“There is one and only one social responsibility of business–to use its resources and engage in activities designed to increase its profits so long as it stays within the rules of the game, which is to say, engages in open and free competition without deception or fraud.”
1Can we take Milton Friedman seriously? This famous, pithy and provocative phrase, according to which a company’s only social responsibility is to increase profits and the remuneration of those who own it, is a priori refuted by the whole sociological corpus. Indeed, companies do many things other than increase profits: within them work is organized and regulated (Reynaud 1988; Lallement 2007), identities and affiliations are constructed (Sainsaulieu 1977, 1992), and they can also be places for innovation (Le Masson et al. 2006) and the accumulation of knowledge (Dosi et al. 1990; Segrestin 2004). Undoubtedly, Friedman would respond, but these are only effects induced by their true function: it is in order to increase their owners’ profits that companies bring together employees and coordinate and regulate their work, it is because employees interact within them over the long term to generate value that firms become one of the foundations of their identity; it is in order to endlessly produce increasing wealth that there is innovation and knowledge accumulation within firms. The objective of companies is profit and shareholder remuneration. Can we go beyond this classic opposition between sociologists and economists, that extends a dividing line between the two disciplines over the empirical subject that is “big business”?
2Without departing from the fundamental propositions advanced by sociologists who have studied companies, we have chosen to take Friedman’s proposition seriously in this presentation of studies on big business in recent decades —or, more precisely, to make an enigma of it. Neo-institutionalist sociology has indeed shown that capitalist firms can pursue alternative objectives (Fligstein 1990 and later Davis et al. 1994 on changes over time in American industry; Dobbin 1994, on differences between France, Britain and the United States in the nineteenth century). By challenging the functional evidence for the quest for profit and shareholder remuneration, these studies highlight the importance of normative considerations for the definition of companies’ objectives. This is precisely the furrow we will plough here: since the existence of large capitalist companies, both actors and researchers—and not just Friedman—have often highlighted the objective of profit and shareholder remuneration. While not forgetting that alternatives have existed, it is for us to account for whether this objective was a priority or not depending on the period.
3To better understand whether capitalist big business is by definition linked to the objective of profit and shareholder remuneration, it is helpful to examine its origins, in other words the nineteenth century and even earlier periods when the legal framework that supports it was developed—which we shall do in the first part of this article. First we will focus on the legal framework which brought about capitalist investment which was far from straightforward, then on the conflicts caused by the desire to control and to capture the consubstantial profits of this investment. In particular, we will address the conflict between shareholders, who provide the capital, and managers who have to make it grow—as described by Berle and Means in 1932: the manager does not always obediently follow the injunctions of the owner, and may also have his own objectives. This first part of our literature review draws on works that are as much historical as sociological, or from other disciplines. They all discuss a very functionalist model, that of the business historian Alfred Chandler (1962, 1977, 1990), firmly rooted in twentieth-century US history. From this range of works, which stem from different backgrounds and do not constitute a coherent corpus, we attempt to extract an alternative narrative to that of Chandler.
4Relations between shareholders and managers and their connections to the process of financing companies are at the heart of the second part of the article, centred on the last decades of the twentieth century and the beginning of the twenty-first. This section presents a literature that is essentially in English, that is cumulative and for the most part presents itself as such, and which largely comes under the banner of neo-institutionalist sociology—while suggesting it be supplemented with a diversification of empirical cases beyond the United States. In these two sections, we cannot claim to touch on work on all regions of the world. This restriction is not due solely to our limits: research on big business, and in particular on the questions we raise here, is polarized by the studies carried out on the American case. This is not, however, a representative case, nor a model towards which other national trajectories are compelled to converge: developing work on other countries is thus one of the possible ways to provide a less linear narrative of the relationships between companies and their shareholders.
The birth of big business: From one (functionalist) history to another
Chronological revision and causal divergence
5For a long time, historians were not particularly interested in capitalist big business—with the obvious exception of the history of labour and of the capital-labour conflict—and those who studied the economy often limited themselves to the macro scale. The history of the organization of production, and more recently relations between shareholders and managers, and the legal framework of business, essentially emerged from business schools, and adopted a frequently normative posture (when discussing what business should be, and often praising entrepreneurs) and/or a functionalist one (analysing historical developments as logical responses to clearly-identified problems). While business history originated in the inter-war period in the United States (Fridenson 2014), it was in the 1960s–1970s that, based on the literature and new monographs, Alfred Chandler suggested a functionalist account of the birth of the large, integrated firm that historians and sociologists have since sought both to illustrate and revise, particularly in terms of its functionalism. For Chandler (1962, 1977), in the nineteenth century, it was because the size of markets was growing (thanks in particular to new means of transport) that these companies also grew, and that as a result new technical and organizational issues arose. In response, the railways in particular invented modern forms of management and organization, such as internal labour markets based on a bureaucratic system that formalized recruitment and promotion. It was also to finance this growth that ownership had to be spread among a very large number of shareholders. They ended up no longer having any knowledge of what was taking place within the companies they owned a stake in. The figure of the manager thus emerged to manage these immense companies. And if this figure emerged later in Germany or in England than in the United States, it was because the small size of their markets delayed the growth in the size of their companies: as export barriers fell the same dynamic arose, which led to the same distribution of power (Chandler 1990).
6The argument advanced by Chandler was a major revision of the hitherto established chronology. Indeed, the term “industrial revolution,” and also Adam Smith’s and Karl Marx’s interest in the first but still uncommon factories of their time, had long led to a belief that very large companies had dominated in the countries involved in this “revolution” since the end of the eighteenth and the beginning of the nineteenth centuries. Chandler himself sought to date the birth and subsequent spread of the large integrated firm to the late nineteenth century. The literature on the growth in size of firms and on the wage-earner society in the last twenty-five years has overall corroborated the chronology advanced by Chandler. This has been independently confirmed by legal and business history as well as economic and labour history. Whereas nineteenth-century Western economies can be described as capitalist and growing, they were still dominated by commercial, rather than industrial, and small, rather than large firms (Verley 1997; Gervais 2007; see Lemercier 2014 for an introduction). It was really in the 1880s–1900s, in particular, that wage labour was created as a specific compromise (subordination to an employer and a limited share of the value created, but also protection by the employer and/or the state from various risks) partly reducing the conflict between labour and capital, stabilizing employment relations and enabling the birth of internal markets (Castel 1999). It was only at this point that organizational and legal forms were created that enabled the invention of this compromise and supported its diffusion (Perrow 1991). Prior to this there were no employment contracts (Deakin and Wilkinson 2005), no foremen, and more generally no hierarchy between worker and entrepreneur (Lefebvre 2003), nor human resources departments (Jacoby 1985). Even in the large production units, with the exception of those using forced labour or one of its variants (e.g., plantations and English factories using the criminal justice system in their employment relations—see Stanziani 2010), labour relations were based on the market and were decentralised until at least 1870 and still largely so until just after the First World War. The latter played a major role in establishing the wage-labour compromise and in the spread of the scientific organization of production (Moutet 1997), mostly with the support of states.
7These studies, although they establish a similar chronology, are opposed to those of Chandler in the sense that they seek to provide less linear descriptions of developments and to be less functionalist in their imputation of causes. Firstly, the time before 1870 ceases to be read in terms of what it lacked, as a time when firms par excellence, those of the twentieth century, did not yet exist. Since the large integrated firm model has been called into question as the dominant form of contemporary organization of firms, more and more researchers no longer want to see the nineteenth-century merchant economy as an archaic system destined only to be outmoded: instead they analyse its own dynamics and rationalities—ironically this trend started even before Chandler’s first major book was translated into French, in particular in an article by Sabel and Zeitlin (1985). Therefore, since the period preceding the birth of integrated big business has been reconsidered, it is no longer possible to present its birth, and along with it that of wage labour, as completely natural, or simply as a result of increasingly integrated markets, the dating of which continues to be the subject of debate (Daudin 2010). It is not owners and company managers alone who adopted more integrated forms of organization as a logical reaction to the growth in the size of markets. The transformations that affected companies are better analysed as a result of conflicts between different groups: trade unions and business leaders obviously, but also, and especially, social reformers and the state.
8The same revisions apply to two issues on which we shall elaborate a little more: the growing adoption of the publicly listed corporation as a legal form and a means of finance, and the beginnings of conflict between shareholders and managers. These innovations from the end of the nineteenth and beginning of the twentieth centuries did not stem from the obsolescence of previous forms, and the state played a key role in their conflicted origins.
The invention of profit and use of legal frameworks
9Concerning the invention of the publicly listed corporation, the history of law and accounting has long been presented in as linear and functionalist a fashion as Chandler’s; moreover he included its account in his argument. Here again, it was the growth in size and complexity of companies that required them to be more sustainable (thus to go beyond a simple partnership), to attract greater capital (and thus to have recourse to financial markets), to limit personal liability in relation to their own money firstly of those providing the capital and secondly of the company managers themselves (through different forms of limited liability companies, in particular the corporation) and finally to standardize the measurement of profit in order to distribute to shareholders their fair share. Such linear narratives seem to cohere with that of Max Weber’s founding account: according to him, the formal rationalization of law and accounting in particular goes hand-in-hand with the birth of capitalism, or may even have been one of its necessary conditions. Although it is in part under the auspices of a rediscovery of Weber that there has been growing interest in the law in economic history and economic sociology (remaining with the French-speaking world, see Stanziani 2007 and Bessy et al. 2012), studies in recent decades have clearly revised, or point blank contradicted this old and persistent account. The dominant theme of this revision was largely a return of the political, emphasizing the non-economic origins of legal frameworks used by capitalists, and the importance of conflicts and power relations in their adoption.
10Using the French case as a starting-point, the historian and former economist Alessandro Stanziani (2012) thus recently emphasized how much the emergence and expansion of capitalism in the seventeenth and eighteenth centuries unproblematically drew on law that was much older, designed to regulate interpersonal merchant relationships. The same is true for accounting, which also retained methods more focussed on managing credit relations than on calculating profit until late in the nineteenth century. This was shown in particular by Gervais (2012): even the merchants managing the most profitable eighteenth century companies had no means to estimate the profits of each operation; moreover, it was not what they were interested in. Even in the nineteenth century, in the industrial companies that have most interested historians of accountancy, profit calculations remained tentative, with no consensual norms (Labardin and Pezet 2014). There is, however, a significant exception to this general situation. From the eighteenth century, special partnerships could be set up, sometimes in the form of (short-term) joint stock companies, in order to raise more substantial capital—this was the case, for example, for the chartering of transatlantic ships. When it came time to pay each of the shares, profits needed to be calculated. In the last decades of the nineteenth century, it was the fraud that resulted from the emerging clash between shareholders and managers that gave rise to a normalization of the calculation of profits: some companies offered “fictional dividends”, drawn in reality on their capital, to attract new shareholders, while others hid their profits in order to reinvest them rather than distribute them (Lemarchand and Praquin 2005). Ultimately, in France, it was the state, around the First World War, that codified the legal definition of profit, since the first tax on companies was a tax on war profits. Without a shared literature, economic historians and historians of accountancy have thus largely repeated the foreshortened and qualified Weberian account given by the sociologists Carruthers and Espeland’s (1991).
11For profits to be calculated in the sense understood by Friedman, there thus needs to be both shareholders and a state, each seeking to protect their share of this profit and guard against its possible appropriation by company managers, because those managers better know the reality behind these accounts. The association economists make between the profit objective and the remuneration of shareholders is thus empirically confirmed, but also restricted to the period following the emergence of integrated big business. More specifically still, it applies to the period from the middle of the nineteenth century in the United States—and from the latter decades of the century in France—when corporations became very common. It is their multiplication, with the ensuing conflicts, that gave rise to the creation of accountancy norms: the latter are not a condition of the birth of capitalism, and corporations multiplied when it was still not known how their profits should be distributed or even calculated. Retracing the history of the legal frameworks used by these companies enables us to observe that they were not invented to manage larger capitalist investment; instead it was older frameworks that were updated to do this.
12Historians and sociologists have indeed shown that such legal forms were much older, and used in more diverse ways, than Chandler or Friedman describe. In the context of their more general research on Florence as a laboratory of political and economic innovation in the fourteenth and fifteenth centuries, Padgett and McLean (2006) stressed the precocity of legal forms enabling the separation of capital owners from business management and the restriction of owners’ responsibility for the sums involved, all the while highlighting what the adoption of the forms for international commerce owed to the revolt of the Ciompi. Although this Florentine partnership system can be placed within the genealogy of the modern corporation, it does not mean it is the product of a linear evolution that increasingly differentiates roles or increasingly protects the interests of shareholders. It is not, as Chandler would have it, the simple legal dressing of a technical necessity. On the contrary, many authors from very different intellectual perspectives have since the 1990s stressed the importance of the state and of political conflicts in the history of the corporation, as well as the persistent diversity in legal arrangements between shareholders and managers in most developed countries.
13Well before their work was discovered by economic historians, it was American legal historians who, from the 1970s, were the first to revise, and in particular to politicize, the history of the corporation (for an overview, see Novak 2009). They highlighted that up to the beginning of the nineteenth century, the corporation was a legal status applied indiscriminately to some companies and also to municipalities, churches and universities. As such, it was mostly used to confer an often regional monopoly on these companies in exchange for various public service obligations, in particular restrictions on their profits. Most charters founding these corporations did not provide limited liability for their managers. These corporations, which were thus defined as true political entities, were mostly financial or transport companies rather than industrial ones. The economic historian Naomi Lamoreaux (1994) added to this picture the fact that these corporations were hardly anonymous. While Weber saw their development as one of the symptoms of the depersonalization that accompanies the rationalization of economic relations, Lamoreaux shows that shares were generally few in number, thus worth very large sums, were rarely sold, and their ownership was concentrated in very homogenous social and geographic milieus. Jean Rochat (2014) has shown that this was essentially the same in France until the 1860s, even though the corporations were all firms, not non-profit organizations: these were often considered to be quasi public (their status signaled a separate class of companies considered less risky and more legitimate) and they had few shareholders who were drawn from tight circles. Corporations were thus less a tool for the depersonalization of capitalism than an additional legal means supporting family and friendship ties.
14When, from the middle or the end of the nineteenth century, the first major integrated companies adopted this legal status, it was therefore to do something new: to engage a larger number of shareholders, resell their shares more frequently, and to adopt the aim of providing dividends rather than public services. This change was enabled by a change in laws, which now required a simple declaration and no longer the authorization of government or parliament. In the United States this change was gradual, radiating from the pioneer states as the sociologist William Roy (1997) has shown. He highlighted the role of the populist movement, which had long criticized the monopolies and collusion between political and capitalist leaders. This criticism might have led to reinforcing the constraints corporations would have to submit to, to ensure that the services, which were the bases of their privileges, were actually rendered. It was the opposite option (removing both the privileges and constraints) that was chosen, notably because of pressure from banks and jurisdictional competition between states.
15It was thus as a result of political rather than economic developments that, towards 1900, the corporation became the hegemonic legal status for large (and even mid-size) US companies. These developments were, however, specific to the US: elsewhere, in France and Germany in particular, other forms had long competed, and even continue to compete with the corporation, and this is not just an archaic symptom, as neo-institutionalist economic historians have shown (Guinnane et al. 2007, 2008). They highlight that the many conflicts inherent to the existence of several owners cannot be optimally resolved by a single legal status, in particular if it is a rigid one. They point to the widespread adoption of the limited liability company, invented in Europe between 1890 and 1930 (the German GmbH was born in 1892, the British PLLC in 1908 and the French SARL in 1925). This status, which implies limited investor responsibility, does not allow capital to be raised from many shareholders but benefits in return from low red tape and reduced reporting obligations. According to Guinnane et al., the European popularity of limited liability companies is due to the fact that, contrary to the corporation, they offer a wide range of forms of governance; in particular these companies can better protect shareholders from the misuse of their profits by managers.
16Shareholders and managers are undoubtedly interested in the variety of legal statuses that enable investors to limit their risks, or the company to outlive its founders. But these issues do not naturally translate into one single solution, whose spread in some countries would only have been hindered by archaic states or families, as has long been written. The solutions ultimately adopted were the result of conflicts and debates that differed between countries; we have mentioned the role of social movements and states in this matter, but obviously these conflicts and debates also took place within companies.
Shareholders and managers
17The invention of the legal framework for the development of the large integrated firm, and of this type of business itself, thus took place gradually through debates in which the state played a decisive role. It set the stage and some of the rules of the game for the central conflict of the twentieth century, alongside that between capital and labour: the clash between the professional manager (typically the CEO) and the shareholder. Before looking at the contemporary situation in this conflict, we should briefly outline its emergence.
18The shareholder vs. manager conflict, as was first highlighted by Berle and Means (1932, a jurist and an economist whose book enjoyed immediate success—on the debates that followed, see O’Sullivan 2000) revolves around one simple question: who runs the company, and with what objective(s)? Long before agency theorists (Jensen and Meckling 1976), Berle and Means asserted that shareholders and managers only appear to be allies. The former want the company to be as profitable as possible, while the latter can decide, for example, to increase their prestige and power by growing the size of the company they manage, even if doing so might reduce the profitability of capital. Two rationales, in other words, clash, and we need to understand which gains the upper hand and why: the rationale that claims that the owner of the company’s capital should also be the one who runs it, or one that suggests that when the issues are too complex and the shareholders too many, a body of professional managers develops that ensures the smooth running of these new private bureaucracies.
19The Chandlerian propositions on this matter, which, as we have seen, resolutely take the managers’ side, gave rise to a series of investigations on the transfer of power between shareholders and managers: has it taken place and if yes, when, how and involving which actors and which power relations? Responses to these questions have varied considerably from one country to another; they have often been provided by legal or business scholars and economists rather than by sociologists or historians, who would certainly gain from supplementing or revising such responses, in particular with regard to European countries.
20In the United States, all the empirical studies show that a transfer of power from shareholders to managers has indeed taken place, but that it was much later, and above all, much more confrontational than described by Chandler. For him, the dilution of ownership was at the heart of the development of American big business from the end of the nineteenth century. In fact, this dilution was not fully realised until the end of the 1920s (Lamoreaux 1985), the growth in companies until then being based on self-financing in the United States as well as in Europe. The dilution of ownership and the transfer of power it gives rise to is thus not, again, the direct functional result of the growth in companies, but instead an indirect consequence, the result of power relations. If we follow the argument advanced by O’Sullivan (2000), a historian who came from business studies and who also worked on Germany, we would see this as the result of the confrontation between two groups, managers and shareholders, with the former prevailing over the latter. The antitrust laws of the end of the nineteenth century caused a spectacular wave of mergers, which contributed to building giant companies, but the vast majority of their shares remained in the hands of a small number of investors, who were heavily involved in the management of the companies they owned. To counter this shareholder interference, managers sought to dilute the weight of their participation by manipulating the issuing of shares and the voting rights attached to these. Shareholders did not stand idly by and obtained decisions in their favour from judges and legislatures, but they could only ratify the dilution of capital underway.
21It was also because of political and legal decisions that the managers’ victory was not really threatened before the 1980s: as shown by the jurist (and fierce defender of shareholders) M. J. Roe (1994), a set of laws whose aims were often very different from managing conflict at the head of companies—again, populist inspired laws—permanently consolidated the managers’ position. The aim of these very specifically American laws was, essentially, to limit the control of financial firms— banks, insurers, and investment funds—over other companies. For example, they limited the number of shares in the same company an insurance company could own, they imposed a highly unfavourable fiscal regime for financial investments and they provided very strong protection to small shareholders against takeover bids. The balance of power stemming from these rules was for a long time such that an investor who was unhappy with the way a company was being run had every interest in selling his shares and investing elsewhere: the likelihood of a shareholder emerging who was sufficiently powerful, involved in the running of a company and willing to contradict its management was low to zero.
22This American trajectory, which granted great power to managers from the end of the 1920s, profoundly differs from that of European companies. Once again, this should not be seen as a symptom of backwardness: the biggest American companies in 1913 still had a smaller market capitalisation than many British, German or Japanese companies (Rajan and Zingales 2003). Shareholders—who in the British and French cases were often families—were only deprived of power after intense struggles, as shown by Cheffins (2004) in the case of Great Britain and Fohlin (2005) in the German case (on these questions see also Roe 2003). Cheffins thus highlights the similarities between large British companies of the end of the nineteenth century and those in the United States. In both cases they were family-owned and in fierce competition. Decades later, the distribution of power at the head of British companies had changed radically because, according to Cheffins, British legislation tolerated price fixing until 1956. Therefore, rather than attempting to purchase their competitors, companies sought agreements with them. It was only from 1956 that a wave of mergers began, following which the old industrial families had to sell their shares to institutional investors. This dispersion of capital was accompanied, as in the United States but much later and following a very different path, by a (temporary) seizure of power by managers: after a few decades the concentration of equity in the hands of institutional investors led to a new modification of power relations and of the distribution of value added.
23Reading historical and sociological (and sometimes economic, business or legal) studies of the last thirty years, the elementary constituents of what is sometimes called the large Fordist company appear to be precarious and provisional equilibrium points in conflicts where the shareholder is simultaneously an actor and an issue. This is the case for the invention of internal labour markets or accounting for profits as well as for the separation of ownership and management. These results are based in large part on an equilibrium of power that is always likely to shift. Drawing primarily on economic sociology and organization theory (a very cumulative literature, at least for the United States) we will devote the second part of this paper to what seems to be the principal challenge, in recent decades, to the equilibriums we have just described: the turn towards shareholder value, which we will begin by presenting, before discussing its consequences for managers and for the structure of firms.
Shareholder value and its causes
24The financialization of firms is the general movement by which they are increasingly submissive to financial actors and to the logics they bring and impose (Van der Zwan 2014). More specifically, this movement often relates to the imposition of what Fligstein (1990) calls a new “conception of control”: a particular way of setting a company’s objectives, on the one hand, and of organizing the company to achieve them on the other. The principal symptom of the adoption of a financial conception of control is the quest for a goal that is, in theory, exclusive of all others: the creation of value for the shareholder.
25In the United States, this strategic shift has been very clearly documented (see, among others, Davis 2009; Fligstein 2001; Fligstein and Shin 2008; Lazonick 2010). The 1960s were considered to be favourable for property investment: the profitability of shares after accumulated dividends and capital gains on the value of securities was then 6.63% per year. But in the 1980s, this return rose to 11.67% and to 15.54% in the 1990s (Lazonick 2010). This growth was not the result of increased business productivity but of a shift in the distribution of value added, to the benefit of shareholders. The 2000s, framed by two financial crises, are proof of this: while companies’ finances were stretched, the share of net income devoted to dividends or share buybacks (which help to support prices, thus promoting the realization of gains on share values) rose five-fold. In other words, even, and particularly so, during times of crisis, considerable efforts have been devoted to defending the interests of shareholders.
26The financialization of companies and the turn towards shareholder value has thus been of considerable importance for the United States. The multiple and interdependent causes of these changes fall into two broad categories. The first relates to the changes to legislation from Ronald Reagan’s first term in office, which greatly relaxed the anti-takeover restrictions imposed at the end of the 1960s, very significantly altered taxation on capital gains, abandoned anti-trust provisions that made it impossible to build very large groups specialized in a single market and finally reformed the management of American employees’ pension funds. The second category of causes relates less, as the former does, to changes to the opportunity structures on offer to actors, than to a change in the characteristics of shareholders. Between the end of the Second World War and the beginning of the 1990s, ownership of company equities profoundly changed. In 1965, 84% of the equity in American listed companies was in the hands of individuals, against 16% in those of institutional investors. At the beginning of the 1990s, in contrast, 46% of the ownership of these companies was concentrated in the hands of investment funds (Useem 1996). The money that these funds dispose of still comes overwhelmingly from American employees, but they are no longer the direct owners of company shares: they entrust these sums to funds that themselves own the shares. Ownership is therefore less dispersed, the shareholders more powerful, better informed and better qualified—more likely, in a word, to ensure that companies operate for their benefit. It is hardly surprising, in these conditions, that the share of value added increases in their favour. There was nothing inevitable, however, in the adoption of shareholder value in the United States: the policies benefitting shareholders, such as the share buyback policies, only occurred gradually, over the course of a relatively slow and often confrontational negotiation process between shareholders and frequently very reticent managers.
27The advent of shareholder value and its causes have thus been the subjects of converging and well-documented analyses in the case of the United States. The country’s institutional framework is central in these analyses: there is no a priori reason that the profound legislative changes at the beginning of the 1980s and the transformation of company share ownership linked to the organization of social protection should have happened elsewhere. The European adoption of shareholder value should thus be the subject of an investigation of its own. Two theses that are substantiated to different extents tackle this subject empirically. According to the first, the spheres of influence of American investment funds extend beyond their original boundaries: they have taken possession of companies from old Europe and now impose their logics (Goyer 2006; Lantenois and Coriat 2011, on France and Germany). An importation of shareholder value is thus observed. According to the second, European capitalism resisted this import (Becht and Röell 1999): the ownership structure of major listed companies has not been completely overhauled, nor has the means of recruiting economic elites (Comet and Finez 2010) or labour law. The adoption of shareholder value can, as a result, be more tempered in some countries than in others, and follows more from a conversion of national actors rather than from the imposition of new norms from the outside.
28While more research is necessary to decide between these interpretations, Van der Zwan (2014) emphasizes that work on shareholder value has devoted increasing significance to the institutional heterogeneity of national political economies, thus leaning towards the second hypothesis. In one of the most detailed studies, Fiss and Zajac (2004) thus show that Germany, where the organization of the financial system is based above all on banks, has partially adopted the shareholder value model, but with an intensity and resulting from causes that have little to do with the American case. The shareholder value rhetoric was thus adopted much earlier than the practices it incarnates, and these practices differed greatly depending on the composition of companies’ shareholders. It was not those whose equity was in part owned by American institutional investors who were the pioneers, since ownership essentially remained concentrated in the hands of German banks, the federal state or the Länder, as it was thirty years earlier. What made the difference was that some of these old shareholders themselves embraced the principles of shareholder value, seeing it as a way of increasing their profits. Thus, it was the conversion of the three main German banks (Deutsche Bank, Dredsner Bank and Commerzbank) to shareholder value that drove the companies they were shareholders in to change their management methods. The French case, studied by François and Lemercier (2016), is also evidence of a conversion logic, rather than an importation one. While there are very clear symptoms of the adoption of shareholder value (the share of dividends in net profit increased six-fold between the 1980s and the end of the 2000s), share ownership of the largest French listed companies remains dominated by French financial companies. The social properties of the economic elite have not changed either: trajectories remain similar to those of the 1970s or even the 1950s.
29The increasing similarity between the United States and some European countries shows that despite very different institutional settings, shareholders and their profit are now at the heart of major companies’ objectives. The adoption of shareholder value, however, varies, in its extent, its methods and its causes. The same observation is evident if we look at the consequences of this adoption for managers of companies firstly and secondly for employees and company structures.
Managers against shareholders?
30At first glance, the return to power of the shareholder implies an unqualified defeat for the manager. This is what is shown by some early studies, for example by Rao and Sivakumar (1999) for the United States: company managers devote significant resources to the development of financial communications with the shareholder when they are under pressure from “active” investment funds, i.e., those that are particularly intrusive in the management of the companies they own. However, beyond communication, are managers effectively subjugated by their shareholders, and as such prevented from setting different strategies from those that maximize returns for the latter? The answer suggested by recent studies in neo-institutionalist sociology is more qualified: some indeed suggest that a great number of current managers share the financial conception of the company that is the basis of shareholder value, while others show that managers have retained margins of autonomy (Tolbert and Hiatt 2009)—including to decide, basically, to pay themselves as much as they can.
31One line of work thus highlights that while shareholder value dominates, in the United States at least, it is also because of the career trajectories of CEOs, which implies that some could have been propagandists for this new conception rather than finding themselves confronted with it against their will. Confirming Fligstein’s (1990) older figures, Zorn (2004) shows that the adoption of shareholder value is accompanied by a marked revision of the methods used to recruit managers of large companies. Spending time in financial functions now plays a decisive role in the careers of CEOs, whereas in the 1980s it was production and sales functions instead. We can assume that in these functions managers learn the principles of financialized management and, in particular, management in favour of shareholders. For France, François and Lemercier (2016) make a slightly different observation, insofar as the time they spend in financial functions hardly ever forms a significant part of French CEOs’ careers. However, these short career spells are more common among CEOs and multiple directors of the largest firms in 2009 than they were in 1979. While the trajectories of the French economic elite are otherwise remarkably stable, this change is probably a symptom of the diffusion of shareholder value to the top of firms.
32Other studies highlight that the interest paid to shareholders does not in any way exclude the personal interests of CEOs—even though the shareholder value may remove the temptation to grow or extend a company more than required to make profits. Many studies devoted to the increase in CEO pay since the beginning of the 1980s highlight that according to the principles of agency theory (Murphy 1999, 2013), this pay is increasingly based on mechanisms (in particular stock options) that supposedly encourage CEOs to increase their companies’ share prices (DiPrete et al. 2010; Englander and Kaufman 2004; Bebchuk and Grinstein 2005). This profit-sharing strategy, moreover, sometimes trickles down in the organization structure. Thus, Enron has long been held up as an example of having put in place a decentralized and financialized model that encouraged its managers, at whatever level, to take initiatives to increase the profits of their operations (Froud et al. 2004). In the context of the liberalization of the energy sector, Enron’s pay strategy drove many of its managers to manipulate electricity and gas markets, before—when faced with the collapse of the profitability of their operations—circumventing accounting rules in order to artificially maintain high levels of profits. It has thus served as a borderline case, which is a counter-model for other industrial companies: they are reluctant to generalize such incentives, or accompany them with controls to avoid the development of speculative practices (Reverdy 2011).
33Returning to the CEO, and if we do not limit ourselves to the aggregate measure of an increase in their pay at the point when shareholder value was adopted, we might ask ourselves if they have truly lost as much of their autonomy as they have gained in pay. In the American case, Dobbin and Jung (2010) show that, among the thousand and one techniques supposed to promote the creation of value for shareholders, managers strive to sort those that serve them (stock options) from those that constrain them (boards of directors that are too independent). Alongside the adoption of elements of shareholder value, we have seen new anti-merger mechanisms emerge, such as “poison pills.” After having failed to obtain legislation in this matter, despite draft bills being filed in the 1980s, CEOs chose to act from the inside of the firms they ran (Davis 1991). Many organization theory studies have shown that US CEO pay depends in great part on the margins of manœuvre they succeed in creating for themselves, for example by recruiting individuals who resemble them to boards of directors (Westphal and Zajac 1997; Hallock 1997) or by remaining in position for a long time (Wade et al. 1990; Ocasio 1994; Westphal and Zajac 1994). In the French case, François and Lemercier (2016) show that the high level of remuneration paid to CEOs, and the proportion handed to them in stock options, could be a symptom of their autonomy: indeed this appears to be greater when the shareholders the managers are faced with are weak, i.e., when they control a small share of a company’s equity.
Finance at the heart of firms
34It would be an exaggeration to suggest that the shareholders’ triumph means defeat for CEOs, even if it is true that they are less free to adopt a strategy that is harmful for shareholder profit. But what are the consequences of the adoption of shareholder value for the whole of the firm and its employees? Answering this crucial question remains very difficult: indeed how can we draw a line between the dynamics of financialization and the much older managerial practices that aim to rationalise the allocation of resources and the strategic and organizational choices? The shift brought about by shareholder value occurs at the level of the destination of earnings and not the rationalisation itself. The effects of financialization on the scope of firms, or on their employees are often indistinguishably mixed with those caused by other factors.
The scope of firms
35The growing influence of shareholders on company strategies has prompted a very marked reshaping of the scope and structure of firms. In the United States, the wave of takeovers, mergers and acquisitions in the 1980s contributed to the rejection of the conglomerates that had been the dominant form of organization since the 1950s (Fligstein 1990). Conglomerates have, moreover, been succeeded (for reasons that are not exclusively financial) by the vertical and horizontal division of companies, which has resulted in new forms of modular organization. These enable cost reduction and close control of the profitability of all the business units that now make up companies.
36Situating the adoption of shareholder value in the succession of conceptions of control in American big business, Fligstein highlighted a significant shift within the dominant conception of financial control in the United States since the 1970s. The idea of optimizing the profitability of companies is, in fact, compatible with the preservation and even the expansion of conglomerates (a structure inherited from the earlier conception, defined by Fligstein as commercial). However, from the middle of the 1980s, the dominant legitimate view had shifted toward a focus on the “core competence” of each company. How can this contradiction be understood? Each conception has justifications in finance. In the case of the conglomerate, it is supposed to optimize profitability of the group under the constraint of risk minimization, on the basis of Harry Markowitz’s portfolio theory, which advocates diversification (here, the compensation of losses in one branch by profits from another). In the 1970s, it was this conglomerate organization that prevailed in the United States: only a quarter of the largest listed companies were involved in a single major sector. In contrast, Davis et al. (1994) show that, depending on the sector granularity adopted, the diversification of the largest listed companies reduced by a third or a half during the 1980s. This refocusing was the result of a wave of takeovers of the most heterogenous firms, subsequently made specialized by their buyers, and by preventive specialization by other firms fearing the same treatment. This radical reduction in the scope of companies is based on a change in the convictions of some of the decisive actors in the game, and in particular financial analysts (Zuckerman 2000): the latter now highlight the management costs of conglomerates and the difficulty of assessing them. Portfolio diversification, in reality, is still the order of the day, but its purpose and subjects have changed: in the conglomerate conception, the company was a portfolio (of activities), the diversity of which was managed by its management; in the shareholder value conception it becomes an element of a portfolio (of shares) that the shareholder forms.
37Shareholder value is not, however, the only cause of the disappearance of the huge conglomerates established in the 1950s. They increasingly faced the effects of other, equally profound, dynamics. Piore and Sabel (1984) emphasize changes in demand and therefore in production (Midler 1993; Saxenian 1994). According to them, it was disaffection with standardized goods that explained the move towards less integrated (especially vertically) “network companies” involved in shifting subcontracting configurations. More than a quarter of a century after their description, the move to vertical disintegration has, according to Berger (2006), been radicalized by the establishment of “modular” organizations. In this type of organization, the central company deconstructs its activities into elementary units that it reconfigures ad libitum, deciding, depending on the opportunities offered, to integrate or dispose of them by, in particular, redistributing its geographical involvement. What can be separated out—manufacturing, accounting, customer services and also product design and software development—is thus delegated to subsidiaries, even to companies in countries with low labour costs (Delarre 2005).
38The modularization of companies is undoubtedly not independent of pressure from increasingly demanding shareholders, who find it an effective strategy for specializing in a segment of the value chain, economies of scale, cost reduction and fiscal optimization and thus for increasing profitability (Jacobides and Winter 2005). Thus, according to the current dominant conception of control, the quest for profitability is based on the separation between core competencies (those that make their products unique and are associated with their brand image) and those, less critical, that can easily be delegated to sub-contractors that do not have the same salary constraints. The most extreme form of this specialization strategy is the “factoryless company”, that retains only design activities and delegates manufacturing— this is the case for Apple, which subcontracts all its production to the Taiwanese company Foxconn (Froud et al. 2014). This strategy of radical externalization is motivated by both a profit maximization logic (for Apple, the profit associated with its brand, based principally on software development, product development and design) and the transfer of constraints to sub-contractors (here to a country with low salary costs). Apple makes increasingly large profits and leaves Foxconn to deal with the redefinition of remuneration levels in China that reduces its own margins. Modularization can also be a financial risk control strategy. For example, in the pharmaceutical industry, research and development have been gradually outsourced; they are directly financed and controlled by investors. Gleadle et al. (2008) show that this organization is supposed to offer shareholders greater clarity about risks and profits and enable them to appropriate the surplus value when start-ups are sold to pharmaceutical groups.
39Financialization therefore results in the implementation of a vertical and then horizontal division of organizations, which gives rise to the dismantling of conglomerates. Business activities are distributed among independent business units whose profitability can be much more precisely controlled. They are also much more easily transferable in the mergers and acquisitions market: it is not a matter of acquiring groups in operations that are of necessity uncommon, expensive and risky, but fragments of companies, that can be frequently bought and resold—increasing the opportunities for capital gains. While it may have other causes (Berger 2006), the logic of company disintegration is thus undoubtedly reinforced by financial management principles that have core companies and their shareholders maximize their profits, transferring most economic pressures firstly to their business partners and ultimately to their employees.
Employees facing financialization
40The adoption of shareholder value thus results in a deteriorating situation for employees. In the case of the United States, convergent findings are very gloomy, even though the situation of employees differs markedly according to their recognized skills. The changes in the scope of firms and the focus on core competencies justifies the loss of jobs and redundancy plans (Lazonick and O’Sullivan 2000), even though many studies show that in the medium term, redundancies do not significantly improve company profitability (Fligstein and Shin 2008; Catelli 2000). Those who retain their jobs are subjected to more intense work and caught in the accentuated polarization of labour markets. While those employees considered the best qualified (not only CEOs as we have just seen, but also managers whose skills are less common) gain the greatest material and symbolic rewards for their activities, those who are considered less skilled work for lower salaries and with reduced social protection, while the counterpower of trade unions is increasingly diluted (Fligstein and Shin 2008; Lin and Tomaskevic-Devey 2013).
41In European countries, the negative consequences of shareholder value are often more tempered—although, here too, they are reinforced by more profound transformations of wage labour (Supiot 1999). If the consequences of financialization are less systematically acute in Europe than in the United States, it is because of the strength of counterpowers. In Germany, in particular, the trade unions are sometimes strong enough to counter the logic of financialization (Jackson et al. 2005; Kädtler 2009), as shown, for example, by a comparison of the trajectories of the German chemical giants, Hoechst, Bayer and BASF (Kädtler and Sperling 2003). Hoechst and Bayer have both undergone major reorganizations of their businesses while globalizing. This considerably weakened the trade unions, which struggled to reposition themselves in the new space. In the case of Hoechst, trade unions, supported by local political coalitions, managed to maintain the integrity of the research and development centre in Frankfurt, which adapted to the company’s restructuring. BASF’s trajectory was different; historically its research and development activities were grouped on one site at Ludwigshafen. The skill levels of employees in both types of activity were high, the trade unions were strong (75% unionization) and they succeeded in showing that maintaining the site as a whole was a condition of its performance. Through collective action they neutralized reforms that were too favourable to shareholders.
42There are currently too few systematic and comparative studies for us to extend these German examples to the European area, which is undoubtedly more heterogeneous (for an analysis of British examples, see Froud et al. 2000 and Deakin et al. 2006). Based on econometric methods, Perraudin et al.’s (2008) studies show, in the French case, a situation that is in some respects similar to the US. Human resources management practices of listed companies in France have become increasingly polarized (regardless of the identity of their largest shareholder). A minority of employees have seen their employment conditions (in terms of salaries and access to training) improve markedly, while for others, the use of temporary contracts is more and more common. More generally, listed companies have made labour costs more flexible for all employees, either by the more common use of short-term contracts or by indexing remuneration to company results. This flexibility results ultimately in the growing use of subcontractors where employment conditions can be worse.
43* * *
44The firm is the heart of the conflict between managers and shareholders: it is both the theatre and the issue. It is also the point of confluence in the conflict between capital and labour, and one of the spaces where a compromise can be reached enabling the conflict’s provisional resolution. It remains at the heart of dynamics that affect capitalism, from the emergence of large integrated firms and the contract of employment at the end of the nineteenth century to that of financialized capitalism a century later. The sociology of capitalism necessarily requires a sociology of the firms that incarnate it, people it and confront each other. This sociology of firms, as we have seen, is very unevenly developed depending on the countries and periods investigated. The US case is undoubtedly the best known: fed by multiple disciplinary perspectives—that very often, however, ignore each other—the understanding of the establishment of the profoundly unstable and conflictual ecosystem of the Fordist company and its successive incarnations is more systematic in the US than anywhere else, especially in France. The dangers are great, therefore, if we seek to take into account European peculiarities, of falling back on typologies painted in broad strokes and whose empirical foundations are disputed (Hall and Soskice 2001). Understanding the heterogeneity of capitalism and its transformation is one of the paths that sociologists should continue to follow.
We do not claim to be exhaustive. For other reviews of the literature in English, see Davis (2005) and Davis and Scott (2007, chap. 12 and 13). On the historical aspects, a good place to start in French is Labardin and Pezet (2014); in English, see Cassis (1997, 2007).