CAIRN-INT.INFO : International Edition

1For the past thirty years the French state has been reducing public investment in an economy that had previously been characterized by the public management of a large number of sectors (e.g., armaments, telecommunications, transport, and energy). The waves of privatizations in the years 1986–1988 and then 1993–1994 were seen as one of many signs of increasingly significant withdrawals by the state from economic activity (Morin 1996). Indeed, while the first wave essentially impacted on industry, the second turned towards “historic” companies nationalized in 1946. It also concerned the financial and insurance sectors, with the proportion of bank employees employed by the state falling from 60% to 5%. Finally, the “noyaux durs,” the system of corporate capital interdependence between privatized companies, which made it hard for them to take control of their capital (Loiseau 2002) was relaxed in 1997–1999, acting as the final step in the process. Threatened by changes to the rules of the game and brutal exposure to international competition, the major companies are faced with twin difficulties. Too-obvious a display of dependence on the state hampers their development in neighbouring countries. The second difficulty relates to the uncertainty produced by the instability of legislation governing their activities. Given the role of the state in their share ownership and their governance, many companies are accused of being subsidized by public money and of being unfair competitors, which hinders their territorial expansion. Moreover, the firms need a stable environment to grow (Beckert 2009), which makes their dependence on the state even more crucial when rules are changing at a pace unprecedented in the history of the sector.

2To understand the relations between the state and the largest companies, several approaches are possible. We might look at pricing (Reverdy 2012; Finez 2014). Shifting from a logic of price setting to a logic of the market sheds light on the way in which actors view goods, and the companies that produce them. A second option is to pay attention to the composition of shareholders as well as changes to their behaviour (Zarlowski and Djelic 2005). We might, for example, hypothesize that a public company will not be managed in the same way as a firm controlled by institutional investors. A third option is to study the managers’ careers. In this case we would consider the social characteristics of directors as symptoms of strategies implemented by the large firms, but also as assets for these companies. My thesis is, from this point of view, an instrumental thesis—that managers are resources for the companies that employ them. This choice allows me to encompass all the sub-sectors of an industry that operates according to heterogeneous principles from the point of view of price setting and the shareholder system.

3To do this, I intend to examine the relationship between the state and the large French energy firms. From the creation of the CFP (the ancestor of Total) in 1924 to the 1946 nationalizations of EDF and Gaz de France and the 1970s nuclear programme, the state remained a major actor in this sector until the 1990s. As such, the energy sector is a sector emblematic of state capitalism in post-war France (Kuisel 1984). I will draw on an ad hoc database bringing together the careers of all members of executive committees (ECs) and boards of directors (BDs) of the last twenty years in the largest French energy groups, and on interviews with energy company directors and senior civil servants responsible for regulating the sector. This document base was supplemented with the Direction Générale de l’Énergie et des Matières Premières (DGEMP—Directorate General for Energy and Raw Materials) archives. I will maintain that, firstly, a sociology of elites can be used in the service of a sociology of firms. I will then briefly outline the study’s methodology. Finally, the argument follows the chronology of changes in the sector, setting out how such changes operated in the 1980s, and then the twin progressions followed by the energy sector elites in the 1990s and 2000s.

A sociology of elites in the service of a sociology of firms

4Two principal hypotheses structure my investigation of how directors are selected (Palmer et al. 1986). According to the first, recruitment of economic leaders is a reflection of and the means for class solidarity. It is above all this hypothesis that has been elaborated upon to account for the methods of recruiting directors in France (Jolly 2007). The second hypothesis considers that, for a company, recruiting a director amounts to attempting to increase the resources at its disposal.

5The first hypothesis considers that when recruiting a director, organizational issues are secondary and fade behind (often unconscious) issues that relate more to the preservation of the homogeneity of the social characteristics of the elite group. The resources that enable directors to attain the summits they occupy are heterogeneous. Bauer and Bertin-Mourot (1987, 1997) study the managing directors of the largest French companies and highlight the possession of significant economic capital, loyalty to a company and the asset of the state as characteristic of former civil servants. Of these resources, studies have mainly emphasized directors’ social characteristics that lie outside the organizations they lead. Bourdieu (1989) contrasts private managers, whose reproduction mechanisms occur primarily through economic inheritance, to state managers, whose reproduction is mediated by the pyramidal mechanism constructed through the connections between the state grandes écoles (essentially ENA and Polytechnique), the state corps, and the opportunities they have to head up the largest companies thanks to pantouflage.[1] For a section of the economic elites, the state thus plays a key role in their access to the dominant positions they occupy. It is, however, counterintuitive that when the state is withdrawing from economic life, pantouflages should be more common, former senior civil servants should be just as common at the head of CAC 40 companies (Dudouet and Grémont 2007), and that their role in structuring the network of directors is more critical than it was at the end of the 1970s (François 2010). We can certainly suggest that, at a time when their career prospects are becoming tougher within a pyramidal administration the summit of which tapers more sharply as the state is reformed (Bezes 2005), company headquarters are places for a prestigious career change for senior civil servants. This private sector conversion could become problematic: what are the resources that enable them to impose themselves at the head of firms that are becoming increasingly internationalized as the influence of the state regresses?

6Answering this question requires concentrating less on the social characteristics of managers than on the mechanisms for their selection: how does a company come to recruit this type of leader? Unlike the first hypothesis outlined, this perspective puts the organization at the centre of the argument. For Dudouet and Grémont (2007), this increased control is based on nothing more than itself: long installed at the head of the largest firms, the grands corps[2] would have the self-sustaining ability to maintain their position. This corporatist argument has been greatly qualified by Comet and Finez (2011): working on co-optation within BDs, they show that the links created partially reflect the solidarity of corps; ideological affinities and friendship relations also play a role. In this perspective, it is not the company that chooses directors based on its agenda, but its directors who co-opt based on their affinities. In the American case, Westphalia and Zajac (1995) show that, when they are powerful, managing directors use their power to appoint directors who resemble them, in particular to guard against these new arrivals challenging their power. If we are interested in the causalities this highlights, this first hypothesis makes the company the background rather than the actor. Nevertheless, while the company directors are chosen by virtue of their class or corps backgrounds, they are above all recruited for what they can bring to the firm that employs them. It has been shown that directors are sometimes recruited from a firm’s partners, clients or suppliers, in order to consolidate existing commercial relations. Mintz and Schwartz (1985) concentrated, for example, on the ties the directors allow to be built with the banks their finances depend upon. Other studies consider that this is a way to construct self-regulation of the competition, by co-opting some of these competitors (Pfeffer and Salancik 1978), an explanation that Davis (1996) shows however, is taken less and less into account in the choice of directors, and that the decline of these logics in part explains the collapse of North American networks of elites.

7Here I wish to use this second hypothesis—that, for a company, recruiting a director consists of attempting to increase the resources at its disposal—to account for the recruitment of directors of major French energy companies. More specifically, I draw on two ideas. Firstly, I suggest that if directors are a resource for a company, the type of director recruited will differ depending on the company’s situation. It is often suggested in the literature that firms in difficult economic situations are more likely than others to appoint bankers to their boards (Mizruchi and Stearns 1988): also more likely to need financiers, they appoint them, whilst bankers want to monitor these vulnerable partners more closely. In the case of the energy sector the problem is twofold: faced (at least potentially) with increased competition in their domestic market, the companies have to internationalize—and this internationalization dynamic involves significantly loosening the ties that may still associate them with the state: to find industrial investors and partners, they have to prove to them that the French state is no longer in a position to interfere in defining their strategy. But at the same time the state is a crucial resource in the competitive struggle.

8In order to clarify the idea that companies accumulate resources when recruiting their directors, I depart slightly from the works on which I am drawing. Whether focussed on the ties companies attempt to forge with their suppliers, clients, banks or competitors, these works commonly take as their starting point the partner the companies attempt to control when choosing their directors. I wish to focus above all on the resources the companies attempt to accumulate by recruiting a particular profile. A director can enable the recruiting company to accumulate three types of resource: human capital, social capital and symbolic capital. The first aspect is undoubtedly the most obvious: directors are chosen for their skills and their experience. This is not to argue that the most competent are chosen, but to maintain that directors may be chosen on the basis of their past experience and the skills attributed to them (Marchal and Rieucau 2010). A company will favour the recruitment of a director with knowledge of the banking sector, for example, if it depends greatly on loans granted by this sector for its growth (Mizruchi and Daniel 2005).

9The directors recruited by a company give it the opportunity to accumulate social capital. This is a very common hypothesis in the literature on the accumulation of positions on BDs (Caroll and Sapinski 2010). However, I am less interested in the synchronic ties highlighted by this literature, which uncovers the links forged between organizations and the directors who accumulate positions within them, than in the successive presence of directors in several organizations. For a company, establishing such links with its environment provides access to information on how that environment is evolving (Useem 1984). Some individuals represent a communication channel through which organizations can voluntarily exchange information, playing an intermediary role (Burt 1995). Given the age of the directors considered here (55 on average) we might assume that their social capital was built up over the course of their careers. The social capital transmitted by their families has faded or matured during their careers, which have impact on it (Coleman 1988). Essentially, it is the career that can be used as the most concise indicator of the potential social capital of a director. Finally, directors are status symbols: in other words, they bring their symbolic capital to a company. According to Bourdieu’s argument in La Noblesse d’État (1989) [The State Nobility (1996)], recruitment of students from the most prestigious grandes écoles is due to the prestige associated with them. This argument also informs North American literature on directors of various companies. Thus, Mizruchi (1996) cites legitimacy among the reasons for recruitment, taking up the argument made by Selznick (1957) that BDs fulfil an important role in terms of a firm’s reputation.

10To assess the relative importance of these three dimensions—human, social and symbolic capital—in the recruitment of directors in the energy sector, I propose to examine two bodies and their composition and renewal: boards of directors (BDs) and executive committees (ECs). The recruitment rationales of these two bodies differ significantly. The BD, which brings together people from outside the firm, is charged with guiding the latter’s strategy. It thus ensures the visibility of competent and experienced people whose careers are described in each annual report and highlighted on the companies’ websites. Symbolic dimensions are thus likely to be decisive. The EC directors, employed full-time, are themselves a source of cultural capital for the firm. This hypothesis will be deployed in the last part of this article in particular. In analysing how the composition of these two bodies has evolved in energy companies, I will show that the recruitment of their directors is a means for them to accumulate capital in a profoundly changed competitive environment.

Energy sector directors

A sectoral approach

11Many studies of the energy sector (Reverdy 2012; Hecht 2009; Nouschi 2001) focus on a sub-sector, such as electricity or oil. But if we want to take into account the role of the state, whose energy policy is precisely to link and coordinate the different means of energy production, the whole of the sector should be studied. Two arguments support this approach. From the point of view of the state, the companies considered here form a well-defined area of public policy and share the same supervisory authority, the DGEMP. Its directors represent the state on BDs where it has holdings. From the point of view of these senior civil servants, the companies are tools, instruments enabling the implementation of an undifferentiated energy policy. The firms are linked by strong shareholder ties to the state, which controlled all the large energy firms until the end of the 1980s. From the firms’ point of view, there are many links between the different sub-sectors, be that in terms of financial contributions or personal ties. Thus, by concentrating on a particular sector of the economy, we can look at its specific dynamics and the social characteristics of its directors more clearly than if we were interested in French capitalism overall. By focussing on energy, I isolate a set of phenomena (e.g., liberalization, privatization, internationalization, etc.) characteristic of the period under consideration and interpreted as a manifestation of the withdrawal by the state from company management.

12The energy sector is unique in that its operators share a large number of directors, through their BDs. Thus, during the period studied, the companies under consideration are linked together by directors sitting on the boards of several of them. Financial links, in particular between operators and suppliers—oil firms and oil services firms, and nuclear operators and nuclear construction companies—complete this picture. This commonality, besides the fact that it indicates a certain sectoral coordination and where appropriate common governance, ensures a certain homogeneity between companies in the sector. Finally, in France the energy sector is highly concentrated: the gas, coal and electricity sectors are monopolies; and the oil, oil services and the nuclear construction sectors having been dominated, since the beginning of the 1990s, by one or two operators, correspond as such to the schema suggested by Fligstein (2001). This concentration ensures that, by focusing on the ten biggest companies in the sector, practically all the private sector positions of power are covered. Since the beginning of the 1990s, the energy sector has been dominated by ten or so operators with close ties to the state, overseen by the same supervisory authority, whose governing bodies are linked by common directors. Is it, however, apposite to put the directors of these companies in a common category (energy sector directors) on the same level: to bundle the head of a company with 5,000 employees, such as Technip, with one of a multinational, such as Elf? The careers of directors moving from one operator to another are consistent: a director of COGEMA (the company from which Areva stemmed), such as François de Wissocq, joined Elf’s executive board in 1990. Pierre Vaillaud, number two at Total, became head of Technip in 1992. The difference in size of the two companies does not prevent them from exchanging executives. This takes place within ECs, bodies I propose to take into account.

Definition of the population

Company selection

13Depending on the period, this study encompasses the 5 to 8 largest companies in the energy sector (see Appendix 1). They were selected according to two criteria: they are among the largest companies dominating the energy sector; and more than half of their revenues come from activities relating to energy. COGEMA is a company specializing in uranium mining and the manufacture and reprocessing of nuclear fuel. It stems emerged from an industrial subsidiary of CEA in 1976. It merged with Framatome (a nuclear boiler manufacturer) in 2001 to become Areva. Elf Aquitaine is a group whose main activities are oil exploration and production and the refining and distribution of oil products. Created in 1976 by the merger of two public companies, Elf remained in public hands until 1994. Total, another French oil company, with a historically less marked state presence, merged with Elf in 1999. EDF and Gaz de France were created in 1946 by the merger of commercial entities active in the electricity and gas fields. They held monopolies over the distribution (EDF only), production, importation and exportation of electricity and gas until 2000 and 2001 respectively. Suez emerged from the nationalization of the Suez Canal in 1956. For a long time a holding company, it has diversified since the beginning of the 1990s into the fields of water and electricity. Its progressive focus on these areas justifies its presence in the database from 1996, the date at which these activities represented 50% of its revenues. Gaz de France merged with Suez in 2008 to become GDF-Suez. Finally, Technip is a company specializing in oil infrastructure construction. Relatively small in the 1990s in comparison to the other companies selected, it grew during the period under consideration through external expansion, the most notable of which was the acquisition of Coflexip in 2000 (which was not taken into account because of its size and lack of information on its directors).

14I have excluded Alstom, responsible for the construction of the conventional part of nuclear power plants, but characterized by a conglomerate structure with substantial activities in transport (Alstom is the TGV constructor), which makes it difficult to identify it as a pure energy company and necessarily has an impact on its relations with the state. Finally, with the scope of this study being the major French energy companies, I have excluded the companies that emerged with the liberalization of the gas and electricity markets, such as Direct Énergie and Altergaz.

Which individuals?

15Three methodological approaches are distinguishable in the literature on large populations of directors. The first approach concentrates on CEOs, who are more easily identifiable (Bauer and Bertin-Mourot 1997, 2007). However, the rationale underlying the preeminent role given to CEOs has been eroded in recent years, as firms have redefined their operations (Mizruchi 1996). In many companies during the 1980s and 1990s, CEOs set up teams bringing together the most senior managers, with the aim being for them to meet frequently and to take decisions jointly. [3] The importance of these teams is reflected in the firms’ annual reports, where a list of executive officers appeared systematically in the 1980s. Recent studies have chosen expanded populations as their subjects, such as the surveys by François (2010) and Comet and Finez (2010), which focus on the directors of the one hundred largest French firms. But these are most often CEOs or COOs of very large firms, and if we only select directors we screen out most of the EC. At least this is the case for the energy sector, where few executive directors have non-executive director roles. Finally, there are very few studies that include ECs. Dudouet and Joly (2010) take them into account but restrict themselves to companies in the CAC 40. Levy-Leboyer (1979) includes members of BDs but does not make a distinction between executive directors and non-executive board members, which is crucial to this project. Their random selection for a population of CAC 40 directors favours the creation of a representative sample at the expense of sectoral coherence.

16The directors were identified based on lists provided by the firms in their annual reports. The biographical data concerning the directors mostly came from Who’s Who France. The entries in this directory rarely mention party political membership, even when this is known as for E. Alphandéry, former economy minister for the UDF, which would lead us to underestimate political and trade union activity and explain their absence from our study. The principal bias observed concerns the overrepresentation of senior civil servants in the 1990s in the database constructed, due to the use of Who’s Who (which favours public service and thus former senior civil servants). For the more recent years, it has been possible to overcome this shortcoming by using the biographies contained in annual reports. To assess the representativeness of the database constructed, it seems more interesting to think in terms of positions occupied rather than people. Effectively, the same director can occupy several positions in different companies simultaneously. Thus, 435 directors occupy 608 positions during this period. The statistical analyses on which this study draws is based on 80% of positions occupied.

The historical domination of the grands corps d’État

A sector historically linked to the state

17Analysis of the data drawn from annual reports of the main firms show consistency in the recruitment of members of ECs during the 1980s (Table 1). More than 70% came from the senior civil service, a statistic that remained stable throughout the decade. They mostly had a “technical” profile, with a third of them coming from government bodies responsible for energy policy (DGEMP, CEA, IFP). Furthermore, only 25% to 30% had experience of a ministerial cabinet (among a high population within the central government hierarchy). This background was considered legitimate, prestigious even, as evidenced by the titles that accompany the names of the former senior civil servants in annual reports: “ingénieur général des mines” (general mining engineer) or “inspecteur général des finances” (inspector general of finances). Not only is the use of these titles an indicator that the energy sector directors from the grands corps d’État were proclaimed by their companies, but that their roles were considered an extension of their careers as senior civil servants when they were in the service of the state, as highlighted for example by Suleiman (1979). This domination by senior civil servants was also reflected, until the beginning of the 1990s, in the major prestige accorded to representatives of government, prestige that was directly linked to the power granted to them. Thus a former government commissioner recalled in an interview that before the liberalization of the electricity market, the representative of the state sat to the right of the EDF president in meetings of the BD.

Table 1

Careers of energy sector directors in the 1980s (BD and EC combined)

Table 1
Foreign (%) Civil servant (%) Energy department (%) Cabinet (%) Élysée/Matignon (%) Obs. 1980 0 71 35 27 10 51 1984 0 73 30 27 11 64 1988 0 74 30 32 12 57

Careers of energy sector directors in the 1980s (BD and EC combined)

Interpretation: In 1980, 35% of directors had spent time in the department of energy and 27% in a ministerial cabinet.

18Until 1992, the date of the state’s withdrawal from Total and the abandonment of monopoly responsibility for oil, the state very largely dominated a sector that government departments had played a key role in creating, or at least organizing (Jobert and Muller 1987). The main groups in this strategic sector were public or had a shareholding largely dominated by the state or its agencies. Thus, while EDF and Gaz de France were public companies, COGEMA (the company from which Areva originates) was controlled by the CEA (Commissariat à l’Énergie Atomique— Atomic Energy Commission). Elf was 65% controlled by the state, which was a minority shareholder in Total and Framatome, Finally, the companies in the oil services sector, such as Technip, were controlled by the IFP (Institut français du Pétrole [French Oil Institute], a para-public institute for research in the field of oil) and its equity was also held by the two oil groups. This control of capital gave the state the power to nominate the directors of these companies, whether to BDs or ECs. The BDs of public companies (Gaz de France, EDF, and COGEMA) were organized on a tripartite principal: the board contained six state representatives, from different departments, six employee representatives and six “qualified persons,” generally the directors of major French companies. It was also the Ministry of Industry that chose the qualified persons, according to proposals from the energy department (20090049/3). [4] Moreover, the state, through the intermediary of the DGEMP, had representatives on the boards of Total, Elf and Framatome, in proportion to the voting rights granted to it as a shareholder. The BDs were thus characterized in the 1980s and the beginning of the 1990s by the number of senior civil servants within them. Finally, it fell to the government to nominate the executive directors of public companies that had long been managed by a director general–president diarchy. This period corresponds to one of domination by a group of directors from the senior civil service of the governing bodies of major energy groups: both BDs and ECs. This domination was exercised by senior civil servants acting for departments of the Ministries of Industry and the Economy and Finances, as well as by former public service leaders retrained in the market sector.

19Since the 1990s, we see (Table 2) that the profiles of directors and executive directors differ. The directors more often come from the civil service (representatives of the state are chosen from within the senior civil service). There are very few foreigners in these governing bodies. Moreover, most of the directors from the senior civil service come from ministerial cabinets (64%), and a large proportion of them from the Élysée (the president’s office) or Matignon (the prime minister’s office) (21%). This should be seen as a reflection of the careers of civil servants at the head of the most prestigious departments, the Treasury and the Budget, for example, who sit there by right.

Table 2

Profiles of energy sector directors in 1992

Table 2
1992 Foreign (%) Civil Servant (%) Energy department (%) Cabinet (%) Élysée/Matignon (%) Obs. BD 01 90 52 64 21 83 EC 00 69 36 39 14 48

Profiles of energy sector directors in 1992

A senior public servant, Philippe Rouvillois

Son of a general, born in Saumur in 1935, P. Rouvillois went to secondary school in Strasbourg, then to the Lycée Louis-le-Grand in Paris. After studying law at the IEP in Paris, he joined the Vauban class at ENA. He then joined the Inspectorate of Finances. He had been the head of mission at the Direction Générale des Impôts for two years when he was appointed to Michel Debré’s cabinet, at the Economy and Finance Ministry, as a technical advisor. From 1968 to 1987, he moved up through the ranks of the Direction des Impôts, which he became head of from 1987 to 1988, the date at which he also chaired the SNCF board of directors. It was only later that he joined the energy sector: from 1989 to 1995, he was CEO of the CEA and President of CEA Industries. He left shortly before the creation of Areva in 1999, of which he became a director in 2002. During this time, he was also a member of the Conseil Économique et Social (Economic and Social Council) and a director of COGEMA and Framatome.
(Source: Who’s Who France 2009)

20The system described here, in place until the 1990s, the date at which notable changes took place (see Tables 4 and 5), operated according to the logic identified by Bourdieu in La Noblesse d’État. During this period, companies recruited their senior staff from members of the most prestigious corps d’État (Corps des Mines, Inspection Générale des Finances). The public companies were in a dual dependent relationship with the state: not only was the latter their main shareholder, but until this time it also had a monopoly on the attributes of legitimacy.

Liberalization, privatization, internationalization: the upheavals of the 1990s–2000s

21In the 1990s and 2000s the energy sector was changing completely. The privatizations from 1986–1988 then 1993–1994 were the first constraint on the model that had prevailed until then. The withdrawal of the state from corporate capital began in the energy sector with the disposal of its stake in Total, 40% of the voting rights of which were still controlled by the state in 1990. The state then withdrew from Elf Aquitaine, of which it owned only 13% of the capital in 1994 (as opposed to 65% in 1986). This retreat was also evident beyond the oil sector, with these two major companies having large holdings in the oil services companies such as Technip and Coflexip from which they had progressively withdrawn in the 1990s. Under the 2004 and 2005 laws, the shareholdings of EDF and Gaz de France were opened up to the public and both companies entered the stock exchange. Finally, the merger of Suez with Gaz de France led the state to abandon control of the public group, and to limit itself to a minority shareholder role. In such a context, the large energy companies were no longer a natural outlet for senior civil servants. Moreover, the legitimacy of these senior civil servants’ career paths was undermined by the decline in state economic intervention.

22The European integration process, reinvigorated by the Maastricht Treaty, led the European Commission to propose electricity and gas market liberalization directives to member states (Jabko 2009). Under threat of legal action under competition directives if the political process was not completed, governments of member states reached an agreement in 1996 after nine years of negotiations. The 1996 and 1997 directives were implemented in France in 2000 and 2001, through laws on the modernization of the electricity and gas sectors (Chevalier and Percebois 2008). This liberalization transformed the legal monopolies in distribution, transport, importation and production, enjoyed by EDF and Gaz de France since the 1946 nationalization laws, into de facto monopolies. In the oil sector, the 31 December 1992 law that created a large European market for oil products rendering obsolete the 30 March 1928 oil law that subjected the oil sector to an authorization regime, was also largely driven by the European Commission. This process was reinforced by the New Organization of the Electricity Market law of 2010, which expanded the electricity market liberalization to EDF’s means of production.

23The third change concerning the energy sector is linked to the location of the firms’ activities. To address potential loss of market share in France, EDF and Gaz de France embarked on an internationalization process that led in 2013 to them generating less than 50% of their revenues in France (Figure 1). Facing a drop in demand in France, Areva also proceeded to gradually internationalize its activities.

Figure 1

The internationalization of energy sector companies

Figure 1

The internationalization of energy sector companies

Key: As solid lines: COGEMA and Framatome, then Areva. As a dashed line, Suez. As dotted lines, EDF. Gaz de France then GDF Suez are alternating dots and dashes. Finally, Technip is at the top.

24These three developments place the energy companies in a situation of great uncertainty (Reverdy 2012), largely unseen since the end of the Second World War. The challenges they have to deal with occur on two fronts. Facing (at least potential) increased competition in their domestic market they have to internationalize—and this move to internationalization involves in particular the requirements on them to significantly loosen ties that could still connect them to the state: to find investors and industrial partners, they have to prove that the French state is no longer in a position to interfere in defining their strategy or in their management. But the energy companies are simultaneously confronted with a radically changed environment, which in particular exposes them to increased competition. If we acknowledge, as in the hypothesis above, that the directors of firms are a resource in the competitive struggles in which they are engaged, how are these two issues likely to change the social characteristics of those who lead them?

A career space specifying the resources available to companies

25Following the profound transformations in the 1990s–2000s, energy sector leaders still came from the same social backgrounds. Over the two decades, more than 30% of them had a father who was a senior civil servant, industrialist, or a director of large company, and these proportions remained stable. [5] More generally, the stratification of the social origins of these directors was stable throughout the period, with less than 20% of directors coming from the working or middle classes. However, the careers followed by their heirs changed substantially during this period. The social characteristics of directors of energy firms form a very different space, which can be described using multiple correspondence analysis.

Table 3

Description of the MCA active variables

Table 3
Public sector career (6 variables, 19 categories) Career mainly at: the DGEMP, the Ministry of Foreign Affairs, Ministry of Economy, the Ministry of Industry (excepting the DGEMP), the Ministry of Transport or Works, the Ministry of Research, in the ENA grands corps, at the Ministry of Interior, or in ministerial cabinets. Worked: —in a right-wing cabinet. —in a left-wing cabinet. Has held political office Worked : —at the IFP. —at the CEA. Private sector career (2 variables, 12 categories) Mainly in the energy, finance,consumer, mechanical and electrical goods, telecommunications, construction, or industry sectors. Number of companies in which a director has worked: 0, 1, 2, 3, 4 and more. Participation in governmental or intergovernmental commissions(4 variables, 8 categories) Sat on a governmental commission (Conseil d’Analyse Économique, etc.). Sat on an intergovernmental commission (Agence Internationale de l’Énergie, etc.). Runs a think tank. Has responsibility in an employers’ association. French nationality (1 variable, 2 categories)

Description of the MCA active variables

26The variables used to build this space match several sets of assumptions about the careers of energy sector leaders. The first set of variables describe the public sector career, the directors of major groups having the characteristic of coming from a grand corps d’État and thus having worked in government and possibly gained ministerial cabinet experience. This set corresponds to the hypothesis that companies need resources from the civil service or the public sphere and thus recruit directors who have a good knowledge of this world and relations with it. A second set of indicators aims to reveal the sectors from which directors from the market sector originate. Here I therefore investigate the energy firms’ dependency on the market sector. I also count the number of firms within which an energy sector leader has spent his career, in order to characterize those whose asset is their loyalty to a company, according to Bauer and Bertin-Mourot’s (1987) expression. Here, it is the fact that a director is recruited on the basis of his legitimacy (internal and due to good knowledge of the company), which is investigated. The third group describes the involvement of directors in various commissions and BDs of public institutions and establishments, according to Useem’s (1984) government scan hypothesis demonstrated in the United States, whereby many directors in the market sector sit on commissions in order to maintain links with government or politics. Although little discussed in the literature on elites, this hypothesis seems to me to be essential to understanding the substantial number of directors with public careers. The final set is interested in energy sector leaders and their possible international experience. I thus draw on the hypothesis that, in a context of substantial internationalization of the sector, energy companies appoint individuals with knowledge of international markets and related social capital. These different hypotheses define the 13 active variables and 41 categories defining the careers of 324 directors whose careers are known.

Figure 2

Characteristics of energy sector managers

Figure 2

Characteristics of energy sector managers

Note: Main plan (axis 1-axis 2) of the multiple correspondence analysis.
The categories that only contribute marginally to the axes are not plotted for reasons of readability. Some directors’ names are plotted in roman letters. John O’Leary’s career is presented on page 334. Anne Lauvergeon became head of Areva in 1999, after a short career in the Corps des Mines, ending in the position of deputy secretary general at the Élysée. Bernard Dupraz is an EDF director, joining after a short career in government (energy). Claude Mandil led the DGEMP during the 2000s after a long career in government and in cabinets. He has sat as a representative of the DGEMP and then in a personal capacity on many BDs.
Interpretation: In bold, the categories contributing the most to axis 1. In italics, those that contribute most to axis 2. In interpreting the axes, other variables that contribute but which are slightly under the chosen threshold are included.

27The first three axes defined by the MCA represent 7.8%, 6.6% and 5.4% of the variance respectively. The first axis drawn from the analysis is mainly constructed around an opposition between the career traits specific to former senior civil servants and the characteristics of directors from the private sector. Thus, past membership of government departments from which public company directors come (e.g., MFA, Ministry of the Economy) and the fact of having never left the civil service are the variables most strongly positively associated with the first dimension. In contrast, the variables acting negatively are those characterizing private sector careers or the fact of being a foreigner. This axis reflects an opposition revealed by Bourdieu (1989) between state managers and private managers. The fact that this dimension is the most structuring shows that the tension described above between opening up to competition—international competition in particular—and seeking state support is the most important for the businesses. The second axis structuring career space is determined by an opposition between characteristics of directors who have worked in the energy sector (e.g., energy sector, DGEMP, CEA, IFP) and the characteristics of directors who have built a career in the private and/or public sector in other industry sectors (e.g., consumer goods, ENA grands corps). Here we also find the attribute identifying directors who have worked in at least four companies, a characteristic that is often shared by financiers in our sample. This opposition clearly marks the strong identity of the energy sector, within which a certain number of directors spend their whole career, whether in government or in companies. The association of companies and government departments within the same cluster demonstrates the circulation of individuals through these various spaces. Something that is also positively associated on this axis is involvement in commissions (e.g., think tanks, governmental commissions, etc.), which can be interpreted as the indicator that the individuals in this clusters meet in these spaces, which mix private and public sector directors. According to my hypotheses, this second axis thus indicates knowledge of the energy sector. The final career space dimension indicates a more subtle opposition (Figure 3). This axis opposed two types of resource for companies in the sector. They have to choose between directors familiar with how they are governed (e.g., DGEMP, CEA, and IFP) and the energy sector, and directors who are more distant from the heart of the industry, but whose resources could be crucial. The latter are from the political sphere among others (directors from the political sphere, or having spent most of their career within ministerial cabinets or employers’ associations), actors whose competence lies in their address books or detailed knowledge of the machinery at the top of government. Another characteristic of these directors to the right of axis 3 is their involvement in international space (e.g., career in the MFA, participation in an intergovernmental commission), at a time when French energy companies are expanding internationally. These directors are also from the financial sector, whose support is crucial in companies’ expansion phases and the telecommunications sector (in the private or public sector “Min corps des Ponts”), which has implemented the same reforms as the electricity and gas sectors and as a result is a reference point. In general, the cluster on the right of axis 3 groups together the directors who represent the resources needed to expand energy company activities internationally.

Figure 3

Properties of energy sector directors

Figure 3

Properties of energy sector directors

Note: Secondary plan (axis 1-axis 3) of the multiple correspondence analysis. In bold, the categories contributing the most to axis 1. In italics, those contributing the most to axis 3.

28I have chosen to plot the characteristics of all individuals from the period under consideration (1992–2010) in a single space. In order to account for changes to the social characteristics of energy company leaders, the variables indicating the presence of a director within the database for several reference years were used as passive variables. This addition enables us to revitalize a method whose static aspect is recognised as one of its limitations (Duval 2013). The temporal variables (Figures 2 and 3) indicate a dynamic along axes 1 and 3 that can be interpreted as the replacement of senior civil servants by directors from the private sector, and more importantly having international experience, but also as a modification of senior civil service careers, which are more political and further away from the energy sector (axis 3).

29The social characteristics space that enables us to describe the careers of energy sector directors is thus arranged around dividing lines partially covered by temporal dynamics. Between the end of the 1980s and the beginning of the 2010s, the dominant profile has come to be primarily directors from the private sector and from international careers rather than directors from the senior civil service, who are necessarily more domestic. These developments within the population of directors show that the 2010 energy company directors are not the same as those in 1990. Corporatism alone is thus not enough to explain the careers of those from the grands corps, of whom there are far fewer in 2010 than in 1990 (Tables 4 and 5). This transformation of the relative weight of career profiles between the beginning and the end of the period can be interpreted as a functional response to two issues energy companies must face. Required to gain a foothold in new international markets, they replace former senior civil servants with foreign directors to demonstrate the loss of the French state’s influence in defining their strategies, and simultaneously to benefit from accumulated experience and social capital that is more relevant to gaining a foothold in these new markets. Moreover, faced with rapid developments in the regulatory and legislative environment, they take advantage of directors with knowledge and social capital that are likely to guide them through these processes, even intercede on their behalf. This is how the comparatively consistent retention of directors who are former Élysée and Matignon advisors can be interpreted (11% in 2014 against 14% in 1992, Table 4).

Table 4

Demography of ECs

Table 4
Foreigner(%) Civil servant(%) Energy department (%) Cabinet(%) Élysée/Matignon(%) Obs. 1992 00 69 36 39 14 48 1995 02 58 31 31 03 49 2000 10 49 24 34 15 48 2004 13 23 23 33 14 54 2009 14 23 08 23 10 56 2014 16 28 18 20 11 56

Demography of ECs

Table 5

Demography of BDs

Table 5
Foreigner(%) Civil servant(%) Energy department (%) Cabinet(%) Élysée/Matignon(%) Obs. 1992 01 90 52 64 21 83 1995 02 82 50 47 14 81 2000 16 67 31 39 15 90 2004 23 64 28 45 24 66 2009 22 62 35 51 24 67 2014 25 52 30 36 18 44

Demography of BDs

30The companies’ internationalization strategies have driven them to recruit more foreign directors: the proportion on BDs rising from 2% in 1995 to 16% in 2000. These distinguish energy sector firms from their historic dependence on the French state and allow them to rely on acquaintance networks outside and beyond the borders of mainland France. “We have taken measures: we have brought a German onto the board from Bavarian electricity” (a former EDF director concerning the internationalization of the EDF in the 1990s). These foreign directors essentially sit on the BDs of major French groups. Half of them have worked in the energy sector, which confirms the hypothesis that French companies need to acquire information on European energy markets; it also serves to indicate their willingness to collaborate with future international partners. A third quarter have spent most of their careers in the financial sector, which provides a comprehensive view of the economy.

31Following on from these initial analyses, it is tempting to conclude that there have been similar developments in BDs and ECs. We have omitted the fact that the state controls nominations of directors in the companies they hold equity in, but not the members of the ECs. Among these leaders are found mostly the directors of the largest branches of government—such as the director of the Treasury—who have spent time ministerial cabinets, and more especially in Matignon and the Élysée. This explains the large proportion of directors in 2014 who have spent time in cabinets (36 %), or in the Élysée or Matignon (18 %).

Twin shifts which have drastically changed the foundations of the sector

32To clarify the scope and robustness of these initial results, it is necessary to return to one of my initial hypotheses, the difference between BDs and ECs. By omitting this up until this point, any divergent dynamic between these two institutions was excluded. Firms were also considered in the same way, whereas I emphasize their differences in the second part of the article. To better understand the developments to these two bodies, two new sets of hypotheses will be introduced in order to understand the connections between the two highlighted dynamics depending on the characteristics of firms and the directors concerned. Undoubtedly, if the replacement of directors with a public sector and national profile by those with private sector and international profiles is a way to tackle the increased internationalization of activities, which has imposed a dramatic reduction in ties to the state, then this substitution must be more marked in the most internationalized companies. Similarly the replacement of directors with skills honed in the energy sector for more “political” directors does not affect all firms in the same way. But other dynamics can have similar implications: independent of the internationalization of their activities, energy sector firms have seen a very significant reduction of their proportion of their shares owned by the state—although the proportions owned are very unequal from one company to the next—and if, as we have seen, the preponderant proportion of senior civil service among energy sector directors has long reflected direct control by the state of the companies they manage, then a reduction of this control could also result, ceteris paribus, in a loosening of the former senior civil servants’ grip. Finally, energy firms are not isolated from firms of a comparable size operating in other economic sectors. Thus, this article will seek to take into account the recruitment practices of companies of a comparable size in France, which practices constitute the institutional environment of governing bodies. The composition of energy company governing bodies will depend on this pool, whose practices will constrain recruitment, whether they serve as a reference to recruiters, or whether imposed upon them because of the need for legitimacy (DiMaggio and Powell 1983).

33The second set of hypotheses I wish to test concerns not the companies but their directors. The directors may constitute resources for the companies they manage in three ways: They may have a specific skill (the human capital argument), an effective network (social capital) and function as status symbols (symbolic capital)—and the companies they govern can benefit from these three types of resources. Presumably, the relative importance of these resources in the recruitment process has an unequal effect depending on the role assigned to the director. More specifically, the status and symbolic roles will be more marked for the most visible directors, those on BDs, than for the more operational managers, those belonging to ECs. Here we may also hypothesize that the replacement of public profiles by private ones concerns would concern primarily those directors whom we might consider to have the most significant proportion of symbolic or status resources among their portfolio of resources gained in companies. If we acknowledge that non-executive directors are, more than members of ECs, likely to be chosen as symbols of a company strategy, then this substitution is likely to affect them more than executive directors.

34To test these two sets of hypotheses, I constructed two regressions whose dependent variables are, on the one hand, the proportion of former senior civil servants on BDs, and, on the other, the proportion on ECs (for a given year): the larger this proportion, the lower the replacement of public sector profiles by private sector ones in a given company. It is based on a hundred observations describing the proportion of senior civil servants on a given governing body within a given company for a year ranging from 1990 to 2010. The companies under consideration are those defined in the second part of this article. As for the explanatory variables, they consist of three sets. The first measures the extent of the companies’ international integration, taking into account the proportion of revenue earned abroad. The second group relates to the place of the state. The first variable describes the proportion of a company’s shareholding held by the state. The second describes the historic proportion of shareholding held by the state measured before the first privatizations in 1986. Lastly, a final variable describes the institutional environment by measuring the proportion of CAC 40 CEOs from the senior civil service for each year. [6] The observations are highly correlated with each other inasmuch as the composition of governing bodies only marginally changes from one year to the next. Yet the ordinary least square model assumes the observations are independent. To take into account the problem of dependency between the observations, I therefore propose to introduce fixed effects for the different companies considered, thus absorbing part of the observations’ dependency.

Legitimation through boards of directors

35In this first model, three variables impact significantly on the proportion of former civil servants on BDs. The share of a company’s equity owned by the state or historically owned by the state both play a part, one positively, the other negatively. The number of CEOs from the senior civil service within the CAC 40 is also significantly positive. Finally, the variable indicating the internationalization of the company under consideration is not significant. The hypothesis linking the markets in which a company is developing and director recruitment is not confirmed, contrary to what the MCA results would lead us to expect. Internationalization does not therefore directly contribute to removing senior civil servants from BDs. Within the CAC 40—the main recruitment pool for company board directors of the companies studied—the number of CEOs from the senior civil service has declined over the past twenty years, which impacts on the composition of BDs, but only marginally. This is explained by the low number of executive directors from CAC 40 companies on the boards of energy companies, above all at the end of the period. The composition of a BD is above all linked to the role of the state in the shareholding of a company: the more significant the shareholding, the more civil servants on the BD. The overabundance of former civil servants on some BDs is explained by the fact that they represent the state, which continued to be a significant shareholder in the sector at the end of the 2000s. In contrast, once free from the constraints of a state shareholding, companies seek to gain independence from directors who are well integrated in government networks (negative coefficient, associated with the historic state shareholding variable). This strategy enables a firm to tackle accusations of unfair competition that are particularly grave in international markets by demonstrating a clean slate through the composition of its BD, which is reasonably representative of the balance of power within a firm. This finding is consistent with the demands by the CEO of Elf and Total, calling for a withdrawal of the remaining directors representing the state, despite the existence of agreements [7] linking the latter to the companies. More generally, the role of the BD should be understood as the board overseeing the company’s shares on behalf of shareholders. A CEO seeking to extract himself from the grip of the state (and especially the image of a company intimately tied to the state) will therefore seek to minimize the number of former civil servants on its board.

Table 6

Linear regression on the proportion of former senior civil servants on energy company BDs (standardized coefficients)

Table 6
Proportion of directors from the grands corps on BDs Model 1 Coefficien Error type State share of shareholding 0.74*** 0.17 Number of CEOs from the grands corps in the CAC 40 0.18** 0.06 Historic state share of shareholding –1.00*** 0.26 Share of company revenue earned abroad –0.13 0.10 Constant 00.05

Linear regression on the proportion of former senior civil servants on energy company BDs (standardized coefficients)

Note: *** p < 0.0001; ** p < 0.01; * p < 0.05.
R2 0.71; 105 observations.

36Within the group of firms considered here, EDF and Total are the furthest apart. I intend to compare how they have developed. In 1992, they looked towards France, and no board director was foreign (Table 7). Senior civil servants dominated the two companies, comparatively less for Total, which did not have the state as its principal shareholder. In 2014, a significant decline in the number of senior civil servants is observable, at both Total and EDF. Moreover, the biographies on Total’s website tend to underplay the public careers of directors. Thus, the following appears concerning Michel Pébereau, who headed the Ministry of the Economy and Finances cabinet from 1978 to 1980: “Mr. Pébereau has held various positions in the Ministry of the Economy and Finances, before becoming in turn director general and then CEO of Crédit Commercial de France.” (Source: Total SA website, consulted on 6 October 2014.). It is above all his experience as a banker that is highlighted. The significance of the numbers of former civil servants at Total is made sense of by the “average” profile of CAC 40 CEOs from among whom Total recruits its directors.

Table 7

The composition of the EDF and Total BDs (%)

Table 7
Board of directors Foreigner Senior civil servant Ministerial cabinet Élysée/Matignon Total 1992 0 62 62 50 EDF 1992 0 69 31 31 Total 2014 23 38 31 07 EDF 2014 15 46 38 23

The composition of the EDF and Total BDs (%)

John O’Leary, an international director at Technip

John O’Leary has, since January 2007, held the post of Chief Executive Officer of Strand Energy (Dubai), a company involved in seeking out investment and development opportunities in the oil and gas sector, and also sits on the Supervisory Boards of Huisman Itrec and Jumbo Shipping. From 2004 to 2006, he was a partner in Pareto Offshore ASA, a Norwegian company specialized in advising customers in the exploration/production sector. In 1985, he joined the Forasol-Foramer group where he successively held the posts of Development and Partnerships Manager (1985–1989) and Vice Chairman for Marketing (1990–1997). After the takeover in 1997 of Forasol-Foramer by Pride International, a company specialized in onshore and offshore drilling, he became the President of the new group until 2004. He began his career as a trader in the Irish National Petroleum Corporation (1979–1980) before joining Total as a drilling engineer (1980–1985).
John O’Leary is a graduate of Trinity College in Dublin, the University College in Cork as well as the Institut Français du Pétrole.
(Source: Technip website, consulted 16 September 2014)

37The recruitment to BDs of foreign directors, or from the market sector, is based on a logic of displaying their independence vis-à-vis the state. Recruiting them enables a company to acquire a certain legitimacy as a multinational.

The growing politicization of senior civil servants on ECs

38The dynamic in the ECs is not of the same nature. The presence of many members of the corps d’État within them cannot be explained as a residue of the past influence of the grands corps: their number has remained stable since 2004 (around 25%, see Table 4). In model 2 (Table 8), two variables significantly impact on the proportion of senior civil servants in company ECs. In equal amounts, the share of equity held by the state in 1986 (historic share) and the proportion of CEOs from the grands corps in the CAC 40 positively influence the representation of senior civil servants. Conversely, neither the internationalization of the company, represented by the share of revenue earned abroad, nor the share of equity owned by the state have an effect on the variable to be explained.

Table 8

Linear regression on the proportion of former senior civil servants in energy companies ECs

Table 8
Proportion of directors from the grands corps in the ECs Model 2 Coefficient Error type State share of equity 0.003 0.23 Number of CEOs from the grands corps in the CAC 40 0.49*** 0.08 Historic state share equity 0.55* 0.09 Share of company revenue earned abroad –0.05 0.13 Constant 0.03 0.06

Linear regression on the proportion of former senior civil servants in energy companies ECs

Note: *** p < 0.0001;** p < 0.01; * p < 0.05.
R2 0.58; 98 observations.

39The results of model 2 show that the composition of ECs is not linked to the composition of a firm’s shareholders, and in particular the role the state plays in relation to this. This confirms what the archives of the government department in charge of regulating the energy sector suggest: the government and ministerial cabinets are informed a posteriori of changes within the executive management of public companies. [8] Despite the significant increase in international managers in ECs, the internationalization of companies, understood by the share of revenue earned abroad, does not impact on the number of senior civil servants. The composition of CAC 40 ECs imposes a norm, the institutional environment playing a fundamental role, without however completely explaining the variation in the number of senior civil servants in ECs. The proportion of senior civil servants within ECs is positively and equally linked to historic financial relations between the firm under consideration and the state. Thus, privatized companies retain managers who have passed through government.

40But, in light of these developments, Dudouet’s (2007) argument, which sees the persistence of the grands corps in executive management as an expression of corporatism, is not an adequate one for the energy sector. Senior civil servants have lost ground and their careers have evolved. My survey shows that the senior civil servants recruited to ECs were increasingly in contact with the political world within ministerial cabinets. These links are forged increasingly higher up; the number of directors who have been advisors to the president or prime minister remaining constant at around 10% while the number of senior civil servants has declined from 70% to 30% (Table 4). Consequently, in 2014, 40% (11% among 28%) of directors from the civil service had come from Matignon or the Élysée against 20% in 1992 (14% among 69%). This development can be understood as a selection: the only senior civil servants who have succeeded in retaining their positions or in being appointed to executive positions are those who have sufficient support, or enough resources that a company depends on. These resources in the 2000s are political resources.

Table 9

EDF and Total executive directors (%)

Table 9
Executive committees Foreigner Senior civil servant Cabinet Élysée/Matignon Total 1992 0 75 75 25 EDF 1992 0 60 30 10 Total 2014 0 33 33 16 EDF 2014 0 22 11 0

EDF and Total executive directors (%)

Bruno Bensasson: A politicized senior civil servant

Born in Paris in 1972, Bruno Bensasson studied at the Duruy and Henri IV lycées. After leaving the École Polytechnique, he joined the Corps des Mines. Head of the nuclear installations division of the Direction Régionale de l’Industrie, de la Recherche et de l’Environnement en Région (Regional Department of Industry, Research and Environment), he then became director of the cabinet of the Direction Générale de la Sureté Nucléaire et de la Radioprotection (General Directorate of Nuclear Security and Radiation Protection) until 2004. A technical advisor on the environment, new forms of energy and nuclear energy to the Ministry of Industry until 2006, he left the civil service after spending time in the President’s office as a technical advisor on industry, the environment and transport. He then joined GDF Suez as deputy director general for strategy (2007–2008), then as director of economic, prices and markets from 2008 to 2011. He became head of strategy and sustainable development and became a member of the executive committee in 2011.
(Source: Who’s Who France 2014)

41As an example, take developments identified when comparing EDF and Total. In 1992 the ECs of the two companies were quantitatively dominated by senior civil servants. The latter are in the minority in 2014, but there are notable differences between Total and EDF. In 2014, of Total’s 6 member EC, two are from the corps des Mines, one of whom had spent time in the Ministry of Defence cabinet and the other in Matignon. In 2014, the EDF EC included one former civil servant among its 9 members, from the Council of State and previously an advisor in the Ministry of Justice. Another director began his career in public service at the Conseil National des Études Spatiales (National Space Research Council) without being a member of a corps. Finally the EC’s secretary was an advisor to the general secretary of the RPR (the Rassemblement pour la République, liberal-conservative political party). The EDF’s investment thus seems less than that of Total; the EDF being a public company, it had less need for this type of investment than Total.

42In contrast to BDs whose recruitment rationales are still largely dependent on state influence and the defence of its interests, or on the willingness of managers to withdraw from it, energy company ECs function in a relatively autonomous way in relation to the state in terms of recruitment. In this context, companies choose a smaller number of directors from the corps d’État (only 28% in 2014), but increasingly such directors have political backgrounds, since the majority have spent time in ministerial cabinets, and more than a third at the Élysée or Matignon. This is why, in the context of a withdrawal of the state from the energy sector, companies continue to employ senior civil servants at the highest level who are more politicized than is generally the case.

From the interests of the major groups to the interests of the state

43The 1990s–2000s were undoubtedly years of a decline of the historically voluntarist action by the state in the energy sector, but also the setting for many developments in the French energy sector, mostly driven by the political sphere. The period is marked by regulatory instability. In an interview, a former EDF director described the political factor as the prime factor determining strategy. In contrast, “we and the civil service understood each other, we had the same figures, and with the same knowledge even if our interests sometimes diverged we could negotiate and reach agreement.” It is the political sphere that represents a source of uncertainty for energy sector leaders. Moreover, the political sphere has become the interlocutor for energy sector leaders at the expense of the civil service. All this article can do is suggest a number of hypotheses; proving them would require more significant work. Technical departments, such as the DGEMP, have lost considerable influence and resources following the privatizations of the companies under their supervision (the DGEMP had 1,200 officials in 1990 in two departments; it has only 200 in 2015). The decisions concerning companies are no longer of the order of controlling investment, but modify the structure of the sector itself. Actions such as arbitration between companies that have become competitors or mergers between firms have become more common, and thus involve the Élysée or Matignon.


—It’s strange, I was expecting to see more about intervention from the state in our discussions. (interviewer)
—Politics is present, but the civil service disappeared in the 1990s. There is one actor, the APE, [9] that intervened because of the opening up of capital. When there was too much tussling between X and Y it intervened. But the only real actors are political actors.
(Former executive director)

45The companies’ managements do not remain passive with respect to these reforms and ensure their voices are heard so that their firms’ interests are taken into account. These major reforms necessitate directors who are familiar with the civil service but above all with the higher echelons of the state, which explains the persistent presence of members of the grands corps d’État on ECs. Certainly, they depend on positions of power acquired in the past (Dudouet and Gremion 2007), but above all they are useful to the company employing them.

46Usually, the government department charged with energy policy is the companies’ interlocutor, as the official regulator. However, some senior managers exploit their experience of the higher echelons of the state and their personal acquaintance with certain actors and short circuit the administration, speaking directly to the Élysée or Matignon. According to Sawicki and Mathiot (1999), the Prime Minister’s cabinet is central in that is it at the centre of the decision-making system and information and power is concentrated within it, in particular the power to arbitrate conflicts between ministers. As such, recruiting someone with experience of this situation is an asset for a company faced with changes to the legislation governing its field of activity. As a former senior civil servant in the hydrocarbon directorate says in an interview: “I don’t think the privatization of Elf and Total has changed many of the relationships between the state and these oil companies. They were companies of a considerable size and so it was a normalization of relations. These are subjects that are dealt with at the presidential level. He has necessarily close relationships with the company presidents.” Despite privatization, the relationships remained close between the political sphere and oil company managers, proving that dependent relationships are not just equity-based ones.

47Having a director from Matignon’s cabinet office or the general secretariat of the Élysée within its management – even at its head – paradoxically guarantees a company some autonomy with respect to the administration. In her autobiography published following her dismissal as head of Areva, Anne Lauvergeon [10] talks of her recurrent interviews with the Prime Minister and the President, as well as her recurrent disputes with the latter. The archives belonging to Alain Juppé’s advisor on industry’s (20000526/5 and 20000526/6) from 1995 to 1997 show, for example, the continuous involvement of G. Ménage, then Edmond Alphandéry, both presidents of EDF, from 1992 to 1995 and from 1995 to 1998 respectively in the negotiations leading to the adoption of the European directive on the internal electricity market in 1996. Ménage was a deputy director, then director, of François Mitterrand’s cabinet from 1981 to 1991 and Alphandéry was Minister of the Economy from 1993 to 1995. The extensive correspondence they exchanged with Juppé and Jacques Chirac bears witness to the cordial relationships they entertained with the top of the executive. In a letter, Alphandéry mentions his “concerns” over the discussions on the European directive on the internal electricity market that Juppé was going to have with chancellor Helmut Kohl the following day “7th May, I must warn you, is a meeting fraught with danger. The plan on the table is an explosive one […] In this context, I believe it is my duty to suggest you get the Germans to change direction. For years the commission has sought, artificially, to introduce a market in all countries that does not exist. Can we not profit from the general weariness with this directive to change footing?” Having been Minister of the Economy, he was able to communicate directly with Juppé, without going through the civil service.

48In the introduction, it was hypothesized that the circulation of directors resulting from pantouflage was emblematic of the interdependent relations between the political sphere and civil service, on the one hand, and the major firms on the other. While I have shown that the companies benefit from this situation, what of the administration (Colignon and Usui 2004)? It is difficult to determine this. Aside from the fact that the market sector is now a necessary outlet at a time when positions in the central administration are becoming fewer, former senior civil servants may act as information go-betweens for the department charged with regulating energy companies. In a note on the new organization of the management of EDF (20090045/3) decided unilaterally by its president, Alphandéry, Dominque Maillard, Director General of Energy and Raw Materials from 1998 to 2007 observed to the cabinet of the Ministry of Industry that “the least we can say is that our usual contacts are not the winners.” The contacts in question, two former senior civil servants from the Corps des Ponts and the Corps des Administrateurs Civils, found their positions marginalized in comparison to the previous organization of the EC. Directors parachuted in from ministerial cabinets to the head of companies are more likely to show attachment to the defence of the public interest and public service values guaranteed by the state, and all the more so when the renewal of their mandates depends on government good will. Thus, EDF and Gaz de France directors prove to be particularly attached to the public service missions attributed to their companies in the course of successive reforms of their respective companies. In their writings, others support a vision of a strong state in its functions of shareholder and regulator, as well as of strategist. It is difficult, however, to distinguish rhetoric from real engagement.

49* * *

50The social characteristics of energy sector managers have shifted in two directions in the last two decades. The first shift affects non-executive directors who were principally recruited from the grands corps de l’État at the beginning of the 1990s. The companies’ internationalization strategies have driven them to recruit more foreign directors, drawing a line which separates the energy sector firms from their historic dependence on the French state and enabling them to draw on acquaintance networks beyond mainland France alone. The second shift complements the first without contradicting it. It has seen the comparative persistence of former senior civil servants within the companies’ ECs. These civil servants have, however, more often spent time in ministerial cabinets than their predecessors, and often in positions that are more political and less technical. The persistence of these profiles in ECs reflects the strategic importance for energy sector companies of direct access to the most senior political leaders—direct access that procures resources that are often crucial in the competition faced by major European companies. Executive committee recruitment is the result of an explicit strategy by the firms. This plays out on two fronts: both demonstrating independence vis-à-vis the state through the composition of their BDs, and using the social capital of their executive directors. This strategy is based on pantouflage, an age-old institution (Joly 2013). These links illustrate the interdependent relationships involving the civil service and the political sphere on the one hand and the major energy groups on the other, demonstrating Fligstein’s (2001) arguments. Despite the recent developments in the liberalization of oil, electricity and gas markets, energy companies depend greatly on the resources of the regulatory state, which allows the grands corps d’État to maintain a strong presence within the energy sector firms’ managements. The appointment of senior civil servants from ministerial cabinets contributes to reducing the uncertainty linked to the political sphere. In this sense, it is in line with the logic of privatisation: by depriving the state of its shareholder role, energy sector managers deprive it of a lever over their firm’s management and thus reduce the uncertainty linked to the political sphere. It is much harder to attribute a strategy to the state, which only has very indirect control of the appointment of directors. As for the grands corps (particularly the technical ones), we can only assume a loss of influence or redeployment to other sectors.

51Interdependent relationships have been revealed by changing scale, focussing the study on a single sector and tracing changes through it, and by adding a temporal dimension, framed by specific events in the sector. This study also calls for a change in perspective concerning elites. It is important to consider economic elites not just as actors capable of calling upon substantial resources to serve their interests, but also as actors who are themselves resources for organizations. Their knowledge and likewise their social capital, are all assets for the institutions employing them. Thus, Bauer and Bertin-Mourot’s (1987, 1997) notion of state asset, or Bourdieu’s (1989) social capital, should be understood both in terms of the manager him/herself and in terms of the organization that employs them.

I would like to thank Pierre François, Sebastian Billows and Vanessa Albert for their attentive reading and fruitful comments. A preliminary version of this text was presented at a sociology of elites thematic network session at the Association Française de Sociologie conference (Nantes, 5 September 2013). As a result I received many suggestions from participants whom I thank here. I would also like to thank the Revue Française de Sociologie reviewers who have improved the quality and accuracy of this text. The article is, nevertheless, solely the author’s responsibility.
Table A10

List of companies

Table A10
Company First year in the database Last year in the database Areva 2001 2009 COGEMA 1990 2000 Elf Aquitaine 1990 1999 EDF 1990 2009 Framatome 1990 2001 Gaz de France 1990 2009 Suez 1996 2007 Technip 1990 2009 Total 1990 2009

List of companies


  • [1]
    A “revolving door” career transition named in reference to the pantoufle that is normally paid by someone who has not completed the ten years state service s/he was committed to do when entering the corps.
  • [2]
    The French administration is divided into corps, institutions that manage the careers of their members. The grands corps d’État, made up of senior civil servants, are the most prestigious and are characterized by considerable solidarity among their members, past and present. Their members have access to positions of power within the administration.
  • [3]
    “You say ‘us’, who is the ‘us’?” “Oh, sorry, I say ‘us’ because we, the EC, take decisions collegially” (interview with a petrol company manager).
  • [4]
    These references indicate the relevant document number at the National Archives in the DGEMP archives.
  • [5]
    Statistics assembled from Who’s Who France entries for energy company directors.
  • [6]
    Based on data from Bauer and Bertin-Mourot (1997), Dudouet and Grémont (2007) and François (2010).
  • [7]
    DGEMP archives, lot n°20090185/1.
  • [8]
    20090049/2, 20090045/3.
  • [9]
    Agence des Participations de l’État (State Equity Investment Agency), government agency in charge of managing the state’s equity investments in companies.
  • [10]
    La femme qui résiste. Paris: Plon, 2012.

Since the end of the 1980s the state has been engaged in a withdrawal from the management of nationalized businesses. The institutional environment for industrial firms has changed considerably as a result. Firms have internationalized and a number of markets have been liberalized. What has been the result of these changes for relations between the state and big business? In this article I set out to respond to this question looking at the case of the energy sector, in which the state has played a major shareholder role since the beginning of the 1990s. In considering the managers of the largest firms in this sector as business resources, I will identify the strategies implemented by these firms, which reveal recourse to the state and notably the political sphere. The social characteristics of energy sector managers have evolved in two directions during the last two decades. The firms have it both ways, exhibiting their independence from the state through boards of directors, while reinforcing informal ties with the political sphere through their executive committees.


  • capitalism
  • economic elites
  • state
  • industry
  • energy
  • stateowned companies


  • Bauer, M., Bertin-Mourot, B., 1987. Les 200. Comment devient-on un grand patron? Paris: Le Seuil.
  • Bauer, M., Bertin-Mourot, B., 1997. Administrateurs et dirigeants du CAC 40. Paris: Boyden.
  • OnlineBeckert, J., 2009. “The Social Order of Markets,” Theory and Society 38(3): 245–69.
  • OnlineBezes, P., 2005. “Le modèle de ‘l’État-Stratège:’ genèse d’une forme organisationnelle dans l’administration française,” Sociologie du Travail 47(4): 431–50.
  • Bourdieu, P., 1989. La noblesse d’État. Paris: Éditions de Minuit. Published in English as The State Nobility. Stanford, CA: Stanford University Press, 1996.
  • OnlineBourdieu, P., Saint Martin, M. (de), 1978. “Le patronat,” Actes de la Recherche en Sciences Sociales 20–1: 3–82
  • OnlineBurt, R., 1995. “Le capital social, les trous structuraux et l’entrepreneur,” Revue Française de Sociologie 36(4): 599–629.
  • OnlineByrd, D., Mizruchi, M., 2005. “Bankers on the Board and the Debt Ratio of Firms,” Journal of Corporate Finance 11: 129–73.
  • Chevalier, J.-M., Percebois, J., 2008. Gaz et électricité, un défi pour l’Europe et pour la France. Paris: La Documentation Française.
  • OnlineCobb, D., Davis, G., 2010. “Resource Dependence Theory: Past and Future,” Research in the Sociology of Organizations 28: 21–42.
  • OnlineColeman, J., 1988. “Social Capital in the Creation of Human Capital,” American Journal of Sociology 94(Supplement): S95–120.
  • Colignon, R. A., Usui, C., 2003. Amakudari, the Hidden Fabric of Japan’s Economy. Ithaca, NY: Cornell University Press.
  • OnlineComet, C., Finez, J., 2010. “Le cœur de l’élite patronale,” Sociologies Pratiques 21: 49–66
  • OnlineDavis, G. F., Diekmann, K. A., Tinsley, C. H., 1994. “The Decline and Fall of the Conglomerate Firm in the 1980s: The Deinstitutionalization of an Organizational Form,” American Sociological Review 59(4): 547–70.
  • OnlineDavis, G. F., 1996. “The Significance of Board Interlocks for Corporate Governance,” Corporate Governance: An International Review 4(3): 154–9.
  • OnlineDudouet, F.-X., Grémont, E., 2007. “Les grands patrons et l’État en France, 1981– 2007,” Sociétés Contemporaines 68: 105–31.
  • OnlineDudouet F.-X, Joly H., 2010. “Les dirigeants français du CAC 40: entre élitisme scolaire et passage par l’État,” Sociologies Pratiques 21: 35–46.
  • OnlineDuval, J., 2013. “L’analyse des correspondances et la construction des champs,” Actes de la Recherche en Sciences Sociales, 5: 110–23.
  • OnlineFinez, J., 2014. “La construction des prix à la SNCF, une socio-histoire de la tarification,” Revue Française de Sociologie 55(1): 5–39. Published in English as “Ticket Pricing by the French National Railway Company (SNCF), a Historical Sociology of Tariff Setting,” Revue Française de Sociologie 55(1, English version): 1–31
  • OnlineFinez, J., Comet, C., 2011. “Solidarités patronales et formation des interlocks entre les principaux administrateurs du CAC 40,” Terrains & Travaux, 2: 57–76.
  • Fligstein, N., 2001. The Architecture of Markets: An Economic Sociology of Twenty-First-Century Capitalist Societies. Princeton, NJ: University Press.
  • François, P., 2010. “Les guépards du capitalisme français? Structure de l’élite patronale et modes d’accès aux positions dominantes,” Les élites économiques en France, Journées d’études conjointes DRM-IRISSO, Université Paris Dauphine, 4 & 5 November 2010.
  • François, P., Lemercier, C., 2014. “Ebbs and Flows of French Capitalism,” in T. David, W. Gerharda, eds, The Power of Corporate Networks. A Comparative and Historical Perspective. London: Routledge.
  • Hecht, G., 2009. The Radiance of France: Nuclear Power and National Identity after World War Two. Cambridge, MA: MIT.
  • Jabko, N., 2009. L’Europe par le marché. Histoire d’une stratégie improbable. Paris: Presses de Sciences Po.
  • OnlineJabko, N., Massoc, E., 2012. “French Capitalism under Stress: How Nicolas Sarkozy Rescued the Banks,” Review of International Political Economy 19(4): 562–85.
  • Jobert, B., Muller, P., 1987. L’État en action. Politiques publiques et corporatismes. Paris: Presses Universitaires de France.
  • OnlineJoly, H., 2007. “Les études sur le recrutement du patronat: une tentative de bilan critique,” Sociétés Contemporaines 68: 133–54.
  • Joly, H., 2013. Diriger une grande entreprise au XXe siècle: l’élite industrielle française. Tours: Presses Universitaires François-Rabelais.
  • Kuisel, R., 1984, Le capitalisme et l’État en France. Modernisation et dirigisme au XXe siècle. Paris: Gallimard.
  • Lévy-Leboyer, M., 1979. “Le patronat français, 1912–1973,” in M. Lévy-Leboyer, ed., Le patronat de la seconde industrialisation. Paris: Editions Ouvrières, pp. 137–88.
  • Loiseau, H., 2002. “1985–2000: quinze années de mutation du secteur public d’entreprises,” INSEE Première, 860.
  • OnlineMarchal, E., Rieucau, G., 2010, Le recrutement. Paris: La Découverte.
  • Mintz, B. A., Schwartz, B., 1985. The Power Structure of American Business. Chicago, IL: University of Chicago Press.
  • OnlineMizruchi, M. S., 1996. “What do Interlocks do? An Analysis, Critique, and Assessment of Research on Interlocking Directorates,” Annual Review of Sociology 22(1): 271–98.
  • OnlineMizruchi, M., Stearns, L., 1988. “A Longitudinal Study of the Formation of Interlocking Directorates,” Administrative Science Quarterly 33(2): 194–210.
  • OnlineMorin, F., 1996. “Privatisation et dévolution des pouvoirs. Le modèle français du gouvernement d’entreprise,” Revue Économique 47: 1253–68.
  • Nouschi, A., 2001. La France et le pétrole de 1924 à nos jours. Paris: Picard.
  • OnlinePalmer, D., Friedland, R., Singh, J. V., 1986. “The Ties that Bind: Organizational and Class Bases of Stability in a Corporate Interlock Network,” American Sociological Review 51(6): 781–96.
  • Pfeffer, J., Salancik, G. R., 1978. The External Control of Organizations, a Resource Dependence Perspective. New York, NY: Harper & Row.
  • Reverdy, T., 2014. La construction politique du prix de l’énergie. Paris: Presses de Sciences Po.
  • OnlineSawicki, F., Mathiot, P., 1999. “Les membres des cabinets ministériels socialistes en France, 1981–1993: recrutement et reconversion. Première partie: caractéristiques sociales et filières de recrutement,” Revue Française de Science Politique 49(1): 3–30.
  • Selznick, P., 1957. Leadership in Administration: A Sociological Interpretation. Evanston, IL: Row, Peterson.
  • Suleiman, E., 1979. Les élites en France, grands corps et grandes écoles. Paris: Le Seuil.
  • OnlineWestphal, J., Zajac, E., 1995. “Who Shall Govern? CEO/Board Power, Demographic Similarity, and New Director Selection,” Administrative Science Quarterly 40(1): 60–83.
  • OnlineZarlowski, P., Djelic, M.-L., 2005. “Entreprises et gouvernance en France: perspectives historiques et évolutions récentes,” Sociologie du Travail 47(4): 451–69.
Scott Viallet-Thévenin
CSO—CNRS-Sciences Po
19, rue Amélie
75007 Paris
Translated by 
Toby Matthews
Latest publication on cairn or another partner portal
Uploaded on on 30/01/2017
Distribution électronique pour Presses de Sciences Po © Presses de Sciences Po. Tous droits réservés pour tous pays. Il est interdit, sauf accord préalable et écrit de l’éditeur, de reproduire (notamment par photocopie) partiellement ou totalement le présent article, de le stocker dans une banque de données ou de le communiquer au public sous quelque forme et de quelque manière que ce soit.
Loading... Please wait