became more widespread (though not as pervasive as the literature might indicate), as key agents – informational systems specialists and large consultancies such as Ericsson and Anderson – developed tools such as intranets, groupware and software toolkits that could persuade high-tech companies that there were practical means of leveraging the tacit knowledge of employees (McKinlay, 2005; Scarbrough & Swan, 2001).
If we examine the process of diffusion and embedding, there is little or no evidence that the discourse articulated the interests or coordinated the practices of most private capital. The dominant supply-side KBE discourse is marginal to the key concerns of large transnational firms and particularly to the financial sector, who were much more interested in free trade and deregulation. Indeed, the Clinton and Blair administrations, for all their KBE rhetoric, were prime actors in pursuing these policies, as we shall see in the next section. The best that could be said is that the innovation and knowledge management strands coincide with the interests of a fraction of capital in the knowledge-intensive industries, who can benefit from extra incentives for investment and related policies, as well as the aforesaid knowledge management practices. To explain what was actually happening in the macro economic domain, we have to shift emphasis and look to a much less high-profile discourse, that of shareholder value.
Under the Radar: Financialized Capitalism and the Discourse of Shareholder Value
The KBE discourse, in a variety of guises, was highly visible in academic work during the 1990s. In contrast, the discourse of shareholder value (SV), while appearing regularly in the financial pages of the business press and to a lesser extent specialist business literatures (e.g. Doyle, 2000), was largely ignored by social scientists (Williams, 2000, p. 1). Even in comparative political economy literature, ‘financial agents, financial markets and financial products have only a nebulous existence in the representations of contemporary capitalism’ (Engelen & Konings, 2010, p. 604). Though subordinate to the KBE discourse, at least in academic circles, it has had a profound impact on the shape of contemporary capitalism and on managerial practice within contemporary organizations. In this part of the paper, we provide an account of the emergence, spread and influence of SV discourse.
SV discourse can be traced in part to broader notions of shareholder sovereignty that emerged in the United States in the 1970s, as part of a struggle between managers and investors for control of companies (Jensen & Meckling, 1976). According to Green et al.: The industrial decline of the 1970s provided shareholders with a unique opportunity to rhetorically attack management control of the firm…Simply put, if managers were the rightful stewards of the American firm, why was the economy falling apart? Allies and proponents of the [investor control] logic answered these questions by suggesting that the rightful stewards were the shareholders. (Green et al., 2008, p. 57)
In the terms of the logic of SV discourse, firms should be judged on the basis of value delivered to shareholders. Similarly, managers should be judged on whether they undertake actions that deliver such value. Two key themes can be identified. First, texts directed at promoting a market for corporate control. The initial driver was agency theory directed towards weakening the existing forms of managerial capitalism and promoted an investor-led version. Studies by a group of financial economists centred on the Chicago School of Law and Economics (Jensen, 1986) provided credibility via apparent quantitative rigour to new institutional actors such as investment, equity and sovereign wealth funds. Initially, those ideas helped to legitimize a wave of hostile takeovers (Icahn, 1988). Williams (2000) also reminds us of the role played by another institutional actor, arguing that the term ‘shareholder value’ was introduced in the US in the 1980s by consultants selling ‘value-based management’ to companies that were under pressure to increase their returns.
A second theme connected the discourse of SV with the emergence of what would later become known as neoliberal political projects, at that time focused on the Reagan presidency and a new generation of policy-makers and regulators involved in the Council of Economic Advisors and Securities and Exchange Commission (SEC). SV discourse played an important role in strengthening broader shifts towards marketization in the economic and political spheres. There is evidence of changing terms of debate in business organizations (and business schools) away from stakeholder models (Khurana, 2007). Most crucially, as Green et al. (2008) and Tett (2009) detail, SV discourse underpinned key legal and (de)regulatory changes including case law, SEC decisions and, later, the blockage in Congress of bills designed to limit speculative activities by the finance sector.
In sum, the combined effect of these two themes of SV discourse was to legitimize increased shareholder activism and deregulation. Importantly, it seems clear that SV discourse did not cause shareholder activism, but was used to legitimize and buttress it. There was, therefore, a causal process which involved the interplay of discursive and extra-discursive phenomena, rather than a simply one-way process. By the mid-1990s, what had occasionally been referred to as financial engineering had become the emergent framework for a financialized economy. As one of the few research groups with a financialization perspective in this period argued (Froud, Haslam, Johal, & Williams, 2000; Williams, 2000), in such economies the driver of capital accumulation shifts from competition in product markets to the requirements of capital markets. The latter are no longer merely intermediaries for firms seeking sources of capital, but a regulator of firm and household behaviour. In the language of labour process theory, this represents changes in the circuits of capital that rework the logic of accumulation (Thompson & Vincent, 2010).
Growth strategies for firms are directed to a simultaneous squeezing of labour and more active management of corporate assets, manifested in delayering, disaggregation, downsizing and divestment. This fuelled the wave of corporate restructuring that became prominent in the 1990s. In the context of the long bull market, 80 per cent of shareholder returns in the 1990s came through the rise in share prices that allowed companies to escape the limits of returns on earnings in product markets (Erturk, Froud, Johal, & Williams, 2004, p. 689). As Blackburn (2006, p. 43) notes, in such an economic context, the corporation becomes a ‘plaything’ of capital markets, ‘an accidental bundle of liabilities and assets that is there to be re-arranged to maximise shareholder value… the corporation and the workforce are, in principle, disposable’.
On the face of it, such disposability might seem to run counter to the interests of senior managers. That, however, ignores the now widely observed rise of stock options as the main source of executive reward. Spurred on by a ‘chorus’ of finance professors (Erturk et al., 2004, p. 690), such policies succeeded in tying the interests of the corporate elite to permanent restructuring and often short-term and speculative behaviours (Henwood, 2003; Lazonick, 2009). The growing political power of this new alliance of financial and corporate interests was manifested in further waves of deregulation, the most prominent being the repeal of the US Glass-Steagal Act in 1999 that had previously kept the commercial and investment activities of banks separate. Nor was such influence confined to that country. The general approach to SV was endorsed in a 1999 OECD publication, The OECD Principles of Corporate Governance (Lazonick & O’Sullivan, 2000, pp. 13–14). In the UK, at the same time as promoting the virtues of the knowledge economy, the New Labour government was lessening regulatory requirements on the City, which they now, belatedly, admit was a mistake.
It is worth emphasizing at this point that the trends mapped out above are incompatible with claims concerning the hegemony of the KBE. The KBE discourse was very much consonant with that of ‘soft’ HRM and ‘high performance work systems’, in which organizations could compete by nurturing talent and harnessing it through developmental and humanistic HRM practices. The negative consequences for the sustainability of ‘high road’, mutual gains work practices and stakeholder-oriented corporate governance was set out systematically by Thompson (2003) in his analysis of ‘disconnected capitalism’. The approach to corporate management which has been justified by SV discourse undermines the required degree of stability and capacity for investment in firm-specific asset. In an environment where cost-cutting is imperative, labour is usually ‘the first casualty’ of corporate restructuring (Froud et al., 2000, p. 771). As Clark (2009) demonstrates in his research on PLC firms governed by a private equity business model, financialization makes it more difficult for firms to sustain stakeholder governance structures that are employee-friendly, given a shorter-term focus and requirement to service debt.
Of course, these tendencies have been manifested in different ways in different national contexts, reflecting the prominence and power of economic and social institutions that could ‘translate’ the discourse and accelerate or constrain the spread of financialized practices. Business journalists such as Will Hutton (1995) had run a long campaign to contrast the virtues of stakeholder ‘Rhenish’ capitalism to its Anglo-American ‘shareholder’ variant. In the Special Issue of Economy and Society, where the first serious social science discussion of financialization took place, contributors took different views of the extent to which this phenomenon had penetrated the walls of the German and French business models. Attention was focused on factors that affect differential exposure to capital market pressures such as: the size and character of stock markets; the relative openness of the market for corporate control; the stabilizing role played by banking institutions; and the presence of new entrepreneurial actors such as global consultancy firms and equity funds. The dominant view was that long-term trends were working against traditional stakeholder corporate governance models (see also Faust, 2009; O’Sullivan, 2000). Those long-term trends promoting the interconnectedness of the global economy, in part promoted by financialization, were soon to be become glaringly obvious.
The credit crunch: financialized capitalism emerges from the shadows
This is not the place for a blow-by-blow account of the global financial crisis for which the general condition and consequences are well known. We want to emphasize a few key points that are pertinent to our general theme of financialization as something that emerged largely under the discursive radar. The crisis may have emerged from the interaction between the financial sector and the housing market, but its roots are wider and deeper. By the early 1990s banks, companies, pension funds and other asset managers had begun to turn to corporate loans and bonds, notably derivatives, to make big gambles and realize large returns. Such moves were facilitated by new instruments (e.g. value at risk measurements) and actors (e.g. high-risk hedge funds, the International Swaps and Derivatives Association). Global consultancies such as McKinsey continued to play a central role in hyping the performance of shareholder-oriented economies and their own calculative techniques (McSweeney, 2008).
Banks needed a means to disperse and diversify credit and shift risk off their books and this was provided by collateralized debt obligations (CDOs) that bundled lots of deals together and thus appeared to pool risk (securitization). In the wake of the collapse of the dotcom bubble at the end of the decade, the focus of attention in the world of cyberfinance shifted to real estate and mortgages. We know the story – banks and other lenders issued more and more risky mortgages, so that those loans could be repackaged into more and more CDOs in order to make up for the declining profit margins. The key development was that banks turned toward households and individuals as sources of profit, a dangerous manoeuvre in a period of falling or static real wages. In essence this was a crisis waiting to happen. With regulation dismantled and risk dispersion spread throughout the financial system, this new version of financial engineering was taken to a level of complexity which was unsustainable.
Marazzi (2008) argues that to explain the workings of financial markets in the era of post-Fordism we need a linguistic theory of their operations. Yet it is not clear what such analysis could reveal, as this was a crisis marked by ‘the failure of the private and corporate actors to understand what they were doing… the financial system created a fog so thick that even its captains could not navigate it’ (Muller, 2009, p. 2). As the detective work of financial journalists such as Gillian Tett (2009) shows, most non-bankers had no idea at all that these new and toxic credit investment products existed and most people inside the banks did not know how they worked: ‘Nor were SIVs (structured investment vehicles) mentioned in the mainstream British press. The financiers had created a vast shadow banking system that was out of the sight of almost everybody outside the specialist credit world’ (2009, pp. 115–16).
That financial discourses were largely submerged or opaque does not mean they were insignificant or silent. The second, broader SV theme provided a legitimating narrative that underpinned the marketization of risk associated with securitization, derivatives and the democratization of finance (Montgomerie & Williams, 2009, p. 99). Particular discursive subplots such as the efficient market hypothesis provided rationales for key economic actors, from universities to Wall Street, to shun regulation even when asset bubbles were threatening to burst (Taylor, 2004). As can be demonstrated from MacKenzie’s (2006, 2009, 2010) work on knowledge and language in financial markets, a third theme certainly existed based in the local evaluation cultures developed inside small subgroups. Complex technological texts legitimated products and metrics, but an already ‘impenetrable’ jargon reached the limits of the language when mortgage-backed securities were repackaged into CDOs and other instruments. The effects were devastating, though the discursive resource was only accessible to a tiny number of people.
Despite the trail of post-crash havoc, our capacity to understand events has also been partly obscured by the emergent dominant narrative ‘that unregulated financial markets have become increasingly de-coupled or disembedded from the real economy’ (Montgomerie & Williams, 2009, p. 101). Financialized capitalism was not called into being by the specific form the crisis took. Its immediate forms can distract us from the longer-term ‘systemic transformation of mature capitalist economies’ (Lapavitsas, 2011, p. 611). That transformation, as Lapavitsas (2011), Blackburn (2006) and Lazonick (2009) show, involved large corporations as players as well as victims: whether raising external finance on open markets; financing the wage bill through issuing commercial paper; making substantial portions of their profits from consumer credit and leasing arms; involvement in bond and equity trading as part of the successive waves of takeovers and restructuring; or boosting share price through buybacks of their own stock. Financialized capitalism may have undergone a systemic crisis, but none of the fundamentals have changed. It has been widely observed that despite the challenges to varieties of SV and neoliberal discourse and the public recantations by some of its former advocates, the power of financial elites and financial circuits of capital remain largely intact. This raises important issues that we discuss in the next section.
Discussion
How can we explain the fact that the SV discourse, in spite of its apparent role in contributing to change in the extra-discursive realm, flew under the radar while the KBE gained such prominence? It is important, as we indicated in an earlier section, not to underestimate the attractiveness of the KBE story to academics themselves. New economy narratives have been the dominant means of accumulating reputational capital since the earlier versions of post-industrial society. Despite a paucity of evidence, many academics have been content to repeat the rhetorical dispositions of KBE discourse with minimal critical interrogation. This is hardly surprising when we consider the way in which that discourse permeated the policy and research programmes of key national and international agencies. Academics, after all, are part of the supply-side networks that institutionally benefit from investment in human capital.
In contrast, the SV discourse and a concern with financialization was partially submerged in concerns with the speculative excesses and long bull market of what appeared to be a new business model that peaked and crashed with the Enron scandal in 2001 (Henwood, 2003). There was also a tendency for the story to be buried within a broader discourse on neoliberalism. As Montgomerie and Williams (2009, p. 100) note: Neo-liberalism demarcates the period since 1970 as one of privatisation, deregulation, Washington Consensus and all the rest, which essentially takes ideological propositions at their own word and, consequently, fails to distinguish between rhetoric and practice.
SV and financialization cannot simply be bundled up with a variety of other discourses and practices as if they constitute a single project with specific purposes and outcomes.
Explanation of why one discourse is more or less prominent among academics is not the same thing as the relative contribution of competing discourses to change within political economies. In terms of the way we set out some of the criteria in friendly dialogue with Fairclough and Jessop, a number of observations can be made. The KBE appeared to meet some of the developmental conditions of emergence, hegemony, recontexualization and operationalization, For example, it did have high translation capacity and did articulate some interests and identities (a facet of recontextualization), but only to particular territories and via particular actors. Narrative hegemony can exist without dominance of key socioeconomic practices. KBE discourses had their greatest impact on supply-side policies, creating some new practices (the operationalization measure favoured by Jessop and Fairclough) and, even then, predominantly in terms of public rather than corporate policy.
KBE discourse may also be consistent with the experience of knowledge-intensive sectors such as biotechnology and pharmaceuticals, though the extent to which the discourse significantly influences practices is open to debate. More importantly, many of the major economic and corporate trends of the last two decades – perpetual restructuring, low cost competition, low investment in R&D, continuation of lean practices and extension of those into services and supply chains, and low-level service job growth – show the weakness of correspondence claims between KBE discourses and contemporary capitalist political economy
For critical realists, a dominant discourse can be blocked and countermanded by a more powerful extra-discursive force. We would argue that this has been the case in key domains, notably investment in long-term firm-specific assets. As Lazonick (2007) illustrates, SV discourse and practice is a major constraint on innovative enterprise, draining companies of the necessary financial and human resources. While KBE discourse might appeal to ‘plant level’ managers, under conditions of financialized capitalism they are far less autonomous and much more subject to ‘the close surveillance of a board of directors who represent exclusively the interests of shareholders’ (Dore, 2008, p. 1103). Within large corporations there is evidence of a decline in culture-led, soft HRM rhetoric and a corresponding ‘surge’ in the rhetoric of market rationalism (Kunda & Ailon-Souday, 2005).
In contrast to KBE, SV discourse did not appeal to ‘diverse institutional orders’. Its propagators (e.g. finance professors) and institutional agents (e.g. large consultancies, equity fund managers) were even more ‘localized’ than their KBE equivalents, but had greater performativity as that location was at the heart of capitalist political economy. Its ‘fix’ had more institutional materiality. All contexts in capitalism are not equal. Supply-side measures, in particular, have less power to generate effects than demand-side factors susceptible to influence by corporate actors. While less visible and more opaque to ‘outsiders’, SV discourse spoke to and for a project to transform the macro-economic order or, in the language of regulation theory, the regime of accumulation.
In such circumstances it is perfectly possible for an ‘economic imaginary’ to fill a discursive space, but largely with textual noise that is marginal to core economic practices. It is also possible that a plurality of discourses can share a space and perform different functions in overlapping but largely non-competing domains. This was not like a boxing match in which the two sets of discourses and practices were always in direct competition and could deliver a knock-out blow. As the work of Green et al. (2008) indicates, the competitor to SV rhetoric was managerial or stakeholder capitalism, not the KBE. As SV discourse was articulating the interests and legitimating the practices of corporate actors, the KBE equivalent was primarily a resource for governance projects at national and supranational levels. The ‘endless inflated claims’ concerning the prospect for the ‘new’ knowledge economy (Peters, 2001, p. 12) constituted an optimistic story with broader cross-party and public appeal that asserted the capacity for limited state action but masked the actual adaptation to financial markets, as evidenced by weakened regulation and economic intervention.
Finally, it is worth emphasizing that under financialized capitalism there are complex interactions between the discursive and non-discursive elements. This is not a simple correspondence or mirror model. The discourse can be understood in terms of generative mechanisms with performative potentials, but that is not to say that FC is a single, successful, stable regime of accumulation. There are a variety of contingent and contested trajectories of financialization (Engelen & Konings, 2010, p. 617), in part as Froud, Leaver, Johal, & Williams (2006, p. 7) note, because ‘shareholder value as social rhetoric can be appropriated and inflected by various social actors, financialization is not associated with one invariant set of consequences in terms of firm performance or management behaviour’.
Conclusions
The primary objective of this paper was to develop new ways of thinking about the interplay between the discursive and the extra-discursive in exploring processes of social and economic change. Our secondary objective was to add to existing attempts to explain dynamics and shifts in capitalist political economy and to add to empirical studies of the nature of financialized capitalism.
With respect to the latter, by providing an account of two competing discourses, we hope that we have added to existing understandings of the emergence and growth of financialized capitalism and the role of the discourse of SV in driving key changes. We have sought to show how the SV discourse ‘flew beneath the radar’ of most observers, who focused instead on the rather less significant discourse of the KBE. While we argue that the SV discourse has been more important than KBE, this is not to imply that its triumph reflects superiority of argument or outcome. As McSweeney (2008, p. 52) argues, ‘instead of solid empirical support the rhetoric of maximising shareholder value is dominated by anecdotes; crude notions of causality; exaggerated predictive power; ridiculously simplistic views on the internal workings of companies, and an unreal and utopian/dystopian notion of markets’. In terms of developing an understanding of the likely future trajectory of developments in the capitalist political economy, we would argue that, even if the SV discourse plays a role in a new ‘fix’, it is unlikely to be a stable or permanent one.
In terms of our primary aim, by evaluating Fairclough and Jessop’s CDA approach and its application to the KBE discourse, then providing both an alternative reading of KBE and an argument about the significance of the SV discourse, we are able to draw some general conclusions about the application of CR to discourse and to some extent to get beyond the apparent logjam in the positions discussed in the opening section. By taking discourse seriously, we have been able to demonstrate its importance as a potential causal mechanism without a priori privileging discourse as the dominant phenomenon or explanation.
More specifically, while the framework of focusing on the conditions of emergence, hegemony, recontextualization and operationalization set out by Jessop and Fairclough offers a significant step towards a framework which integrates the discursive and extra-discursive in explaining social and economic change, we have sought to demonstrate that it did not offer a wholly credible account of the KBE. That is not to say that the cultural dimensions of political economy analysis are insignificant. We have no problem agreeing with Willmott (2010) that cultural discourses on SV can shape as well as describe financialization. However, research on the potential causal powers of discourse needs to pay less attention to intertextuality and more to actors and institutions, and to extra-discursive conditions and constraints. With respect to the former, recent emphasis on new financial elites (Hall, 2009) and corporate intermediaries such as international legal, accounting and consultancy firms (McSweeney, 2008) offers promising directions. It might also be wiser to consider extra-discursive factors in a more bounded way. The idea of hegemony can be misleading if it implies that discursive dominance is necessarily system-wide. Utilizing the kind of analysis offered by Green et al. (2008), more use could be made of the idea of competing logics within or across distinct institutional fields whose characteristics may limit translation or enactment capacity. However we seek to make sense of the specific roles of actors and institutions or to employ the concept of competing logics, our overarching conclusion is that discourse is too important to leave out of the equation, but not so important that it can bear the burden of explanation on its own.
Footnotes
Acknowledgements
The authors would like to thank the three anonymous reviewers for their comments on earlier versions of this paper.
Funding
This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.
Author biographies
Paul Thompson is Professor of Organisational Analysis in the Business School at the University of Strathclyde. His research interests focus on skill and work organization, control and resistance, organizational restructuring and the changing political economy of contemporary capitalism. He is the Convenor of the Steering Group of the International Labour Process Conference and co-editor of the Palgrave series Management, Work and Organization.
Bill Harley is Professor of Management, Department of Management and Marketing at the University of Melbourne. He is General Editor of the Journal of Management Studies. He has published extensively on issues of HRM, work organization, high performance work systems, employee involvement and labour process theory.